Commons:Village pump/Archive/2013/09

From Wikimedia Commons, the free media repository
Jump to: navigation, search
Archive This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Contents

Abuse filter in the Khmer Wikipedia

Every time it try to visit a page in the Khmer Wikipedia from the red number links for global usage in galleries I en up in km:ពិសេស:AbuseFilter. Any ideas? -- Tuválkin 12:55, 28 August 2013 (UTC)

Here is the Wikimedia Commons, which is independent from Wikipedia projects. You may contact an administrator of Khmer language Wikipedia. Best regards. – Kwj2772 (msg) 13:10, 28 August 2013 (UTC)
Even Commons community can help you, I didn't understand what you said. :( – Kwj2772 (msg) 13:13, 28 August 2013 (UTC)
It's about the links from Gadget-GlobalUsageUI. Tuválkin, can you provide an example please so I do not have to search? Thanks in advance. -- Rillke(q?) 13:47, 28 August 2013 (UTC)
I failed to find an example myself, trying stuff like http://km.wikipedia.org/wiki/%E1%9E%AF%E1%9E%80%E1%9E%9F%E1%9E%B6%E1%9E%9A:Wikipedia-logo-v2.svg#globalusage seems too not use this gadget? I'm confused... Tuvalkin, please provide a testcase. Thanks in advance! --AKlapper (WMF) (talk) 10:35, 29 August 2013 (UTC)
I guess Tuválkin is clicking a link in this tipsy popup and is redirected to km:ពិសេស:AbuseFilter but I would be also glad if I would have an example … -- Rillke(q?) 11:57, 29 August 2013 (UTC)
I assume this is covered in bugzilla:53498. --AKlapper (WMF) (talk) 12:20, 29 August 2013 (UTC)
Question for Tuvalkin: What happens if you try clicking the same link again? Does it still redirect you to AbuseFilter? I've found that it only happens the first time I get auto-logged-in, and after that the link works fine. If this is happening to you repeatedly with the same page on the same wiki, then this bug is even more serious than we thought. — PinkAmpers&(Je vous invite à me parler) 14:50, 29 August 2013 (UTC)
Sorry I went mute on you about this, and thanks for all the feedback. Meanwhile I had the same problem when accessing a couple other random projects, such as one of the Wikisources (I forgot which): I click a link to a specific page and I end up in that project’s Abuse Filter page. I’ll read the bugzilla thread above carefully and will come back with more guinea-pig reports if anything interesting happens to me again. -- Tuválkin 02:38, 31 August 2013 (UTC)
I can confirm this; Recently, I visited ru.wikivoyage, I think, and ended up on the AbuseFilter page despite I wanted to visit their "village pump" following a language link in the sidebar. -- Rillke(q?) 17:50, 1 September 2013 (UTC)

Prompt deletion of personal information about children

I recently added a speedy deletion tag against a mobile uploaded selfie image of a 14 year old where the title included their full birth date and legal name. In addition the image appeared to have been uploaded by someone else, probably another child in the same family. The speedy was rejected by one of our Bureaucrats, so I have been forced to raise a DR as they recommended, though this makes suppressing this information much harder as it starts to get propagated (by repetition of the file name) across Commons in various ways. You can find the DR via Special:Redirect/revision/102855749 (I am trying to avoid repeating the filename further). [Update: since raising the matter here, the image has been removed by one of our Oversighters and the Deletion request has itself been deleted.]

This is a situation where the proposed courtesy deletions process might help (see the discussion at Commons_talk:Courtesy_deletions), however I believe even the current guidelines are sufficient when it comes to personal information about children, however they are not being implemented in a pragmatic way that avoids the unnecessary and potentially irrevocable propagation of personal information, which does nothing to fulfil the scope of this project. I welcome comments on the DR linked to above, and further feedback and example case studies to be added to the courtesy deletion proposal. In general I would recommend that the respect guideline in Photographs of identifiable people fully applies to personal information such as full birth date, home address, which school a child attends etc. and any administrator or other trusted user can redact and remove any such irrelevant data on request, hopefully early on before it becomes a permanent archived record elsewhere. Thanks -- (talk) 16:21, 28 August 2013 (UTC)

And nobody thought of notifying me? It is not that nobody can make a mistake once in a while. If I never hear about it how am I supposed to learn? ;-) --Dschwen (talk) 17:06, 28 August 2013 (UTC)
What, bureaucrats make mistakes? :-) I'll remember that for my next massive cock-up. -- (talk) 17:28, 28 August 2013 (UTC)
I fully agree with a policy similar to enwiki authorizing us to delete or revdel personally identifying information about young minors, and warn them. That said I'd probably be okay with an older minor (e.g. 16 or 17) uploading an image of themselves for their user page - user page images are explicitly permitted by SCOPE and is relatively harmless in the absence of other identifying info. I would purge info like birthdates/street address/etc. regardless of the minor's age. Dcoetzee (talk) 05:56, 29 August 2013 (UTC)
Since the problematic data is in the filename and not in the media itself, and should the image be otherwise considered in scope, shouldn’t the security/privacy issues be soved by a swift renaming of the file, completed with admin expunging of the old filename? (Ditto, m.m., for sensitive data added to the file description, EXIF, or anywhere else.) -- Tuválkin 02:45, 31 August 2013 (UTC)
Yes, I agree, files containing sensitive data in filename should be renamed (without a redirect and without leaving any trace in the logs of the former filename - I think the deletion log might have to be purged?). For the file description, revision deletion suffices; for EXIF data, stripping and reuploading the image, and revision deletion of the old file contents should be sufficient. Dcoetzee (talk) 08:12, 1 September 2013 (UTC)

August 29

Wikidata as replacement for interwikilinks?

Is is possible in commons to replace interwikilinks in categories and galleries, in particular for biological species, by wikidata entries? Regards, --Burkhard (talk) 21:07, 29 August 2013 (UTC)

No. There's a Template:On Wikidata, which at least provides a cut-and-paste list of language links. ghouston (talk) 23:51, 29 August 2013 (UTC)
It would be easy for bots to add or update interwikis and the {{on Wikipedia}} template using data from Wikidata. Before Wikidata started, some bots used to keep interwikis updated by copying them from wikipedias, and the same work should be done now using Wikidata.--Pere prlpz (talk) 10:45, 30 August 2013 (UTC)
We are working on getting interwiki links via Wikidata to Commons soon now like we've done for Wikivoyage recently. I'll let you know as soon as I have a date. --Lydia Pintscher (WMDE) (talk) 10:07, 1 September 2013 (UTC)

Automatically added category Category:Uploaded via Campaign:*****

It would appear that sometime between and , using Upload Wizard Campaigns will cause media to be automatically categorized with Category:Uploaded via Campaign:*****. I.e. in the case of Campaign:dk, media is tagged with the (currently non-existing) category Uploaded via Campaign:dk. In the campaign we are however also autoadding the category Uploaded from da-wiki, so media are in effect added to two categories automatically, which (depending on the purpose of Category:Uploaded via Campaign:dk) seems a bit superfluous. I haven't been able to find any announcements/reason (but I probably simply don't know where to look?) on the Category:Uploaded via Campaign addition, so I don't know exactly what it is for, but I am tempted to remove the Uploaded from da-wiki (which we are using to "vet" the files uploaded through Danish Wikipedia by Danish administrators) at some point and simply use Uploaded via Campaign:dk. --heb [T C E] 06:03, 29 August 2013 (UTC)

This is currently not configurable by campaign editors by design and was added to UploadWizard with this commit. That's all I can say. -- Rillke(q?) 08:22, 29 August 2013 (UTC)
Hello! That was me! Sorry about not giving more wide notice (I mentioned this in the #wlm channels). The 'Uploaded via Campaign:xxx' categories are being added to be able to accurately and unambiguously track which images were uploaded from which campaign. This lets us do a number of cool things (look at prototype http://blue-dragon.wmflabs.org/wiki/Campaign:show-off-campaigns, discussion at http://lists.wikimedia.org/pipermail/design/2013-August/000839.html - and more things coming to mobile, etc). I'd highly recommend just using those and removing the Uploaded from da-wiki category and just using the automatically added one. Thank you! Yuvipanda (talk) 14:48, 29 August 2013 (UTC)
Thank you both User:Rillke & User:Yuvipanda. Given the potential of "accurately and unambiguously track[ing] which images were uploaded", and the way we currently use Uploaded from da-wiki, I think we should keep both as files are removed from Uploaded from da-wiki after being "vetted". Removing files from Category:Uploaded via Campaign:dk doesn't seem to be a part of the scope for that category. --heb [T C E] 11:09, 2 September 2013 (UTC)

Template:Gadget-desc performance

For anyone familiar with how this template works: I noticed that the slowness of this template is causing Special:Preferences to take a long time to load. You can add forceprofile=true to the URL to see details (using the browser's "view source" tool). This also leads to more deadlocks on preferences changes done via the API. I can hack the code a bit to work around the deadlocks, though I don't see any easy way to avoid the slowness without the template being simplified or otherwise improved for performance. Note that parsed interface messages are not cached (the text is, but not the output) and caching would require extra dependency tracking that does not exist at the moment. It would be nice if anyone was interested in taking a crack at this. Aaron Schulz (talk) 17:38, 30 August 2013 (UTC)

So to translate this into a language that I do understand: Special:Preferences is not stored in the parser output cache. Because we use some templates, they have to be loaded and parsed each time the user preferences are switched via API (watchlist message, Upload Wizard dismiss tutorial, …) or the user goes to Special:Preferences. We could use MediaWiki messages but they should not handle text that takes a long time to parse.

-- Rillke(q?) 19:46, 30 August 2013 (UTC)

Seems kind of odd the api would even be parsing gadget descriptions... Bawolff (talk) 23:23, 30 August 2013 (UTC)
Indeed. You can trigger the same hook that happens for prefs/ApiOptions by calling GadgetHooks::getPreferences() in eval.php (this is were I first profiled). In theory one could hack that function to not call $gadget->getDescription() in API context probably, which would avoid deadlocks and slowness in ApiOptions, though wouldn't help Special:Preferences. Aaron Schulz (talk) 00:21, 31 August 2013 (UTC)

I filed bugzilla:53677 so we can drop all {{LangSwitch}}es. I guess titleparts is also not very parser-friendly… but somehow we have to be able to efficiently manage information without having to edit each single translation… -- Rillke(q?) 11:12, 2 September 2013 (UTC)

Categorization and labelling of works depicting Chelsea Manning

On English Wikipedia due to Bradley Manning's recent coming out as transgender, the article was moved to en:Chelsea Manning, but on Commons most of the associated filenames, categories, and file descriptions still say "Bradley Manning". Category:Chelsea Manning is a catredirect to Category:Bradley Manning, and so attempts to update the category of images to "Chelsea Manning" were of course reverted. I propose that:

  • Category:Bradley Manning is renamed to Category:Chelsea Manning, and the former catredirect to the latter, rather than vice versa. All files in the category are moved to the new one.
  • Files including "Bradley Manning" in the filename be renamed to use the name "Chelsea Manning". (The filename may also be modified to indicate that she was formerly known as Bradley Manning.)
  • File descriptions/category descriptions are updated to change the name "Bradley Manning" to "Chelsea Manning". This has mostly already been done. (Again they may indicate that she was formerly known as Bradley Manning.)

Since the second point above runs counter to Commons:File renaming, I would not want to do it without prior consensus - but this appears to be a new situation that was not anticipated by this policy. Dcoetzee (talk) 08:06, 1 September 2013 (UTC)

Fæ's photo of a protest at Gay Pride 2012 supporting Chelsea Manning—a candidate for renaming.
  • Symbol support vote.svg Support I hope the embarrassingly stupid edit warring on en.wp does not spread, and we can create a credible consensus and stick to it. I note that one of my photographs has already had the categories changed and changed back again by someone more active on en.wp than they are on this project. Should anyone wish to discuss the issue of respect on Wikimedia projects for LGBT identity in a non-hostile environment, I recommend joining our Wikimedia LGBT email list. :-) -- (talk) 08:25, 1 September 2013 (UTC)
  • Symbol neutral vote.svg not so fast - we consider photos legally (in regard to copyright) as works of art. If the artist/photographer has given his work a title, this title is legally binding, as expressedly mentioned in the legal code of the CC licenses (see section 4.c.[1]). For me, a supplementary note (refering to current event) in the description of such images would be o.k., but completely and intentionally changing filename and description would require serious evaluation. --Túrelio (talk) 08:37, 1 September 2013 (UTC)
    • It is worth noting that categories and filenames are not the same thing as the titles of works represented. Commons is not obliged to use the artist's title for a work in either. Where a photograph as a creative work is given an 'official' title by the photographer (vanishingly rare for current event photographs), this might be represented in the description and that title should of course be preserved, along with helpfully adding the changing context. -- (talk) 09:17, 1 September 2013 (UTC)
    • Section 4c reads "reasonable to the medium or means You are utilizing." In the situation where the uploader supplies a formal title (which is not the same thing as a filename), we are in compliance as long as we retain mention of the title in the file description. We have never observed a legal obligation to avoid renaming files, and all the files in Category:Bradley Manning have simple descriptive filenames that I would not construe as titles (particularly the ones that I uploaded and supplied the filenames for myself, File:Bradley Manning 1.jpg and File:Bradley Manning 2.jpg). Dcoetzee (talk) 09:21, 1 September 2013 (UTC)
"reasonable to the medium or means You are utilizing" just relates to the manner/location of credit, not to its content.
The thing is that many images in this category are from external sources. --Túrelio (talk) 09:23, 1 September 2013 (UTC)
Túrelio, you are well known as a sensible contributor, so I am a bit worried that I don't think I properly understand the point you are making in relation to these images representing Chelsea Manning and the controversies surrounding her in various ways, we might be talking at slight tangents. Given that it is easy to retain any original descriptive text provided by the uploader (we can just supplement it), if there is a relevant example in these categories that you can point to in order to illustrate how we might be breaching 4c(ii) "the title of the Work if supplied" by changing either category names or filenames, this would probably make your concern a more pragmatic issue of implementation. Thanks -- (talk) 09:46, 1 September 2013 (UTC)
I was not refering to category-names, as this is not directly associated to individual works. Whether the Manning-cat should be renamed is completely up to the Commons community, IMO. However, as of now the :en article is still Bradley Manning with a redirect from Chelsea Manning. Other Wikipedias seem to be similarly divided over this issue. For sure, anybody searching for "Chelsea Manning" on Commons should be led to Manning's category.
AFAIK, it is true that so far on Commons we have taken little (or no) consideration about "title of the work" (in the copyright sense) — except for paintings by famous painters — despite its mentioning in the CC license terms. I don't have an informed opinion about whether filename and/or description of files by self-uploaders might constitute such a title, as this would need legal expertise.
So, whether my general Caveat in regard to the description/filename of individual works really applies to any of the currently existing images, needs to be checked. If all these images were works of Commons-users, it would be easy to ask them either for any objection or whether they consider their choosen filename/description (or part of it) as title of the work. But, as most of the files are from external sources, IMO we should honour the description choosen by the author, even if we aren't sure whether we are legally bound to it. This wouldn't preclude a supplementary note about the Bradley/Chelseas issue. --Túrelio (talk) 10:27, 1 September 2013 (UTC)
I don't think we should honor the original description, except possibly in a section labeled "Original description". If we have something we believe was intended as the title, we should put it in a section labeled Title, not Description. If the creator gave us a description that was marked description, not title, I don't think we have to worry about it as a title--and if we do, again, it should be labeled title and we do what we need to do with the description.--Prosfilaes (talk) 10:47, 1 September 2013 (UTC)
  • While I don't agree with Turelio that there are legal reasons to do so, I don't think we should rename files after a name change - there isn't really a basis for this in COM:File renaming, and in general, users are going to recognize both "Bradley Manning" and "Chelsea Manning". With file names we should stick to whatever name the uploader chose. I don't have a strong opinion about which name categories and descriptions should use - perhaps we should choose that by counting what the majority of Wikipedias do for categories, and use both names in descriptions? darkweasel94 11:31, 1 September 2013 (UTC)
  • Just one small point: I don’t believe the file renaming policy strictly runs against one or the other choices we have (I think this case arguably falls under criteria #3) − it’s just the matter of deciding what shall be done. Jean-Fred (talk) 12:01, 1 September 2013 (UTC)
    • Neither name is misleading, in my opinion. darkweasel94 12:21, 1 September 2013 (UTC)
  • Pictogram voting comment.svg Comment Having a look over at the discussion on en:Talk:Bradley Manning the retrospective situation seems rather unclear. Although any photo taken from this point should undoubtedly be called Chelsea Manning, older photos may not require a rename if we follow Pvt Manning's statements, and other precedent on the matter. This statement from the Private Manning Support Network seems to sum it up quite well. Liamdavies (talk) 12:43, 1 September 2013 (UTC)
People, places and things can be known by multilpe names, at the same time or during different times in history. This can be a problem with categories where we have to agree on one name to use. But for images I think the easiest solution is to use the filename given by the uploader (unless it is clearly misleading or un-NPOV). Renaming files is just creates extra work changing image links and maintaining redirects. To avoid confusion helpful explanations can be added to the image descriptions, e.g. "X (also called Y)", "X (now known as Y)", "Y (known as X at time of photography)", ... /Ö 12:52, 1 September 2013 (UTC)
  • Symbol oppose vote.svg Oppose any file renames, that's just a waste of resources. You might want to update the description. As for the category rename, I protected the category so we don't get people moving it around without going to Commons:Categories for discussion. Multichill (talk) 13:03, 1 September 2013 (UTC)
  • number of issues raised Symbol support vote.svg Support category name being at Chelsea Manning, with Bradley Manning being a redirect we should respect the person choice of name just the same as we do if company changes its name. Symbol oppose vote.svg Oppose renaming file names uploaded before the change of name became public knowledge as they were named in good faith. Prior to name change either or both is acceptable, after the name change only Chelsea... What ever the outcome it should be made clear in the category description that files uploaded after 22 August 2013 should refer to her as Chelsea. Gnangarra 13:55, 1 September 2013 (UTC)
  • Pictogram voting comment.svg Comment It seems to have escaped Dcoetzee's notice that the article was moved back to "Bradley Manning" on the English-language WP some time before he made this suggestion. Delicious carbuncle (talk) 15:14, 1 September 2013 (UTC)
    • Pictogram voting comment.svg Comment It seems that we don't have any images of Chelsea, all photos (that are categorized so far) are of the old Bradley. IMHO the redirect is OK right now. Renaming is not a big deal, we are talking about 15 images of Bradley and 15 images of supporters. Again, IMHO it is not a good idea to change the filenames. By keeping the status quo there will be a clear devide between old images of Bradley and new images of Chelsea (I hope we'll be able to get at least a few). Makes sense for users who might want to use before/after photo. Just my 2 cents. --Hedwig in Washington (mail?) 16:07, 1 September 2013 (UTC)
  • Pictogram voting comment.svg Comment I think both names should be acceptable in file names and I do not want edit warring about descriptions. People do change names and pseudonyms and having a rename operation after each such change would be awkward. If we change policy to allow the renaming we will start having rename wars and debates about whether the name really was changed, the new name legitimate or official enough etcetera. Even for new uploads, the old name may be legitimate, at least if the image itself is old. Descriptions could be altered to reflect the new name (and should probably for new photos), but mentioning the old name in descriptions of old photos is reasonable. The category should probably have the new name. --LPfi (talk) 17:19, 1 September 2013 (UTC)
  • Support renaming category, oppose renaming individual photos; as for descriptions/captions, the addition of parentheticals '(later Chelsea Manning)' or '(formerly Bradley Manning)' should resolve that. - Jmabel ! talk 19:45, 1 September 2013 (UTC)

Update: Descriptions are all updated now appropriately in the main cat. I don't think it's likely we'll see support for file renames (preferring instead to clarify the filenames in the descriptions - most of them were accurate as far as we knew at the time of upload), but I hope we can at least still get broad consensus for a category move. As such, per Multichill's suggestion, I've opened a discussion at Commons:Categories for discussion/2013/09/Category:Bradley Manning. Please leave your further comments there. Dcoetzee (talk) 06:09, 2 September 2013 (UTC)

URGENT

My friends! Unfortunately, some one are about to delete the personal image of Oscar Casares as you can see here. I've been asking for permits and pestering people with email from one side to the other, to appear someone and ruin other people's work. The Flickr page of Oscar Casares is Official, has more than 6 photographs (contrary to what is argued in the file for quick removal) and was created by his team with the objective of providing images under ShareAlike Creative Commons Attribution, as requested by Wikipedia. In this context, it is quite inconvenient erase the Oscar Casares image. Again, I ask for help from someone experienced, because I do not know how to prevent this misunderstanding.--António Lagos (talk) 23:49, 1 September 2013 (UTC)

Hi, I've converted the speedy deletion request to a regular deletion request, so it can be discussed for one week. If you can provide some evidence on Commons:Deletion requests/File:Oscar Casares (pintor).jpg to show that it's an official Flickr account, we should be able to keep it. :) darkweasel94 05:16, 2 September 2013 (UTC)
I hope Oscar Casares realizes that anyone can now take File:Nicole_Kidman_(Oscar_Casares).jpg and put it on a tshirt without paying him a dime. I've taken the liberty of uploading the images of his other paintings. Delicious carbuncle (talk) 06:02, 2 September 2013 (UTC)
Do you have evidence for "was created by his team" --McZusatz (talk) 07:41, 2 September 2013 (UTC)

September 02

Upload NEW version of a file??

Hi guys, how to upload on commons a new version of a file? I'm speaking about this file. One word was written partially incorectly, and i edited it, and now can't overwrite this file. If an admin read this post, please consider reuploading this file (new version http://i.imgur.com/LATFsGj.png). Argumentation: the name of sector "Central" is wrong. Officially it's name is "Centru" (from romanian mot-a-mot it's called Center, but not Central). Proof https://ro.wikipedia.org/wiki/Sectorul_Centru . Thanks --XXN (talk) 08:01, 2 September 2013 (UTC)

Hi, you need to be autoconfirmed to upload a new version of a file. So basically you just need to wait four days, unless an admin reads this and manually confirms you (you can also ask at Commons:Requests for rights). darkweasel94 08:07, 2 September 2013 (UTC)

Replace old Wikivoyage logo with new one (File:Wikivoyage-Logo-v3-icon.svg)

Hi. Please replace the old logo with the new one. See m:Wikivoyage/Logo 2013 for more info. Specifically, replace it in highly-visible pages, like Category:Sisterproject_templates. PiRSquared17 (talk) 14:30, 2 September 2013 (UTC)

 Doing… -- Rillke(q?) 14:34, 2 September 2013 (UTC)
Note: do not use a script to replace all instances, everywhere, at least not yet. Doing it in that category, and maybe some others should be good for now. There are still valid uses for the old one. PiRSquared17 (talk) 14:35, 2 September 2013 (UTC)
Note this request. Can this be somehow organized, please? -- Rillke(q?) 17:24, 2 September 2013 (UTC)

Copyright following corporate dissolution

Hi all, can anyone give an opinion as to what happens to a company's copyright on its dissolution where there is no successor in law? I am currently disputing the speedy deletion of File:EWO1920Ad2.jpg on the basis that EWO disappeared in the 1950's - The background is here. All advice welcome. Thanks in advance. Philg88 (talk) 18:45, 1 September 2013 (UTC)

  • There might be some countries that are exceptions to the following, but in most countries any intellectual property rights, like any other assets, will pass to some entity on the dissolution of a corporation. That can be a successor corporation, a government, or some rights that didn't originate with the corporation might revert to prior owners. Unfortunately, it can often be very hard for a third party to determine who owns these rights; it can often be hard enough for the rights-holder to determine, which can leave such materials effectively orphaned. The short is, images that would otherwise remain in copyright are indeed still in copyright, but now it would be much harder to determine whose permission you'd need. They lapse out of copyright at the same time they would if the corporation still existed. - Jmabel ! talk 19:54, 1 September 2013 (UTC)
  • More specifically, though, your claim of "own work" on File:EWO1920Ad2.jpg is obviously false. Even if the work is out of copyright, simple intellectual honesty requires attributing it correctly and accurately indicating your source. - Jmabel ! talk 19:58, 1 September 2013 (UTC)

┌─────────────────────────────────┘
Thanks Jmabel. Given the history behind Ewo Brewery, I suspect that the rights now vest in the government of the People's Republic of China. Rather than embark on the Sisyphean task of asking them for permission, I will instead move the files to en:wiki and claim fair use of promotional material. Thanks for the advice. Best Philg88 (talk) 06:25, 2 September 2013 (UTC)

Checkmark This section is resolved and can be archived. If you disagree, replace this template with your comment. Jmabel ! talk 19:20, 2 September 2013 (UTC)

Royal Air Force pics on Twitter

The RAF have just confirmed that pictures posted to their Twitter account @RoyalAirForceUK are "under an Open Government Licence (OGL) and are crown copyright". Does anyone have the wherewithal to run a script to hoover these up? Is there a tool for the job? Andy Mabbett (talk) 10:48, 2 September 2013 (UTC)

You could start a request at Commons:Batch uploading … --El Grafo (talk) 15:00, 2 September 2013 (UTC)
I would be concerned about the absence of good metadata on the tweets (in particular the lack of a convincing unique ID), compare with the detail available in this batch upload which contains a large number of RAF photos - I suspect the same images are hosted elsewhere with more data, I'll check it out. The problem with simply hoovering these, is that we are likely to get a host of duplicates which only vary by EXIF data; a nightmare to tease out afterwards. -- (talk) 21:32, 2 September 2013 (UTC)

PDF-ODF files cannot be uploaded

The upload script does not permit uploading hybrid PDF-ODF files at present (produced for instance by LibreOffice, OpenOffice), although it was possible in the past (ex. Deklaro pri livero de permesilo Krea Komunaĵo 001.pdf, Šablona české všeobecné plné moci (A4) 001.pdf). The PDF format is important to show the typesetting, nonetheless the possibility of editing is essential for some types of documents. --Petrus Adamus (talk) 19:26, 2 September 2013 (UTC)

we don't allow mixed zip-other format files currently [since about mediawiki 1.17]. (Odf is a zip format). This is to ensure that gifar type files (hidden java applets) don't get uploaded. There's a bug about this somewhere in bugzilla. The solution would probably be to special case pdf/odf and write code to validate both parts of file (additional issue is odfs aren't on the allowed file type list)Bawolff (talk) 21:42, 2 September 2013 (UTC)
The related ticket is bugzilla:28188. --AKlapper (WMF) (talk) 11:28, 3 September 2013 (UTC)

September 03

Turning off POTY-App for old candidates?

I recently clicked on the link Commons:Picture_of_the_Year/2012/R1/Lyapunov-fractal.png and the page initially loaded with a popup which eventually ended up reading: "POTY-App experienced an error. The data that will be saved and publicly visible if you send the report: Your username, a timestamp, what the App did immediately before and "poty.data[poty.galleryType] is undefined TypeError @line862"". Since the competition is no longer live for such images, would it be possible to turn off the app for them, so that people can just browse those pages without encountering the error? It Is Me Here t / c 10:17, 3 September 2013 (UTC)

Until modified for the Wikivoyage logo selection, it did not throw an error ;-) It is disabled now. -- Rillke(q?) 10:31, 3 September 2013 (UTC)

<Gallery> tag

Hi all.

We've been experimenting with changing the output of the <gallery> tag. We added a new parameter to the gallery tag - mode. This allows triggering a number of experimental modes for the gallery tag. The values of the mode tag can be traditional, nolines, packed, packed-hover, and packed-overlay. This mode parameter is available on all non-wikipedia wikis at the moment (Wikipedia will have it available on thursday) I would really like commons folks to play with it, and see what you think. If people really like it, we could make one of the new modes the default for the gallery tag. Personally I think the "packed" mode would be good for commons. Changing the auto-generated "galleries" on category pages and on special:newfiles is also an option if people like the new version (I think packed-hover would work well for categories). See mw:Help:Images#Rendering_a_gallery_of_images for full details about the options available for the gallery tag. Example gallery:

Anyways, I really would appreciate all your feedback and comments. Thanks. Bawolff (talk) 20:22, 19 August 2013 (UTC)

Thanks for posting and for the wonderful work. In the packed example on mww, are the underendered "pipe trick" links filed as a bug already? Can't check now. --Nemo 21:39, 19 August 2013 (UTC)
Yes, that's a known bug (I totally forgot about that). It happens on all tag extensions. Bawolff (talk) 22:30, 19 August 2013 (UTC)
I've got a question regarding the size of images. I quickly tested both, the widths as well as the heights gallery options with mode="packed", and both didn't work as I'm used to. The gallery images somehow changed their size, but neither did widths limit their widths to the given value nor did heights limit their heights. This even applied to a gallery containing only a single image (where widths/heights don't need to be adjusted to match the other images and the width of the page). Is this a design limitation somehow or an issue that will be fixed eventually? --Patrick87 (talk) 22:09, 19 August 2013 (UTC)
It tries to adjust the size in order to make all rows full the screen (So no ragged edges on right). heights parameter is basically ignored. Widths parameter is used as an initial size. If there's extra room on the end of a row, the width of each image is used as the initial size, and then the widths are increased until the row fills the browser width (or until you hit 1.5 times the specified size, which is taken as the maximum). Bawolff (talk) 22:29, 19 August 2013 (UTC)
OK, I see. Although I think you meant widths is ignored while heights is used as an initial value and is increased up to 1.5*heights to fill rows?
I'd propose one enhancement to the current algo: The last line currently has almost always a height of 1.5*heights (because it doesn't fill the row). If the algorithm detects it can't fill the row with 1.5*heights it should fall back to e.g. the mean height of the previous lines or (in case of only a single line) 1.0*heights. If there is only one line, the specification of heights should be obeyed, if there are multiple lines, the last line should not look larger than the other lines needlessly (as the current algo tends to). --Patrick87 (talk) 23:31, 19 August 2013 (UTC)
I'll try and do some experiments to see how that looks. Bawolff (talk) 16:55, 20 August 2013 (UTC)
I did some experiments over here - http://tools.wmflabs.org/bawolff/gallery/index.php?title=Featured_pictures/Non-photographic_media . I switched it so that it will try to fit the last row first (with up to 1.5 times the size). If its impossible to fit it with 1.5 times the size, it will then multiply the size by the largest multiplier so far (So if a previous row had been multiplied by 1.5, it would use that. If the biggest multiplier is 1.3, then that's what will be used for the last row. If there is only one row, it will first see if it can fit it with 1.5 times the size, failing that it will display the images at 1.0 times the size). Do you (and everyone else here) think that behaviour is better? (I think its better). Bawolff (talk) 18:08, 21 August 2013 (UTC)
Great, thank you for the tests! Yes, it looks much better, line heights are much more even with your changes. Three questions, though:
  • Why did you decide to fall back to the maximum multiplier that occured before when the last row can't be filled, instead of a mean value? This would make heights even more similar. The maximum previous multiplier is often quite high (e.g close to 1.5), whereas the mean multiplier is often substantially lower.
  • Did you think about lines in between that can't be filled? They set the maximum multiplier to 1.5*heights, therefore the last line will be 1.5*heights, too. Even if all other lines are lower and only one single line can't fill a row.
  • Do you think the same algo as for the last line could also be useful for lines in between that can't fill a row? I'm not sure personally, since it will make the gallery look more "ragged", but it will also make heights more even and might still look better then.
--Patrick87 (talk) 20:39, 21 August 2013 (UTC)
I updated the test wiki to use average instead of maximun (please test and tell me what you think), I agree that looks better. I don't think the algorithm for the last row should be use for the others (I think the ugliness of more ragged edges are more than the ugliness of larger images). Most of the time, the non-last line won't be too bad for having to resize to make the line fit. Bawolff (talk) 22:25, 21 August 2013 (UTC)
Wow, thank you for the blazingly fast update Face-smile.svg. I think it looks great in most cases now. I'm looking forward to see this online on other Wikis and to hear what other people say. --Patrick87 (talk) 23:40, 21 August 2013 (UTC)
Cool. Thanks for your feedback. 00:04, 22 August 2013 (UTC)
I missed the deadline to get those changes into the deployment on monday, so those changes are probably going to go live the monday after (September 2, 2013). Bawolff (talk) 23:41, 23 August 2013 (UTC)
If some new mode is made default, is there a way to force, in one’s preferences, the grownup’s old and boring mode as one’s default? -- Tuválkin 08:41, 20 August 2013 (UTC)
Probably not (Although maybe a gadget could be created). Preferences have a tendency to increase complexity just because the number of options explode, so we try to avoid creating new ones unless there is a really good reason why some people would want to view something differently. Additionally we also try to avoid introducing new preferences that split the parser cache (cause different html to be rendered) because it reduces cache efficiency. Bawolff (talk) 16:55, 20 August 2013 (UTC)
Well, if it can be turned off, then it should never be changed into the default behaviour. That’s basic respect. -- Tuválkin 23:33, 21 August 2013 (UTC)
Like most configuration changes, individual wikis can decide as a group (by which I mean have a discussion to determine community consensuses) whether or not they want the feature on. Of course any such discussion would happen after any initial bugs are worked out, and after people have had time to play with the feature so they can make an informed decision. I'm also looking into adding a url parameter to category pages/special:newfiles so that people can experiment with autogenerated galleries too. Bawolff (talk) 23:41, 23 August 2013 (UTC)
Hmmm, that’s funny. The gallery on User:PierreSelim uses packed-over ; but when viewing it logged in on Firefox, I have the rendering of packed ; output is as expected unlogged with Chromium. Jean-Fred (talk) 09:28, 20 August 2013 (UTC)
I've the same problem when I'm logged with this account, but not with my second account. There is probably a problem with a gadget or a personnal script. Pyb (talk) 10:17, 20 August 2013 (UTC)
That's really odd. Looking into it. Bawolff (talk) 16:55, 20 August 2013 (UTC)
I think the old css (prior to the change being introduced) is still cached possibly. Does doing a hard refresh fix the issue? (This sort of thing should automatically happen, but sometimes doesn't seem to) Bawolff (talk) 16:58, 20 August 2013 (UTC)
For me, it did. Caption hidden and shown on hover after purging browser's cache. -- Rillke(q?) 07:22, 21 August 2013 (UTC)
Same here. All good now, thanks for the tip. Jean-Fred (talk) 12:03, 21 August 2013 (UTC)
That's really good news. Thank you for this major improvement! -- Rillke(q?) 12:41, 20 August 2013 (UTC)
+1 WOW!--Steinsplitter (talk) 12:47, 20 August 2013 (UTC)

Neat. --Michaeldsuarez (talk) 12:44, 20 August 2013 (UTC)

Like it! --Kippelboy (talk) 05:01, 21 August 2013 (UTC)
+1 DOUBLE WOW. Very beautiful. --SJ+ 09:50, 21 August 2013 (UTC)
This makes galleries look WAY more modern. Good job! Looking forward to seeing this rolled out all over. — Scott talk 12:04, 21 August 2013 (UTC)
In my opinion, this is totally not neat nor wowworthy. It is going to break really ugly when instead of carefully picked examples and preset galleries (for which we had already presentation tools other than <gallery />…) it starts to be used for categories. Any image much higher than wide or much wider than high, any icon of a non-visual media file, any thing that doesn’t match the narrow-minded preconceptions behind this “innovation” — will break the whole look stacked/packed look. Leave this kind of trendy eye-candy for Microsoft, lets forcus on work here. And what works, for an all-emcompassing presentation of visual elements of unpredictable aspect ratios is… yes: have each of them padded to square cells, which is what we have. Stop this nonsense and do some real work instead — there’s a lot waiting to be done. (I mean this for categories, only — for the rest, especially wikipedia articles, these new changes wont make it worse: Even with better looks, galleries is still lazy article editing.) -- Tuválkin 23:33, 21 August 2013 (UTC)
Well at some point, there's a limit to how well we can display something with extreme aspect ratios. Very wide and short images do end up taking up a lot more room, which can look less than ideal if they are also low resolution. However, I don't think the results are particularly horrid with such images. (obviously that's a matter of opinion, and I have quite a bit of bias). Example gallery with image choices that look less "good": Bawolff (talk) 00:06, 22 August 2013 (UTC)
Traditional gallery
packed
An example of a gallery with very wide images, that looks a little better is, if you do the featured pictures of Space exploration. - User:Bawolff/Space,_the_final_fronteir Bawolff (talk) 20:28, 22 August 2013 (UTC)
It looks like there is a compatibility issue with perrow. See below using perrow="4" ! — t a r u s¡Dímelo! 03:28, 23 August 2013 (UTC)
Given that the packed mode has images with same height, but very different widths, the number of images in a row depends a lot on the aspect ratio, and each row could have a significantly different number of images on it. I don't think we can easily make such an option work with a the new mode (Future versions of the code will totally ignore the perrow option). However, if you want to limit how long the rows are, you can use code like:
<div style="width:60%">
<gallery mode="packed-hover">
File:Russell Falls Mt Field National Park.jpg|Russell Falls in Mt Field National Park, Tasmania, Australia
File:Caracol Falls.jpg|Caracol falls, near Canela, Rio Grande do Sul, Brazil
File:The Spit Bruny Island.jpg|The Spit, Bruny Island, Tasmania, Australia
Image:Male-total.jpg|Malé, capital of Maldives
Image:Mont Blanc depuis la gare des glaciers.jpg|The top of Mont-Blanc seen from the Gare des Glaciers
Image:Eyjafjallajökull.jpeg|Eyjafjallajökull (Iceland)
Image:Stromboli Eruption.jpg|Stromboli volcano eruption, Italy
Image:Lower_Antelope_Canyon_478.jpg|Antelope Canyon in Arizona
Image:GeysirEruptionNear.jpg|Eruption of Strokkur at close range
Image:Alpamayo.jpg|Alpamayo, Andes, Peru
</gallery>
</div>

To make:

(You can use other units like px instead of % too) Bawolff (talk) 23:41, 23 August 2013 (UTC)


Just as an update, per request of (somebody?) you can see what categories and special:newfiles would look like with the different gallery via a url parameter gallerymode on test.wikipedia.org (The url parameter will work on commons on monday) - https://test.wikipedia.org/wiki/Special:NewFiles?gallerymode=packed Bawolff (talk) 21:52, 29 August 2013 (UTC)

Thanks for all the feedback. Given a large amount of it has been positive, and a number of the initial bugs have now been fixed, I would like to propose enabling this gallery code by default. To that end I've created Commons:Requests_for_comment/Changing_default_gallery_mode. Bawolff (talk) 16:15, 4 September 2013 (UTC)

<Gallery> tag – nolines option

I just experimented a little bit and noticed the nolines version of the gallery isn't aligned nicely in vertical direction. Would it be possible to align images in the middle vertically and to align the captions so they start all at the same vertical position? E.g. exactly like the "traditional" gallery format with borders and padding removed? --Patrick87 (talk) 17:18, 4 September 2013 (UTC)

August 20

File does not display existing license

File:La Cène de Guillaume Fouace, église Notre-Dame, Montfarville, France.jpg has a proper license, but does not display it (for me at least). Anybody knows why? I can not figure it out. I know we can probably move the license to a different location, but I am trying to figure out what is wrong with this syntax. --Jarekt (talk) 13:21, 3 September 2013 (UTC)

Moving the license templates out of the Artwork-template solved the problem. However, such a problem shouldn't occur. --Túrelio (talk) 13:28, 3 September 2013 (UTC)
Same in File:La Basilique Saint-Pierre de Rome de Guillaume Fouace, église Notre-Dame, Montfarville, France.jpg or File:Le Sermon sur la Montagne de Guillaume Fouace, église Notre-Dame, Montfarville, France.jpg. I know we can fix it by moving the license, but lets keep the license where it is and try to fix the problem with the template, or whatever else is causing the issue.--Jarekt (talk) 13:53, 3 September 2013 (UTC)
Same for categories included in templates: it shouldn't be done: hotcat and cat-a-lot can't do their jobs.~~---- — Preceding unsigned comment added by Havang(nl) (talk • contribs) 13:57, 3 September 2013 (UTC)
I gave up and fixed the issue by reshuffling the fields in the template. I do not know why the previous version did not work. It makes no sense. --Jarekt (talk) 13:28, 4 September 2013 (UTC)
I had the same problem with the Artwork-template a couple weeks back and end up fiding the same kind of solution (see most items this category). -- Tuválkin 17:22, 4 September 2013 (UTC)

Something went wrong at upload

Hello, something went wrong at upload of this file File:Amettes (Pas-de-Calais, Fr) maison natale Saint Benoît Fabre, extérieur.JPG, two files uploaded at once; the first thumbnail is a bad one, buit it seems to block the upload of the other file. Can some-one delete that bad lowest thumbnail, it probably solves my problem. --Havang(nl) (talk) 16:01, 4 September 2013 (UTC)

This does not work. Neither deleting nor moving works. API request failed (unknownerror): Unknown error: "backend-fail-synced" at Wed, 04 Sep 2013 16:15:43 GMT served by mw1207 -- Rillke(q?) 16:17, 4 September 2013 (UTC)
Storage backend problem, sometimes it resolves itself after some synchronization several days later but sometimes it does not resolve itself. It's a long standing bug (seen it too often) with two uploads stored by the same user at the same time, probably some kid of duplicate saving/backend sync error upon saving. The same problem at File:St Bernardusabdij Westkant.JPG--Denniss (talk) 18:08, 4 September 2013 (UTC)

September 05

I can't recover my password

Hi, When I try to get back my password (recuperar contraseña) https://commons.wikimedia.org/wiki/Special:PasswordReset, it sends me the message "El nombre de usuario Tingundan no está registrado en este wiki, pero existe en el sistema de identificación unificada." and then I can´t get back the password neither create the user again because already exists. Should I have to create a new user? -- 15:26, 5 September 2013‎ 200.72.45.10

Hola. Prueba en es:Special:PasswordReset (es.wikipedia) que es el homewiki asignado. Un saludo. Alan (talk) 17:35, 5 September 2013 (UTC)
Te habrá llegado al email desde el que registraste el usuario una nueva clave temporal.
Restablecimiento de contraseña

Se ha enviado un correo electrónico para el restablecimiento de tu contraseña. 
Un saludo. --Alan (talk) 18:04, 5 September 2013 (UTC)

Freedom of Panorama for transient sculptures

Commons:FOP#Permanent vs temporary says that it's fine to use FoP to create free photos of transient sculptures (such as ice or sand sculptures). However in a recent deletion request it was suggested that this could use further examination. There doesn't seem to be much legal precedent in the area. From the point of view of the sculpture, it's on display for as long as it lasts. However from the point of view of the location, the sculpture is intended to be temporary. Can a transient object be considered to be "permanently situated in a public place or in premises open to the public" (as per UK copyright law)? I have no idea how a court would judge it.

Is there anything more that could be done, such as obtaining a legal opinion, or adding a template to such images to warn reusers that it's a legal gray area? Or is the advice in Commons:FOP#Permanent vs temporary perfectly sound? ghouston (talk) 01:19, 31 August 2013 (UTC)

There was a case in Germany about an ice sculpture which ruled that way, I'm pretty sure, or maybe some legislative commentary on the wording in their law. I don't know of anything which contradicts that. When you think about it, it does make some sense. For countries which do have FOP, they are basically saying that normal derivative rights on works which are part of day-to-day public life can intrude too far, so those rights are scaled back in those situations. However, the "permanent" qualification means that temporary display like an exhibition should not deprive authors of those rights -- authors should continue to enjoy full derivative rights. But for something like an ice sculpture or sand castle... it's not like it can ever be used again, or put up for display, be sold, etc., so preserving those rights don't make much sense. The works entire purpose was to be on public display for the entirety of its life, really. Carl Lindberg (talk) 20:35, 31 August 2013 (UTC)
Just to note, the section on the FOP page came about via this discussion. It has a cite and some links, though they appear to be dead links now. Carl Lindberg (talk) 20:42, 31 August 2013 (UTC)
Thanks, that helps me understand where that wording came from. A word like "permanent" will have a legal meaning, which may or may not relate to how it's used in normal English. "Permanent employment" doesn't mean you will be employed for the rest of your life, it means that there isn't any fixed termination date. The point about "day-to-day public life can intrude too far" seems to have a big problem when sculptures can be placed in a public place for years at a time, and still not be "permanent" in a legal sense. How long will it be before a sculptor allows a work to be displayed for 999 years, like you find in legal leasehold agreements, so that they can still make use of the copyright until it expires? (see also Commons:Deletion requests/Files in Category:Genghis Khan sculpture, which I thought was a slightly different issue, but now seems to be the same one.)
For the ice sculptures, there could still be a problem, e.g., if space was allocated for the exhibition for a week or so, after which the intention was to clean it all up. Then it's still a "temporary" exhibition. The idea that "preserving sculptors' rights in transient sculptures makes no sense" could be argued, since they can still create or licence derivative works such as photos and movies.
This would have been a lot easier if copyright law had said FoP applied if a work was intended to be displayed for at least 3 months, or whatever, and addressed transient works explicitly. ghouston (talk) 00:17, 1 September 2013 (UTC)
I think their definition of permanent includes the notion that a work will be on display for its natural life. If an artist planned to move the ice sculpture into cold storage so he could display it at arbitrary later times, that is probably where issues would come up. If the ice sculpture was going to be destroyed (or melt) at the end of the exhibition, it would still be permanent. Carl Lindberg (talk) 20:45, 1 September 2013 (UTC)
Note that "permanent" may mean different things in different countries. The wording on COM:FOP only seems to be based on German law. --Stefan4 (talk) 11:55, 6 September 2013 (UTC)

Newer, better looking Campaign pages coming soon!

Hello everyone!

I've been working on making Campaign pages more useful. Details about the system are at https://www.mediawiki.org/wiki/User:Yuvipanda/Campaigns_namespace_proposal, but even better - you can see a live demo! I've also added two extra properties, 'title' and 'description' to the JSON config of the Campaign namespace, that lets you specify wikitext for the title and description that is shown on the newer Campaign pages.

The idea being that you can pass on links to the Campaign page itself and people can upload from there. They can also see a description, and other images that have been uploaded (it currently shows the last 24 images). This is far more useful than the current version, http://commons.wikimedia.org/wiki/Campaign:wlm-nl, which is just a JSON structure. We also record (privately) how many people sign up from each campaign, and will build out a dashboard for this eventually. It also has some statistics about uploaders, and we can add more as requests come in.

This will be deployed on Tuesday (9th September 2013) to commons. For people who still want to look at the current JSON structure, you can use the new tool I built - masscamps, which lets you do that for a lot of campaigns effectively and efficiently. I'll be adding more features to the tool as people ask for them, so feel free to leave a message on my talk page. Thanks! YuviPanda (WMF) (talk) 21:52, 5 September 2013 (UTC)

I started reading this and when I got to «making Campaign pages more useful» I was like yay, but then as I read more I grew disappointed because the changes was not what I expected. What I was hoping to read was that the whole thing was being taken down and that WMF was going to focus on enabling users’ work in a better environment, both technical (less bugs, more stability, adding features requested by the community, more and better servers, etc.) and social (gearing the UI to minimize targetting by vandals and trolls, and discourage frivolous misuse of the projects by fence-sitters): Sadly, it aint so. -- Tuválkin 00:40, 6 September 2013 (UTC)
This is certainly good news for curious visitors; for campaign managers the table is however more useful. Thank you for building a tool so WLM can run without disruption. There is a bug full of the pros and cons so we don't have a lot to discuss here, I think. -- Rillke(q?) 09:45, 6 September 2013 (UTC)
Indeed. I've poked the people involved in that bugzilla thread on IRC about the tool and they seem mostly happy about it. I'll implement any changes if they're requested. YuviPanda (WMF) (talk) 21:14, 6 September 2013 (UTC)

September 06

Wikimedia UK to offer photography workshops for volunteers

Hello everyone. Wikimedia UK is looking at hosting some photography training workshops for Wikimedia volunteers. If you're interested please take a look at this section on the Wikimedia UK Water Cooler. Stevie Benton (WMUK) (talk) 12:45, 6 September 2013 (UTC)

Pages going blank, can't resolve

Right now I'm editing in Firefox because I can't even view this page in my normal browser: IE 8, Windows 7, Monobook skin. When I try to load COM:VP or COM:ADMIN, the page goes completely blank; the only way I can view anything other than whiteness is by going to the history page and clicking "edit", although previewing or looking at a specific revision of either page (e.g. [2]) has a similar result, with the added "feature" of producing the "Internet Explorer has crashed" message. I've only encountered this on these two pages; I've had no trouble uploading, and pages such as Commons:Administrators/De-adminship are loading without difficulty, and I've had no problems with the same browser/operating system/skin anywhere at en:wp. What could be the problem? Nyttend (talk) 03:17, 7 September 2013 (UTC)

Does it happen with other browsers too? Does it happen with the Vector skin? In general, I wouldn't suggest that anyone use any version of Internet Explorer. darkweasel94 06:08, 7 September 2013 (UTC)
It works fine in IE10 and Chrome on Windows 8 (vector skin). Ruslik (talk) 16:28, 7 September 2013 (UTC)

"This file might be corrupt, or have the wrong extension."

When trying to upload photos from the Environmental Protection Agency, I get the error "This file might be corrupt, or have the wrong extension." - The pictures display in the preview but Commons won't take them. The Houston Chronicle article "EPA photos of Texas from the 1970s." compiled by Dana Guthrie and posted on September 3, 2013 shows a series of photos taken in the 1970s by Marc St. Gil of the EPA.

Filenames of some:

How do I fix them?

Thanks WhisperToMe (talk) 18:18, 7 September 2013 (UTC)

Those are GIF files. Just change the file extension to *.gif and you should be fine. --Patrick87 (talk) 18:23, 7 September 2013 (UTC)
Thank you so much! I am uploading them WhisperToMe (talk) 18:49, 7 September 2013 (UTC)

Both Flickr to Commons bots down?

I can't get either of these to work for the last couple of days. Flickr2commons won't load images with valid Flickr URLs, see https://meta.wikimedia.org/wiki/Talk:Flickr2commons#Message.2C_Invalid_photo_ID

Bryan's Flickr upload bot leads to a "You can't edit this main page" screen that is utterly irrelevant to the task being attempted. Specifically:

View source for Main Page ← Main Page

You do not have permission to edit this page, for the following reasons:

  This page is currently protected, and can be edited only by administrators. 

My setup: MacOS 10.6 + Firefox 23

Help? TIA, Pete Tillman (talk) 19:25, 30 August 2013 (UTC), a regular Flickr -> Commons prospector, see Category:Files uploaded by PDTillman, with many hundreds from Flickr

Did you check the file name? Maybe it IS protected? The main page message certainly is a bug, no doubt. --Hedwig in Washington (mail?) 21:15, 30 August 2013 (UTC)

  • For other users:

http://wikipedia.ramselehof.de/flinfo.php

Help page for Flinfo: https://commons.wikimedia.org/wiki/User:Flominator/Flinfo

-- is a reasonable alternative, and it WORKS. Cheers, Pete Tillman (talk) 17:48, 2 September 2013 (UTC)


For Flickr2Commons http://toolserver.org/~magnus/flickr2commons.php

-- the trick (for a single picture) is to use ONLY the Flickr ID#, eg for http://www.flickr.com/photos/reykjavikmuseumofphotography/8076826134/ enter 8076826134

This wasn't at all obvious (to me) in the online instructions. HTH, Pete Tillman (talk) 21:15, 4 September 2013 (UTC)

I think the problem with Flickr Upload bot is HTTPS. I desactivated it in my preferences. Nemesis III (discuter) 14:22, 8 September 2013 (UTC)

Unidentified plants

I always get frustrated when I try to categorise plants. Latest example File:Lagune Lamraro - Ambavarana 11.jpg. I understand the reason that everything is classified in latin, but it is not made easy. There is no guide to at least identify the correct upper classification. It would help if there are at least picture examples in the main categories. There are a lot of uploaders who are ready to do some research if it is made easy. Most uploaders dump the pictures in "Unidentified plants" (983 entries and 182 latin subcategories) or dont even bother about classifying interesting plants in the uploaded pictures. Smiley.toerist (talk) 13:36, 4 September 2013 (UTC)

You may get help if posted at Wikipedia talk:WikiProject Plants. JKadavoor Jee 13:57, 4 September 2013 (UTC)
Searching on common names should work, given the common name in the language in question is given at the category page. Knowing the common name one can also go via Wikipedia, where scientific names should be given for any species (or higher taxa). If the problem is identifying the plant in the first place, there is little one can do here. A guide at Wikibooks would probably be the best approach. Adding picture examples to any of the categories is probably welcome, as long as they are reasonably representative. --LPfi (talk) 11:00, 5 September 2013 (UTC)
Do we have a similar Wikiproject xxx for animals/reptiles and/or Lizards? I have File:Lizard in Anja Park II.JPG, File:Lizard in Anja Park.jpg and File:Fort Dauphin rocks 11.jpg. Smiley.toerist (talk) 11:16, 8 September 2013 (UTC)

Number of watchers

Every revision history page has a link to the Number of watchers - the MZMbride tool which is disabled. Should we remove the link or replace it with something else?--Ymblanter (talk) 07:12, 5 September 2013 (UTC)

Replaced it. In theory, we could even remove the second link (Page view statistics) because it is included in the page information. -- Rillke(q?) 08:19, 5 September 2013 (UTC)
Thanks. Yes, I think it might be reasonable to leave only the links to the page info (may be explaining what it means).--Ymblanter (talk) 09:55, 5 September 2013 (UTC)
I don't think we should remove the page view statistics link because it further buries the information (adding an additional click if one is already on the history page). I've modified Rillke's change slightly by appending #mw-pageinfo-watchers to the link so it brings users to the number of watchers like before. The general page info page is already linked in the Toolbox on the sidebar. Killiondude (talk) 00:35, 8 September 2013 (UTC)
How are the page view statistics and numbers of watchers related to the page history? I guess it was added there in the first place because there was no page information page but now, there is. IMHO the only valid position for this kind of links is the page information page where the statistics tool is explicitly and correctly listed as an external tool. -- Rillke(q?) 08:31, 8 September 2013 (UTC)
As an aside, it should be noted that the info action only shows number of watchers for pages with more than 30 watchers if you are not an admin (Don't have unwatchedpages right which currently only admins have). Bawolff (talk) 13:59, 8 September 2013 (UTC)

SVG masks not rendering

So I guess libRSVG isn't very good at rendering SVGs correctly, but with masks it fails completely to display them—see File:SVG mask.svg. Time to switch to a different renderer???—Love, Kelvinsong talk 12:57, 7 September 2013 (UTC)

The lacking rendering capabilities of librsvg are a known problem but are only considered low priority by WMF. I tried to make them aware of the problem that librsvg is essentially umaintainded in bug 51555 and suggested to support development to get those bugs fixed or switch to another renderer otherwise, but there was not much promising feedback since.
There's already a long list of librsvg bugs in MediaWiki's Bugzilla (and even more in GNOME's bugzilla which librsvg is part of). Feel free to add your bug if it is not already present and voice your opinion. Maybe at some point WMF realizes the importance of the problem, too. --Patrick87 (talk) 14:08, 7 September 2013 (UTC)
It's not so much missing features which are annoying, as inconsistencies in some of the features which are claimed to be supported (such as font rendering)... AnonMoos (talk) 02:24, 8 September 2013 (UTC)
Okay I reported bug 53899, but looking at the history of RSVG bug reports, I doubt anyone's going to fix it… -_- —Love, Kelvinsong talk 12:13, 8 September 2013 (UTC)
yeah, sorry but its unlikely to be fixed in the near term. Bawolff (talk) 13:51, 8 September 2013 (UTC)
That's encouraging.—Love, Kelvinsong talk 15:27, 8 September 2013 (UTC)

September 07

Only display language of the user?

Can Template:Translation table be modified to only display the language of the user? There doesn't seem to be any point in displaying all these other languages. For an example, see Category:Université Paris Descartes. Or is this a misuse of the template? --Robert.Allen (talk) 23:08, 7 September 2013 (UTC)

The use of "langswitch" type templates for things other than interface messages has been controversial in the past. For example, it makes it highly inconvenient to check for translation errors if you can't see more than one language at a time... AnonMoos (talk) 01:44, 8 September 2013 (UTC)
In case of {{LangSwitch}} you can always look at the source to see all the languages on the same page. That is not the case with most other internationalization approaches. --Jarekt (talk) 01:51, 8 September 2013 (UTC)
That means you have to edit the page before you can even check whether or not there might be a problem, which is distinctly inconvenient. Also, what if a there is no description provided in the user's own language, but there is a description in another language that the user happens to know? -- AnonMoos (talk) 02:04, 8 September 2013 (UTC)
In this example, it is only the title of the page that is translated. I see no reason to display the title in every possible language. The translated title only needs to be displayed in the language of the user, and this is really only important if the translated title differs from the one given to the page on Commons. If there is a problem, it should be seen immediately by users of these other languages, and they are the ones who can fix it. Other users shouldn't have to deal with all these other languages. I agree with Jarekt, the LangSwitch template should be used for this purpose. --Robert.Allen (talk) 09:49, 8 September 2013 (UTC)
There is no such thing as the «the language of the user», people can and some do speak several, not to mention that many users can take advantage in seeing file and category page content even in languages they are not fluent on. (The phrase «have to deal with all these other languages» is so comically evocative of the stereotypical anglophone monoglot I cannot stop grinning.) Anyway, Special:Preferences#mw-prefsection-gadgets includes two checkboxes that people scared by language diversity should perhaps tick on:
  • MyLangNotify: Emphasizes your language description, or asks you to add one if it is missing.
  • Language select: Collapse foreign languages on multilingual pages (i.e. pages using {{mld}} or {{en}} or similar) and create a language selector
-- Tuválkin 17:41, 8 September 2013 (UTC)
Personally I am in favor of as much internationalization of the site as possible, but what we're talking about here is the language preference setting at the top of the page. I suppose we could suggest that it be changed to allow for display of multiple languages at one time, i.e., check boxes could be added. That could be used to accommodate your concern. Users who do not want to see all these other languages should not have to. A lot of progress in internationalizing Commons has been made, but turning it into an unreadable Tower of Babel is not the right approach. --Robert.Allen (talk) 19:46, 8 September 2013 (UTC)
Robert.Allen -- Re: your message of 09:49 -- You may think that way, but there are a significant number of other people who regard the simplistic rigidity and inflexibility of the "langswitch" approach to be highly annoying, when "langswitch" is used for other things than what it was originally intended for (i.e. user interface messages, not image or category descriptions). It may be OK when there are a few limited deterministic alternatives ("oil on canvas" vs. "engraving" or whatever), but when applied to free-form text fields, the narrow langswitch approach succeeds in making information inaccessible in many cases, more than it succeeds at anything else... AnonMoos (talk) 22:40, 8 September 2013 (UTC)

September 08

Possible copyright problem

Could I ask someone with more knowledge of copyright matters than I have (which isn't hard to achieve) to take a look at:

To my uneducated eye, the copyrights appear to belong to w:Studley, Inc.. However, the uploader asserts here that he has no professional connection with that company, and has "verbally been given approval" to upload one of the images. Is that how we do things here? Justlettersandnumbers (talk) 09:18, 8 September 2013 (UTC)

No, that's not how we do things here. We need a written permission from the copyright holder that the file(s) may be released under an appropriate license. See Commons:OTRS for details. --El Grafo (talk) 14:21, 9 September 2013 (UTC)
Thanks for your reply; and yes, I thought not. So what is the next step for those files? Should they be proposed for deletion? Justlettersandnumbers (talk) 16:51, 9 September 2013 (UTC)

Duplicate files: how to merge information

The files File:Andringitra National Park Office.jpg File:Andringitra, Madagascar by Effervescing Elephant-11.jpg are technicaly identical. Both from Flicker. (by the way: It has nothing to do with Isalo national park, the name used in Flicker). Wich one has to kept under wich name? Smiley.toerist (talk) 11:30, 8 September 2013 (UTC)

You can keep any of them but one that is kept may need to be renamed. Ruslik (talk) 17:30, 8 September 2013 (UTC)
I checked. The picture is duplicate within Flicker, but from the same uploader, Chris. Smiley.toerist (talk) 14:19, 9 September 2013 (UTC)

File:Andrea Previtali called Cordeliaghi - Saint Francis Collecting the Blood of Christ - Google Art Project.jpg

Hi, the painting has the metadata of Saint Francis Collecting the Blood of Christ, but it is Portrait of a Man (like this). How to get the right metadata? --jed (talk) 18:41, 8 September 2013 (UTC)

I guess it is fixed now. See Category:Ritratto di uomo (Andrea Previtali). -- Tuválkin 03:33, 9 September 2013 (UTC)

September 09

Font change

Today the edit screen font seems to have been changed to some non-proportional font. Why? Was this discussed anywhere? Wutsje 19:52, 3 September 2013 (UTC)

Did you notice this in Visual Editor, the Vector skin edit window, or in the grownup’s edit window? If the latter, that doesn’t surprise me, as it is an HTML TEXTAREA element, which are recomended to show a monospace font. Maybe you changed your browser to its defaults? -- Tuválkin 23:14, 3 September 2013 (UTC)
My apologies, I should have made clear that I'm using FF 23.0.1, Win 7 and Monobook (no VE for me, tyvm). Anyway, I didn't change anything in my config or prefs, the change suddenly was there. Wutsje 05:14, 5 September 2013 (UTC)
I observed the same change (using Chrome, Win XP; Monobook, of course no VE; does not occur w IE 8, Win XP), rather unwelcome change. --Túrelio (talk) 07:14, 4 September 2013 (UTC)
I can't reproduce it with Firefox Aurora on OS X 10.8. Everything as usual. darkweasel94 07:25, 4 September 2013 (UTC)
It's very unwelcome - to see if I am reading onl or editing is not so easy any more. -jkb- (talk) 07:29, 4 September 2013 (UTC)
I can confirm this with FF 24 beta/Windows 7. And I don't like it too. Please check Special:Preferences#mw-prefsection-editing, option Edit area font style. It is set to Browser default by default. Something changed with the last MediaWiki update. I guess the CSS for the edit window was changed. To solve this issue set the option to Monospaced font. Raymond 07:33, 4 September 2013 (UTC)
That worked. Thanks. It's a bit frightening that a change in MediaWiki does influence something that is set to my "Browser default". --Túrelio (talk) 08:33, 4 September 2013 (UTC)
Works for me too. Thanks! Wutsje 05:17, 5 September 2013 (UTC)
Don't be scared. Something, initiated by UniversalLanguageSelector, explicitly applies fontFamily sans-serif. Not everywhere, but on language subpages like that one. First, it sets fontFamily to an empty string, then it sets sans-serif despite I have told it in my prefs that I wish to use the browser's default font. So it is a bug. -- Rillke(q?) 16:22, 4 September 2013 (UTC)
Is the bug reported? I do want to use browser default and be able to set that default in the browser. --LPfi (talk) 08:55, 10 September 2013 (UTC)

September 04

Nudity images

I know we have a policy, COM:PENIS, which states we really dont need more of that subject uploaded. I want to know if there are other categories, outside nudity, which we have ever considered for placing a limit on. I am concerned about the categories for cups of coffee: Category:Cups of coffee, Category:Coffee cups and all the similar ones. we have a huge number of them, basically everyone with a camera phone, in a cafe, thinking "hey, why dont i take a photo of this cup of coffee?". granted, its not as serious an issue as excessive nude shots, but i think we should have some way of limiting images in this area as well.Mercurywoodrose (talk) 03:46, 5 September 2013 (UTC)

  • nopenis is NOT POLICY. -- it is the opinion of a faction of editors, which has NEVER been discussed, voted on, & approved by the community as a whole. Lx 121 (talk) 06:00, 6 September 2013 (UTC)
also, the community has NEVER agreed on a policy of "quantity-limits per topic, subject, or category". nor should we, for that matter. commons is a media file catalogue, not a picture-album of one-of-everything. Lx 121 (talk) 06:04, 6 September 2013 (UTC)
No, Lx 121. Commons is NOT a media file catalogue but a free media repository with a specific SCOPE. There IS a DIFFERENCE. FLICKR can be a media file catalogue. --High Contrast (talk) 13:52, 6 September 2013 (UTC)
you are engaging in spurious wordplay. free media repository = catalogue of free media files. call it a "database", call it a "collection", net result is we provide a "catalogue" of free media material INTENDED for people to (re-)use elsewhere. Lx 121 (talk) 21:01, 6 September 2013 (UTC)
commons is also an "archive", de facto, & by default; as a result of being these other things :p Lx 121 (talk) 21:03, 6 September 2013 (UTC)
NO spurious wordplay. You still did not UNDERSTAND that the specific SCOPE is the difference. You know, the project scope. The project scope! --High Contrast (talk) 21:02, 7 September 2013 (UTC)
Whatever -- speedy deletions or admin-initiated out of process deletions can be semi-controversial in some contexts, but Commons:Nudity#New_uploads IS an official policy, and there is widespread sentiment that Commons does not need a large number of drunken cellphone-camera genital selfies. Furthermore, you are highly aware that this has been "discussed" a number of times, since you have participated in many of those discussions... AnonMoos (talk) 11:29, 6 September 2013 (UTC)
yes, & that is EXACTLY why i object to anyone claiming "nopenis" is a commons policy (or referring to it as such). policy should not happen by "creep". Lx 121 (talk) 21:01, 6 September 2013 (UTC)
I've placed a header on this section, although I don't know if you were trying to add a new section or respond to an older thread. Magog the Ogre (talk) (contribs) 03:52, 5 September 2013 (UTC)
Also, the coffee cups category doesn't have people signing into single upload accounts uploading copyright-questionable images, legally questionable images (proof that subject consented and is of age?), and images for the sole purpose of prurient voyeurism (or even vandalism) which makes a mockery of the purpose of Commons and ruins it for children and those with religious and/or moral convictions (and yes, the amount does make a difference; school libraries and foreign governments and such are more likely to block a website with 1/1000 files being dirty, compared to a website with 1/5000). Magog the Ogre (talk) (contribs) 03:56, 5 September 2013 (UTC)
  • i) "the coffee cups category doesn't have people signing into single upload accounts uploading copyright-questionable images" -- actually, we have LOTS of "hit & run" uploaders, on many topics. the penis-pics just get noticed more, because they have a sub-community of avid followers... Lx 121 (talk) 06:42, 6 September 2013 (UTC)
  • ii)"and images for the sole purpose of prurient voyeurism (or even vandalism)" -- you are making undue & unproven assumptions about the intention of the uploaders. aside from violating agf, you have a lack of evidence to support your conclusion.
as re: vandalism; ALL the major wikiprojects now have (adequate) tools in place to prevent mis-use of images, so that arguement is, respectfully, superceded ;p
  • iii)"and ruins it for children and those with religious and/or moral convictions" -- commons is NOT censored. (among other things) that means it is not our job to cater to the restrictions of people who hold such views. if you want, create some tools to "protect" these delicate souls from "accidentally" viewing the more unholy items in our collection, OR make some "child-safe" categories and/or pages for them, OR create a fork of commons: "child-friendly & christian (or whatever) commons" Lx 121 (talk) 06:42, 6 September 2013 (UTC)
Thanks for the headers. I agree its not very serious, and this is a new section, not a thread/response. if other editors are used to this number of similar images in a category, i wont worry. i did find a few low quality and bring them up for deletion/discussion. ps thanks to all the people keeping our total image load of questionable images lower than it would be.Mercurywoodrose (talk) 04:37, 5 September 2013 (UTC)
The numbers being thrown around here appear divorced from any statistical facts, it would be refreshing to be provided with some. Commons has over 18 million photographs. I would be interested to see the unbiased statistical analysis to demonstrate that more than 18,000 of these are "dirty". It would be particularly interesting to compare the same statistical analysis to Flickr, Google Image search or Facebook photographs. In practice the community sees a lot of oh-my-god style soap-boxing around a handful of images which, when fact-checked, turn out to be measured a magnitude less than the 0.1% quoted here. If anyone spots a problematic photograph, raise a deletion request, the "Nominate for deletion" link is right on the page you are looking at. If more people were genuinely concerned and would do that before trolling Jimmy Wales, members of parliament or posting on off-wiki attack sites under the guise of making their names as committed "citizen journalists", we unpaid volunteers might actually get around to manage, annotate or delete these images rather than wasting so much time on psycho drama. -- (talk) 10:51, 5 September 2013 (UTC)
But unfortunately all I see is heavy protest from many so called "we unpaid volunteers", including you; neglecting COM:PORN and COM:NOT. JKadavoor Jee 12:53, 5 September 2013 (UTC)
What you can read here is me asking for facts rather than polemic and just getting unhelpful soapboxing in reply. Jkadavoor, I see that you are fond of quoting wikilawyerish shortcuts like they mean more than the document you are linking to, in this case you have given two shortcuts pointing to the same guideline. Were you to compare my deletion requests raised over the last few months to this guideline, you will see I am hardly "neglecting" it. Reducing the number of poor quality images featuring nudity, and as a consequence accepting better ones to replace them, is at the heart of the intent of this interpretation of scope. -- (talk) 18:15, 5 September 2013 (UTC)
Formally, there isn't really a difference between images of penises and those of coffee cups - both cases require "educational value", and there isn't a restriction on the number of them. The difference is more one of practice - images of penises are more intensely patrolled, and some admins practically apply higher standards to them. However, I've also seen DRs against e.g. low-quality images of cats. If you want to file DRs against low-quality (e.g. unfocused camera-phone) photos of cups of coffee, you can always do so if you can show that there's another file that provides the same educational value with better technical quality (please also check the licenses, if the low-quality file has a less restrictive license, reusers may want to use that one, so it should be kept). darkweasel94 05:38, 5 September 2013 (UTC)
But a school library will not block a website because it hosts millions of cat photos or even above 50% percentage of its contents are low quality photos of cats. JKadavoor Jee 07:10, 5 September 2013 (UTC)
I don't think we should really take such things into consideration. zhwiki didn't self-censor when it was blocked in China either, and does anyone think it should have? Also, we aren't even close to having 50% images that are in some way "sexually explicit" (a culturally relative term btw, see e.g. ace:Wikipedia:Policy on awrah images in Acehnese Wikipedia); it's really probably less than 1%, and even if we didn't apply higher standards to them I doubt that more than maybe 3% were (just look at the live feed of new files and wait until you see something that's considered sexually explicit in your culture). darkweasel94 07:43, 5 September 2013 (UTC)
I watched the live feed for a while. About 400 inoffensive pictures went by before I saw a penis shot (DR here). Based on that, I would guess the rate is less than 1%. --Avenue (talk) 20:15, 5 September 2013 (UTC)
Less than 1% or even less than 0.4% is a much higher figure considering the other eight million seven hundred thousand species on Earth and 5,490 of them have penis. :) JKadavoor Jee 03:34, 6 September 2013 (UTC)
  • again, there is no quanitity-limit rule on commons. & if we are looking for categories with "disproportionate representation"' then there are other categories that are FAR more "over-populated". Lx 121 (talk) 06:31, 6 September 2013 (UTC)
See; the question is why we have no policies like COM:CATS or COM:CUPS whereas we have COM:PENIS / COM:VAGINAS. I think the answer of Magog the Ogre has a perfect sense. JKadavoor Jee 12:41, 5 September 2013 (UTC)
  • comment: again, these are not "policy" they are guidelines. this is a VERY important difference. (& for the record, "nopenis" is also NOT A GUIDELINE. it is, in fact, NEITHER a policy, NOR a guideline...) Lx 121 (talk) 06:10, 6 September 2013 (UTC)
However, Commons:Nudity#New_uploads is in fact a guideline. AnonMoos (talk) 11:37, 6 September 2013 (UTC)
i did say that, so we are in agreement. i was just belabouring the point that nopenis is NOT anything "official" (or any kind of community consensus) @ commons. it has no more weight or bearing, as policy, than an essay-page expressing the opinion of an individual or group of users. Lx 121 (talk) 21:23, 6 September 2013 (UTC)
Whatever -- Template:Nopenis is not "officially" endorsed (whatever that might mean), but it basically just restates what's at Commons:Nudity#New_uploads, which IS in fact official policy (and in fact COM:PENIS redirects there). Frankly, just about everything you have opposed concerning the handling of drunken cellphone-camera genital selfies is now accepted policy -- except for expedited deletion -- and you should face up to the fact... AnonMoos (talk) 04:21, 7 September 2013 (UTC)
I think it's reasonable to have a policy for maintaining images of commons subjects by quality and value merits. Search for useful images should not resemble search for needle in hay heap. --EugeneZelenko (talk) 14:48, 5 September 2013 (UTC)
the problem is not that we have "too many files", the problem is that mediawiki is CRAP at search functions, or even just basic sorting & organization of large databases of material. the tools are inadequate to the job, & nobody seems to be doing much to improve them. Lx 121 (talk) 06:10, 6 September 2013 (UTC)
Good idea. Even though we have badges like FP, QI and VI; it is difficult to find a “useful” image from tons of craps in a category. JKadavoor Jee 17:58, 5 September 2013 (UTC)
A lot of that problem could be solved by technical means - such as sorting categories by size in pixels, or even just displaying the size in pixels in category views; and for reusers, perhaps adding file licenses in category views - including filter systems based on all that. The category feature of MediaWiki really isn't very good or user-friendly for media files. But that isn't something that we can solve through any amount of discussion or policy-making here. Files that are totally unusable are routinely deleted, and that's good. darkweasel94 21:56, 5 September 2013 (UTC)
  • we could also consider many other useful criteria for ordering; number of views, useage on other projects, "popular vote", etc. Lx 121 (talk) 06:31, 6 September 2013 (UTC)
There are tons and tons of clever things we can and should do to improve search; however it remains the case that we will get porn results on unexpected queries, and that's just the nature of search - it can't reliably detect whether or not a user is actually trying to find a pornographic item. At least if search were improved, we'd be more likely to get higher-quality porn on unexpected queries. :-) Dcoetzee (talk) 23:21, 5 September 2013 (UTC)
I had tried to create sub heads like “Promoted files” under each gallery and added icons like Featured pictures, Quality images and Valued images under each picture, after a discussion with Archaeodontosaurus who is the top one quality contributor of biological pictures in Commons. I think it is useful for people visiting the categories; but the maintenance is time consuming. I think it will be nice if we have a provision to display such labels and other parameters like size, resolution, etc. in category itself (as darkweasel94 suggested). I see no solution for the unexpected porn results in search; but I think we can live with it as Commons in not censored.  :)
  • for anything like that to work, we need the tools to be automated. it should (& needs to) be at least as easy to use as hot-cat... Lx 121 (talk) 06:31, 6 September 2013 (UTC)
I think it is the time to stop the meaningless quarrel between the groups arguing for tight censorship and no censorship. In my opinion, both these groups are very small in the community. I think majority of this community have the same opinion of Kat Walsh. There is no problem of having educationally-relevant images of the full range of topics covered (including sexual contents), where subjects have given their informed consent, have enough quality and reliability, and the motif of creation and hosting is not questionable (See the resent DR). But there is no need to host poor amateur quality works or created and published solely for the purpose of trolling Commons or anybody else. (See the current DR. I don’t think an armature porn has any role in describing a scientific topic like Frenulum breve.) JKadavoor Jee 03:21, 6 September 2013 (UTC)
  • technically, EVERYTHING on commons is "amateur". & again, if we are complaining about the quality, there are other categories with equally poor-quality images. Lx 121 (talk) 06:31, 6 September 2013 (UTC)
That's not really true, but regardless, we can still distinguish between good quality and poor quality, and we are perfectly free to decide that we don't need multiple redundant low-quality images. AnonMoos (talk) 11:37, 6 September 2013 (UTC)
  • so why aren't there hardcore groups of diehard deletionists stalking travel pics, or cute animal photos, or etc...? Lx 121 (talk) 21:23, 6 September 2013 (UTC)
I really couldn't say much about those areas, but there are fair number of deletion nominations of perceived low-quality and/or redundant flag and coat-of-arms images (to mention an area which I more closely follow). But we don't get a constant stream of people uploading their own crude crayon drawings of national flags, for some reason... AnonMoos (talk) 04:21, 7 September 2013 (UTC)
Symbol support vote.svg Support User feedback to provide quality ratings in search prioritization. Symbol oppose vote.svg Oppose deletions based on arbitrary or undefined quality measures or rejecting images with reasonable educational rationales and no identifiable people issues, because "the doors are locked as Commons is full". -- (talk) 21:43, 5 September 2013 (UTC)
  • agree with fae. Lx 121 (talk) 06:31, 6 September 2013 (UTC)
This is a belated response to the reaction to my statement above: I never intended my figures to be literal. I was using the percentages as examples. Magog the Ogre (talk) (contribs) 01:02, 6 September 2013 (UTC)
  • well, with all due respect, i think that if we are going to wave number around, then we need some REAL numbers. my estimate for penis-centric images would be something under 10k (at most) files. taken against 18 million files in total, that comes out to 0.000555556.... that's not REALLY a lot. Lx 121 (talk) 06:31, 6 September 2013 (UTC)
Here you go -- there's some real numbers. And not all of these are of genitalia/nakedness/etc. Around 10,000 is a good guess though :) — Preceding unsigned comment added by russavia (talk • contribs)
Too much nude pictures are not only the problem (even though they have a highlighting nature); every over populated, unsorted category is a pain to potential end users. We need to slim them. How is the only question? If there is no consensus for deleting the craps, we can think about other solutions. What about creating a sub-category like "Category:Unused or low quality works" (or a more friendly name to avoid the discriminating feel) under ever category and move all those craps there? So people who really need the craps can go directly to those "craps folder" without disturbing others. JKadavoor Jee 08:36, 6 September 2013 (UTC)
There being too many human sexuality images is your opinion, and obviously the opinion of others, but do you really think your opinion is the overall community opinion? Above you have raw figures of how many sexuality images there are, and you truly believe that 0.05% of images on Commons being related to human sexuality is too many? As to your other proposal, it is totally unworkable in any form. An image which is unused today can be put into use tomorrow. So it would be a waste of human resources to even begin thinking about such a category. As to low quality images, again, who decides what is low quality? If an image is so low quality that it is basically unusable one can use COM:DR and quote COM:SCOPE in that it must be realistically be able to be used for an "educational" purpose. russavia (talk) 08:47, 6 September 2013 (UTC)
So you agree with deleting poor quality works (my preferred choice)? Otherwise, could you suggest an idea "for maintaining images of commons subjects by quality and value merits. Search for useful images should not resemble search for needle in hay heap. (EugeneZelenko)"? We need solution; not hard argument to defend any attempt to progress. JKadavoor Jee 09:01, 6 September 2013 (UTC)
I think (almost) everybody agrees with deleting unused poor quality works that can be replaced without loss of educational value, or loss of liberal licensing (some reusers may desperately need cc0 rather than cc-by and don't care so much about quality). And we currently do that, but whether it's "poor quality enough" and "can be replaced well enough" should be, and is, decided on a case-by-case basis through a DR because I don't trust any single person to make a judgement about that without listening to counter-opinions. darkweasel94 09:11, 6 September 2013 (UTC)
What I could imagine is some kind of a rating system. I wouldn't make it by file, but by file and category: to take an example from my uploads, File:Rampini Seitenansicht.JPG would have a higher rating in Category:Wiener Linien Rampini buses than in Category:Time 11:47 because it illustrates that better. Then you could sort according to that rating in a category. darkweasel94 08:55, 6 September 2013 (UTC)
The problem is, if you don't know those numbers, a random library or school won't know those numbers. It's going to be perception, not reality, and just one "Virgin Killer" will sink us. I don't think it would matter one bit to perception if we deleted all nudity except what's in use.--Prosfilaes (talk) 03:51, 6 September 2013 (UTC)
  • "just one "Virgin Killer" will sink us" -- o__0 please explain this comment? because that doesn't really make any sense, & i don't understand what you meant to say. Lx 121 (talk) 06:31, 6 September 2013 (UTC)
& yes i know what "virgin killer" refers to; it's the "just one will sink us" part that i fail to understand? sink us how? if we find illegal material, we delete it. it happens & we deal with it. end. of. story. Lx 121 (talk) 06:57, 6 September 2013 (UTC)
The original "Virgin Killer" case seemed to discredit the UK "Internet Watch Foundation" far more than it ended up discrediting Wikipedia... AnonMoos (talk) 11:29, 6 September 2013 (UTC)
Yep. Read this old story from 2008 at simple:Virgin Killer, it didn't "sink" Google or Flickr either. This is a good lesson as to why Commons should remain non-partisan and avoid making our own knee-jerk censorship based on knee-jerk reactions by regional bodies attempting to control the inter-web. This is not a argument to allow dumping on Commons of masses of blatantly poor quality or non-educational images regardless of whether they contain images of the human body. In my view "poor quality" includes pointless "web quality" grainy scans of PD material, I consider these a waste of our volunteer capacity. -- (talk) 12:24, 6 September 2013 (UTC)
The "Internet Watch Foundation" was huge and noticeable. I suspect we got added to several school blocklists over that and never got taken off. And the people who really care about such stuff are going to go to w:Human penis or Category:Human penis, be shocked, shocked I tell you, to find human nudity and add us to the blacklist. The point was, percentages don't matter; as long as we have the in-use images, that will be justification enough for us to be blocked in the eyes of people who would block us now, and hence arguing about people blocking us is irrelevant.--Prosfilaes (talk) 21:59, 8 September 2013 (UTC)
I believe the reason we have COM:PENIS and not COM:COFFEECUPS is threefold. Firstly, there is an aspect of exhibitionism associated with penises that is not shared by, for example, coffee cups. It is unlikely that someone will think it is exciting or funny or in some other way gratifying to replace the main image in Wikipedia's coffee cup article with a picture of their own coffee cup. There is nothing "naughty" about coffee cups and very very few people would get a thrill thinking that someone might be looking at their coffee cup on the internet. There are contributors here who have done nothing but upload images of themselves naked, for years. It is hard to draw any conclusions other than they are exhibitionists. Commons should host images of penises and other aspects of anatomy and sexuality, but there are better places for exhibitionists to share their images. Secondly, there are issues with penis images that are not shared by images of coffee cups, namely issues of consent and legality. Setting aside the "self shots" Unless we receive a complaint, it is difficult to know if the subject has consented to having an image of their penis uploaded here. It is difficult to know from a close-up image of a penis if the subject is 14 or 24. I report explicit images of underage subjects to the WMF probably every other month. If Commons wants to welcome any and all images of penises, they are going to end up inadvertently hosting illegal images. Thirdly, most of these are just not good quality images simply because they have been taken in less than ideal circumstances. Someone taking an image of their own penis while masturbating is not likely to take as good an image as someone taking an image of an outdoor monument on a sunny day. Delicious carbuncle (talk) 13:19, 6 September 2013 (UTC)
Hm well I don't know how you define "getting a thrill" and "gratifying", but I am certainly glad and proud when photos I've taken are used on Wikimedia projects or elsewhere, because that shows I've taken good quality and illustrative photos. Do I have a mental illness? ;) As for single-purpose accounts - that may be true, but it doesn't necessarily mean that they're exhibitionists: if I wanted to upload photos of human sexual organs, I'd probably use a sockpuppet for that too (I don't plan to do so though), because otherwise those images may be associated with me IRL if somebody finds out the connection between my real-life and normal Commons identity. Also, many people mostly stay in one topic area anyway: my uploads are mostly about trams - and I don't find it a priori less legitimate to upload mostly photos of penises than to upload mostly photos of trams. darkweasel94 14:08, 6 September 2013 (UTC)
And if someone decided to masturbat outdoors on a sunny day, how would we look at the issue then? That gives me an idea....I just hope it doesn't hurt. russavia (talk) 13:34, 6 September 2013 (UTC)
I hope it does. Delicious carbuncle (talk) 14:18, 6 September 2013 (UTC)

Maintaining images of commons subjects by quality and value merits

What about restarting this discussion focusing only on EugeneZelenko’s concern: "I think it's reasonable to have a policy for maintaining images of commons subjects by quality and value merits. Search for useful images should not resemble search for needle in hay heap"? JKadavoor Jee 12:49, 6 September 2013 (UTC)

We already have that policy, it is called COM:SCOPE. What is "useful" can differ depending on your needs. For example, if you want to illustrate a building for a Wikipedia article, you may want a high-quality photo of the entire building. But that doesn't mean that lower-quality files of certain details of the building aren't useful for those who want to know about those details. And it does not mean that historical photos of the same building aren't interesting. (See Category:Exterior of Vienna State Opera House for the particular category my words are inspired by. When I needed a photo of it for eo:Viena Ringstrato, I picked one that hadn't been in use before, and I was glad that there were so many to choose from, and nobody had chosen to delete a lot of them on the basis that they aren't in use.) I don't think we need to change our policies. Technical advances in the search and category system, such as those I mentioned above, would be great though. darkweasel94 13:02, 6 September 2013 (UTC)
I understand your concept; we need different quality images for different purposes. So it is not good to delete small/unimportant files, keeping only the best ones. But I think it is not the concern raised by Eugene Zelenko. I think he suggested that we need an order of quality and value merit; may be a threshold for the bare minimum quality and value is also good. Now all the files are scattered without any order; so finding one in a 10+ page category is really like searching for needle in hay heap. Further, no information other than file size in the preview; making the task more difficult. JKadavoor Jee 13:42, 6 September 2013 (UTC)
  • what we need is better search/sort functionality in mediawiki, or add-ons to it, combined with some kind of file-ranking system, to allow ordering (& re-ordering) of topically-relevvant files by characterists like quality, useage, number of page-views-over-time, "awards", popular vote, etc. anything that can generate a greater-lesser "desirability"-rating for the files.
what we don't need is to thin the ranks, to make it easier for people to find the "good" files. that's futile (& unwise, & undesirable); we have 18 MILLION media files, & growing daily. the goal of the project is to collect LOTS of useful free media files, to create a large & open catalogue of free material for re-use; not just a handful of "bests" in a gallery-collection. NOR a "picture-wikipedia".
Lx 121 (talk) 21:01, 6 September 2013 (UTC)
In a 10+ page category, it can be useful to create subcategories, which should make files easier to find as well. I'm against solving this on a policy level by excluding images that are low-quality but still useful just because somebody might find them in (correctly added) categories where they were expecting to find other things. I'm totally in favor of working together with the developers to develop more sophisticated category features - such as flat views of all subcategories (but perhaps excluding some, just like OgreBot does), CatScan built right into MediaWiki, rating systems, sorting mechanisms, license filters, RSS feeds (or watchlists) for changes in the images in a category, etc. The problem is just that somebody has to actually write that code, and there aren't a lot of developers. Anyone here want to open relevant RFCs and/or Bugzilla reports? :) darkweasel94 13:58, 6 September 2013 (UTC)
Yes; that is what we want. Now we are on the same track. JKadavoor Jee 14:19, 6 September 2013 (UTC)
I think it is pointless on the VP or AN or BN or Jimmy Wales' en.wp talk page. If you have a proposal and want to do more than soapbox this topic, then create a definitive RFC for change in policy. -- (talk) 13:08, 6 September 2013 (UTC)
From a technical standpoint, in the question of "Low-quality pictures of the windows on building 111 in New York" vs. "High-quality pictures of building 111...", the difference shouldn't be whether the quality is high, but what the picture's subject is. So one of the other features I'd suggest to add in the near future is better ways of finding categories that are even-more-specific, as opposed to limiting your searches on "low quality". That also means that you can search for "Pictures of the windows..." and then sort by quality - maybe you can find a high-quality image from a window-washer or something after all :) --MarkTraceur (talk) 19:15, 8 September 2013 (UTC)
We had a Multimedia Roundtable at Wikimania this year. The discussion revolved (amongst a few other points) around improving the discoverability of images. It seems like the foundation has finally allocated some manpower to improve the media experience on the Wikimedia projects. One idea that was floated was prioritizing search results based on quality metrics. Basically images with higher scores would be floated to the top of the search results (the search backend will be replaced anyhow. The current one I would rank between abysmal and completely useless). These could be existing ones (such as FP, QI, VIC), and/or completely new metrics, fe example based on youtube-like thumbs up/down. Such metrics have the potential to generate a much higher rating throughput at the expense of being less rigorous. Before this idea is immediately dismissed here let me say that the development overhead for such a feature is likely to be small, and that it seems to be worth a shot to try this (after all similar schemes seem to be working great for a whole bunch of top websites). --Dschwen (talk) 15:48, 6 September 2013 (UTC)
Good news; thanks for the info. JKadavoor Jee 16:01, 6 September 2013 (UTC)
Sounds like a good idea. Some factors that can be used to calculate such rating could include
  • FP, QI, VIC and assessment on local wikipedias
  • large/small size
  • file type (like SVG for graphics)
  • number of uses on local wikis
  • some thumbs up/down system
  • {{Watermark}} or other problem templates
  • License:
  • Licenses like {{GFDL or cc-by-nc-3.0}} could decrease the rating and {{CC-zero}} or PD could increase
  • Lack of license or use of decommissioned Licenses like {{PD}} or lack of US related license could decrease the rating
  • files without one of Infobox templates like {{Information}} or {{Artwork}} would be rated lower. Files using various internationalization templates like {{size}} or {{technique}} would rate higher.
  • number of categories (quality of categories ?)
  • number of edits and users editing the metadata
In other words we could use the search engine to encurage well written file pages. However we would have to be careful not to encourage some pointless actions that increase the ratings, like unnecessary edits. --Jarekt (talk) 16:42, 6 September 2013 (UTC)
  • i support the ideas, & am glad that other people are interested in this, BUT i am a little concerned at the complexity that is creeping in.
what you're describing involves a LOT of coding, & it would involve creating algorithms to do the calculations & "decision-making".
2 basic problems with that:
i)it means a lot more work just to get it up & running.
ii)it means developing "specialized expertise" in creating "the best" algorithms.
now, i do think that an open-source search/sort algorithm project has possibilities; but i'm not sure if wikimedia is the place for it (& if it is, it needs to be more than just commons doing it). i'm not against that, but it goes off into a whole project of its own; possibly something that should be set up through wikidata (and/or toolserver). AND we've got a LONG way to go, to catch up to the likes of google, amazon, ebay, & co... (or even bing)
what we do have (lots of) is "human intelligence". we have LOTS of people going through our material, every day. so, what i would suggest, is to use our strengths. design ways of ordering/ranking our file that use "people-power" (or maybe "human brain-power") to do the decision-making.
keep the automated parts of the plan SIMPLE. if you're going to use algorithm-based organization, stick to basic concepts & do them well. BUT it's the part where we have people looking at our stuff, & deciding how to quality-rank it, that gets us the most useful results.
in designing a file-ranking scheme, i'd suggest 2 things.
define (lots of) different, separate & specific criteria for file-ranking, which can be applied by people looking at a file & making decisions about it AND/OR by extracting useage-data (page views, file use (on & off wm projects), traffic on pages using our files, user interactions with in-use files, etc., etc., etc.).
allow users to see search {results, cats, whatever} display in order, by ANY of the ranking-criteria. allow them to re-order the results to a DIFFERENT ranking-criteria, & allow them to COMBINE (or cross-index) different ranking criteria; & allow then to use these powers EASILY (i.e.: without having expert-level abilities in boolean; or needing to know anything about boolean, for that matter)
that way, we get something that's useful, relatively easy to set up, & easy to expand upon.
i could go on ad-nauseam, as this is a problem i've spent much time considering (for all the wmf wiki-projects), but if we're getting into that level of discussion, it's probably time to set up a project-page. :)
Lx 121 (talk) 22:03, 6 September 2013 (UTC)
  • tl;dr = i think it's better to create separate ranking-criteria (& allow users to cross-index results), rather than try to come up with a (complex, compound, "algorithmic") "magic-number" rating-system. Lx 121 (talk) 22:24, 6 September 2013 (UTC)
That's nice there was a "roundtable", but seriously getting the WMF to do anything about it is like pulling teeth. And I don't hold my breathe on anything soon coming from them. Simply put, they simply does not care about projects other than English Wikipedia. It's more important for them to spend tens of thousands of dollars promoting editing on Wikipedia in India, for which the return was nothing but copyvios. And let's not get started on the clusterfuck that is Visual Editor. Why am I so disillusioned?
Commons:Requests_for_comment/improving_search#A_little_bit_of_intelligence saw support amongst many in the community and nothing was done. But then we hear in the media about searches for nefarious terms returning results that might be unexpected. The Commons community constantly cops flak over such things. And several admins (myself included) sent a pretty loud message to the WMF at Commons:Deletion_requests/File:Masturbating_with_a_toothbrush.jpg. And still nothing is done. Why? Can we really believe that no-one on the WMF board is aware of these issues? Surely Jimmy knows, so perhaps people can ask him why nothing is being done by the WMF, and why he isn't pushing for these things to occur.
The WMF rakes in tens of millions of dollars in donations, it's about bloody time they pull their fingers out of their proverbial arses and listen to what this community has been saying for years, and start spending some funds on fixing one of the most oft-talked about problems on our projects. russavia (talk) 16:20, 6 September 2013 (UTC)
Spending tens of thousands of dollars promoting editing on Wikipedia in India? Smells a good news for me; at least. JKadavoor Jee 16:30, 6 September 2013 (UTC)
Note, the WMF is currently hiring for multimedia positions (And has been for something like 5 months). Senior Software Engineer | Multimedia and Software Engineer | Multimedia Systems, which involve things related to commons. So I believe its safe to say that the foundation intends to invest more resources in multimedia and commons related areas once they hire more people to full those positions. Bawolff (talk) 01:27, 10 September 2013 (UTC)
Actually the visual editor is coming along nicely. --Dschwen (talk) 20:31, 6 September 2013 (UTC)
Dunno what planet you are on, but VE is a clusterfuck of major proportions. And I am sure that German and Dutch Wikipedia communities would totally agree with me there. But even then, here we are on Commons in a state of discussing the same thing that has been discussed for years, and will likely still be getting discussed for years to come. It's clear that the WMF simply doesn't give a rats about Commons. russavia (talk) 04:50, 7 September 2013 (UTC)
Well I hope the cynics are wrong and some cash can be thrown at Commons to improve search and move over to some kind of keyword system rather than the awful category hierarchies we have which make it harder to find images the more precisely they are categorised. Even some way to increase the thumbnail size on search results would be helpful. Colin (talk) 17:13, 6 September 2013 (UTC)
  • "category hierarchies we have which make it harder to find images the more precisely they are categorised" -- how do you mean this/pls explain? o__0 Lx 121 (talk) 21:01, 6 September 2013 (UTC)
    • I think he means that if you have a very general category and just want to see all the images that are there and in all subcategories, that isn't really possible - you may need to dig your way through all kinds of detailed subcategories until you find an actual image rather than just seemingly infinite hierarchies of subcategories. I think this can be solved through a "flat view" as I proposed above. I'm not convinced that a "keyword" system would be much better - "keywords" for searching are supposed to be in the file description, and if "keywords" aren't standardized, they aren't really useful to see "all images that show X", and if they are standardized, that's really what we have now with categories, or am I misunderstanding that? What is true is that our category system is sometimes insufficient to portray the real world's complexity (e.g. buildings being categorized into the streets that (some of) their entrance(s) is/are on, sometimes leads to less than perfect situations) - I could go on about that but that needs more space and isn't really in the scope of this discussion. darkweasel94 21:34, 6 September 2013 (UTC)
i think the "keywords vs categories" thing is slightly irrelevant; in that they are effectively the same thing/achieve the same result.
being able to view the entire contents of a category "flat" is absolutely a tool that we need, & i cannot imagine why we do not have it already (& of longstanding)?
the issue of relationships between discrete pieces of data (& how to define them for organizing material in mediawiki databases) is being worked out on wikidata as-we-speak; at least, they're working ON it. how soon/how useful & when-does-it-get-to-commons are open questions. their focus seems to be more on inter-wiki, rather than in-wiki tools; but these are the people who are focussed on these issues. so, we should probably be talking with them about it.
Lx 121 (talk) 22:16, 6 September 2013 (UTC)
LX says that «being able to view the entire contents of a category "flat" is absolutely a tool that we need» and I whole heartedly concur. Why don’t we have it already must be because the WMF doesn’t get enough donations to pay staff that could develop that sort of relatively simple programming. Oh, wait! -- Tuválkin 23:36, 6 September 2013 (UTC)
For reference, Efficiently collapsing a (potentially large) graph on demand is a lot harder than you would think. At the very least we would need a better data structure to store the information (and I have no idea what sort of data structure would be appropriate. The current one we have does not lend itself well to such queries). I also think that total flatness is probably not actually all that desirable. The further you go down the category tree, the less related things become. Category:English phonology is just 4 steps away from Category:Australia. I imagine people don't expect to see file:Atlantic Canada IPA chart.PNG when looking up Australia. Bawolff (talk) 17:58, 8 September 2013 (UTC)
Two matters you address:
  1. I’m sure them number of items to display is not the problem — after all there’s this already. Recursiveness of all child elements from a given point of a tree is of course a problem, as there may even be nested loops and other topological paradoxes, but I’m sure that can be easily dealt with by a compettent programmer (which I’m not). Easily being a relative concept here, of course: Easily relative to the amount of money the WMF can set apart for tech staff hirings — they are using copious resources for things some deem unnecessary and even deterimental to the project.
  2. Unexpected results due gathered from remote categories (assuming correct categorization) is not a bug of flat view, it is a feature. People should get what they ask for. If you ask for "Australia" from among 17 million media files you’re bound to get details about “strine” phonology not just kangaroos, didgeridoos, and the Sydney Opera.
A simple suggestion that may be the solution for these two problems: A gadget that adds a checkbox to each sub category on any category page, right next to the "[+]" (or "[×]"). The user checks some of these boxes, at any depth (maybe with options to select parent-only or the whole subtree; reclick to toggle selection); then click "Go!" for a custom category-page-like view of all files in those custom select categories (200 hits per result page). Can it be done? -- Tuválkin 23:53, 8 September 2013 (UTC)
Doing category union of a small number of categories, sounds like something much more likely to be do-able. (I say that as an initial gut feeling.). Another possibility (Just to throw ideas out there), is perhaps flattening to a certain point (So instead of say flattening a category 2 levels deep, do w:Breadth first search (try every subcategory in order) until you find 200 files, stopping immediately once you have 200 files [or have checked X number of subcategories]). That might be less scary performance wise (maybe anyways, I haven't thought very hard about the performance implications of something like that, so I'm not really sure. Downside is results will be fairly limited, and less predicatble [since how far one goes down the category tree would vary with how many images are in the categories). To continue throwing ideas out there, maybe the problem could be slightly mitigated (although not fixed) by a category page design geared towards browsing subcats for images. Maybe something similar to Commons:Requests_for_comment/improving_search#A_little_bit_of_intelligence, but for categories. Perhaps instead of displaying subcategories at the top in their own section, instead each subcategory could be a collapsible section, that when you open shows all the images from that subcategory or something. (Sorry that's a little rough, this was mostly brainstorming off the top of my head). Bawolff (talk) 01:27, 10 September 2013 (UTC)
Just to go back a step. Earlier darkweasel mentioned RSS feeds (or watchlists) for changes in the images in a category. I would like to point out that we do have Special:Recentchangeslinked which can mostly do that (It doesn't include edits that remove a page from a category). For example, for Category:PD-old-80-1923 we have https://commons.wikimedia.org/w/index.php?title=Special:RecentChangesLinked/Category:PD-old-80-1923&feed=atom&namespace=6&target=Category%3APD-old-80-1923 Bawolff (talk) 16:59, 8 September 2013 (UTC)
See also Magnus’s catfood], which does precisely that. Jean-Fred (talk) 12:18, 9 September 2013 (UTC)

File:Martin Luther Kirche, Stretford (1).JPG main preview thumbnail not updated

Hello again,

I uploaded the above photo, but then noticed there was too much of the road in the lower part of the picture. So I then used Cropbot to crop it to a new picture with 2:1 ratio, such that it now just shows mainly the church, as intended. The full sized picture of the cropped version looks completely correct and so does the small thumbnail in "File History". However the default 800x399 thumbnail used for the main preview looks like the original non-cropped photo squeezed into the new aspect ratio, showing the whole road again. I've tried doing a "Purge" on the page, deleting the browser cache and trying alternative browsers, but I can't get that preview to update correctly. Does anyone else see this wrong preview, or is it just me? And if so, is there any way can I can force it to update? Thanks again for any possible reply, Rept0n1x (talk) 17:44, 8 September 2013 (UTC)

Yes, I can reproduce this. darkweasel94 18:26, 8 September 2013 (UTC)
Thank you for the confirmation. Rept0n1x (talk) 18:35, 8 September 2013 (UTC)

I also cannot force a purge. That particular thumbnail is served by cp1061. Looking at the graphs, cp1061 looks almost as if it stopped responding to purges (total purge packets dequeued seems to have flattened where it should be slowly increasing around thursday, if I'm interpreting the graphs right). Bawolff (talk) 18:28, 8 September 2013 (UTC)

Filed this as RT 5735. Bawolff (talk) 18:50, 8 September 2013 (UTC)
Thank you again for shedding some light on the problem. I will simply leave the file as-is for now until technical issue is resolved - the image isn't currently in use, so it's nothing urgent. I'm just grateful that you could identify a reason for this behaviour. (EDIT: also thanks again for reporting the issue) Rept0n1x (talk) 18:53, 8 September 2013 (UTC)

Bawolff, could you have a look at this file? The thumbnail showing at 555px (the preview) contains a copyright violating element. Lemmens, Tom (talk) 11:53, 9 September 2013 (UTC)

Different cause then Rept0n1x's file (Although not exactly sure what that cause is). I can confirm that the 555px size of the coat of arms is not purging. This one is being served by cp1064, which generally is working (For example, the 119px size also gets served by cp1064, and gets purged fine). So I'm not sure what's happening. If I were to guess I'd say something is wrong in the swift layer, and the version in the swift layer is not getting deleted when ?action=purge is happening. To further substantiate that theory, I notice there's a 'Last-modified: Tue, 17 Jul 2012 19:23:50 GMT' header for that size, even on requests that are cache misses for varnish. Bawolff (talk) 17:47, 9 September 2013 (UTC)
I think this is bugzilla:47087 (Which could be summed up as, rarely a specific size of a specific file does not purge and we have no idea why). Bawolff (talk) 18:04, 9 September 2013 (UTC)
I uploaded the same file with a different width, so that the version displayed is different, which should make the copyvio less of an issue (If you directly ask for it in 555px it still displays the wrong version). Bawolff (talk) 18:17, 9 September 2013 (UTC)
Ah, excellent! Thank you. Lemmens, Tom (talk) 19:19, 9 September 2013 (UTC)

┌─────────────────────────────────┘
Also, just to say that the main preview thumbnail of the Martin Luther Kirche photo is now fixed, so thanks again for helping with this. Rept0n1x (talk) 20:21, 9 September 2013 (UTC)

Hmm, the graphs look happy for cp1061 too. Bawolff (talk) 01:01, 10 September 2013 (UTC)

Bulk uploads of large files

Hi all,

I've been working on a long-running mass upload project, using pywikipediabot, and I've got to a collection of files which are ~ 100-150MB. These fail if I try to upload them using the bot.

As far as I can tell, chunked uploading only works through the browser, and only through the Upload Wizard, which means they would need to be done by hand. This involves ~250 items and associated metadata, so I'm not entirely keen on this. As I have a relatively slow connection and each one will take around an hour, it'd be very inconvenient to have to keep coming back at odd intervals to restart the uploads - and, of course, I couldn't leave it running overnight!

I feel like I'm missing a trick somewhere here - is there any obvious way around this? Andrew Gray (talk) 21:47, 9 September 2013 (UTC)

Chunked upload is neither limited to browsers nor to Upload Wizard. I have a VB 6.0 class, for example. The procedure is explained at mw:API:Upload#Chunked uploading. -- Rillke(q?) 22:07, 9 September 2013 (UTC)
Some of the {{Commons upload tools}} support chunked upload. --McZusatz (talk) 22:09, 9 September 2013 (UTC)
Hi Andrew, I thought Pywikipediabot had been updated, however if not, then as Rillke points out, this is possible via the API with a bit of jiggery-pokery, hopefully someone can share an example to save time? The guidance at mw:API:Upload#Chunked_uploading is a bit daunting for a DIY job. -- (talk) 22:12, 9 September 2013 (UTC)
Also wiki-java supports chunked upload, as far as I know. --McZusatz (talk) 22:17, 9 September 2013 (UTC)
Pictogram voting info.svg Info UP! should be able to handle this. --Hedwig in Washington (mail?) 02:54, 10 September 2013 (UTC)
… and Vicuña as well. --El Grafo (talk) 08:19, 10 September 2013 (UTC)
As part of the LSH mass-upload, André is using this Python script with quite success apparently. Jean-Fred (talk) 09:16, 10 September 2013 (UTC)
Thanks J-F, the token handling there is exactly what I would need. The other tools recommended above make for easy solutions too. -- (talk) 09:45, 10 September 2013 (UTC)
  • Thanks all; I'll look into these. I have a pile of files and prepared upload templates, so hopefully Smallman's script is just what I need! Andrew Gray (talk) 18:58, 10 September 2013 (UTC)

September 10

Tram category tree naming RfC

There has been a RfC started at Category talk:Trams#Request for Comment regarding localisation of terms regarding the use of localised terms such as streetcar in tram related categories. The outcome of this discussion may affect the name of many categories in the Category:Trams tree, input is requested and appreciated in building consensus. Liamdavies (talk) 07:49, 10 September 2013 (UTC)

Rotatebot

Has the Rotatebot malfunctioned in the past 2 days? It has not rotated images since September 8, I believe. Regards, --Leoboudv (talk) 08:10, 10 September 2013 (UTC)

Abusive deletions

Please check this: Commons:Deletion requests/Files of User:Salamjawed and witness how Fastily deems as «blatantly out of scope» «unused and personal images» photos of professional musicians i.a. on stage in the South Asian Bands Festival. -- Tuválkin 10:28, 9 September 2013 (UTC)

I had look just checking the first one I viewed you could add possible copyright violations to the list the end conclusion would have been the same. Gnangarra 12:57, 9 September 2013 (UTC)
No, no, no: You can delete images for being merely personal selfies, or you can delete images for being copyright violations uploaded without permission from their creators — but you cannot claim both rationales, obviously. -- Tuválkin 13:31, 9 September 2013 (UTC)
Those are mostly small, low quality, unused, uncategorized images without descriptions and as such are not usable in the current form. Someone could have dedicated their time and provide descriptions and categories but that did not happen in 2 years since they were uploaded. So I agree that they are out of scope. --Jarekt (talk) 14:12, 9 September 2013 (UTC)
Why do you say uncategorized? I know I categorized these images as much as I could — the topics of music and Afghanistan being far away from my experience, I could not do better. Isn’t the proper policy for unidentified locations or people to advertize the affected media in "unknown so-and-so" categories? -- Tuválkin 15:10, 9 September 2013 (UTC)
OK, I take it back: only some images were uncategorized. 2 images were in Category:Musicians from Afghanistan and 3 in Category:Salam Jawed. But the rest had no categories or only categories like Category:Drummers or Category:Group portraits, which carry very little information.--Jarekt (talk) 12:43, 12 September 2013 (UTC)
Note also, in terms of scope and our goal of offering at least minimal coverage of any notable issue, that we have currently 85 media files under Category:Musicians from Afghanistan — and that this deletion wiped out in a single blow 9 more images: almost 10% of the previous total. This is an incredible misservice the us all. -- Tuválkin 15:10, 9 September 2013 (UTC)
A agree that "blatantly out of scope" is an unreasonable comment if the images show (locally) well known musicians. An administrator who does not know the field should not use such language ("probably out of scope" would have been much better). We, who cannot see the images any more, should be able to trust administrators' comments (it is possible that the images should be deleted as copyvios, but they were not, so this is not relevant for the criticism). --LPfi (talk) 08:43, 10 September 2013 (UTC)
I agree that "blatantly out of scope" is not really a helpful deletion reason when there were users with different good-faith opinions in the discussion - such cases require a more detailed reason as to why the educational value is not there. See Commons:Deletion requests/Files of User:Varkeyveliyath for another such case. darkweasel94 08:54, 10 September 2013 (UTC)

┌─────────────────────────────────┘
It seems reasonable to support the value that administrators are openly accountable for their deletions and with the exception of avoiding the Streisand effect out of respect for living people, or where public comment could make a legal issue more damaging, Deletion requests on multiple files or files with some likelihood of challenging deletion, should be closed with more than vague comments such as "out of scope" or "low quality". Complainants in this thread should consider:

  1. Raising suggestions for improvement at Commons talk:Deletion requests so that Administrators are required to close a DR with more than a blank statement or "close". If Administrators feel over-burdened with the task of processing DRs, then other options (such as refinements to the speedy process) and improvements to the DR workflow and associated reporting and prioritization should be considered.
  2. Take this case to Undeletion requests to have a meaningful review of the educational value of this set of images. If the value is more than zero, then out of respect for this under represented subject, all should be undeleted and those for which there is remaining doubt as to educational value can go through a second, potentially longer review (I see no harm in letting them run for 30 days to gain the benefit of all possible points of view), so long as there are no significant doubts as to copyright status.

I suggest this thread is considered closed and superseded if either of these actions are taken. -- (talk) 09:12, 10 September 2013 (UTC)

Please file Undeletion requests for files which you think should be restored. --Jarekt (talk) 12:46, 12 September 2013 (UTC)

Categorization by user Proshob

Hello. What do you think about the "work" of user:Proshob [3]. He is just changing files from date categories in Category:Media needing categories to uncategorized by name like Category:Uncategorized images from Ronald Saunders. (What is his relation, if any, to User:Matanya ? Traumrune (talk) 21:01, 10 September 2013 (UTC)

I see what you mean, but what matters is a way to add categories to all those files. Having a parallel tree to bag together items according to author/uploader along with upload date seems to be a good idea, as often media from the same author shares many common categories, especially those concerning location or subject, allowing for some Cat-a-lot magic. Move the items away from the date categories is very wrong, however. (Furthermore, Category:Uncategorized images from Ronald Saunders was an orphan, unreachable from anywhere else; I added two suitable parents.) Hopefully is user can be persuaded to add, not replace, these new categories, and it will all be fine. -- Tuválkin 01:58, 11 September 2013 (UTC)
Thanks, you understood the problem. I don't know how other people categorize, but I always start with some day, and only when I find more then one picture by the same uploader I look in the contribution list, if there are more images because people often take several pictures around the same theme. But it's always only in second place. Once I started to categorize Ron Saunders pictures I discovered that it is true in his case. Traumrune (talk) 20:35, 11 September 2013 (UTC)
Just an addition: With cat-a-lot the category media needing category disappears now automatically, the category created by Proshop has to be suppressed in a second step :( Traumrune (talk) 20:47, 11 September 2013 (UTC)
Not supressed, just move the already categorized item up the ladder to his own vanity category. Not terribly important, too. The date pages should remain as the primary split of uncat files; once they are clear from there, they are half way to be categorized in full. (In that regard it makes sense that the uncat cat disappeares as soon as at least one real category is added to the filepage; cat-a-lot doens’t refresh after that, though, so we can keep adding categories with it after the 1st, if we only dont click away from that page.) -- Tuválkin 08:00, 12 September 2013 (UTC)

September 11

Category:Free Software Foundation

Hello, I can’t access the Category:Free Software Foundation, maybe it is full of non-relevant files? Does someone could take a look in it? Thanks! genium ⟨✉⟩ 21:00, 11 September 2013 (UTC)

I don't see a problem. I can access it and the files in it appear relevant. darkweasel94 21:39, 11 September 2013 (UTC)
Thankyou for your reply! Indeed, I can access it with another browser, but I wonder why I have this problem only with this category… Hum I have probably something wrong somewhere… genium ⟨✉⟩ 22:53, 11 September 2013 (UTC)

September 12

New rollback testing page

Hi! I was wondering if I can create a Commons:Rollback/Testing ground for those who got rollback and wanted to try it out, so they can go to that page to do their testing of rollback. Jianhui67 Talk 06:42, 12 September 2013 (UTC)

Just use Commons:Sandbox. Multichill (talk) 16:49, 12 September 2013 (UTC)

Category "Cultural heritage monuments"

How to define a cultural heritage monument? Category: Cultural heritage monuments in Kiev contains thousands of files (statues, apartment buildings, religious buildings, military vehicles, parks etc.). Should all these objects be categorized as cultural heritage monuments? --Kulmalukko (talk) 13:06, 6 September 2013 (UTC)

In many countries there is a formal legal process to make a certain object a cultural heritage monument, which gives it extra protection from being demolished. As I understand it, that's the criterion by which objects should be put into those categories. I don't know how it is in Ukraine. darkweasel94 13:25, 6 September 2013 (UTC)
You might have a look at the Ukrainian WLM-site. There you will find some facts about cultural heritage monuments in Ukraine in general. --High Contrast (talk) 13:38, 6 September 2013 (UTC)
Пам'ятка культурної спадщини (cultural heritage monument) is the phrase used in the Ukrainian legislation for the protection of heritage for objects formally listed for state protection: "пам'ятка культурної спадщини (далі - пам'ятка) - об'єкт культурної спадщини, який занесено до Державного реєстру нерухомих пам'яток України" Man vyi (talk) 13:59, 6 September 2013 (UTC)
Cultural heritage monuments (in particular, in Ukraine) are defined by law (national importance) or by decrees of local authorities (local importance). On Commons, usually we do not make a difference between monuments of national and local importance. If a category contains thousands of images (does not matter what) it is not good, more specific categories should be created.--Ymblanter (talk) 14:42, 6 September 2013 (UTC)
In some cities there are categories for both Cultural heritage monuments of the city and Monuments and memorials of the city. Should the Cultural hertitage monuments.. categories be deleted and put all the monument files for the category Monuments and memorials of the city? (I agree there is no sense to keep separate categories for normal monuments and cultural monuments. At least I think the apartment buildings, vehicles and parks should not be in this kind of categories.) --Kulmalukko (talk) 15:08, 6 September 2013 (UTC)
No, cultural monuments are entities protected by law. These categories are used for proper functionality of Wiki Loves Monuments, for example. Monuments and memorials can be anything, and there is certainly an overlap, though the vast majotrity of cultural monuments are in fact buildings.--Ymblanter (talk) 15:19, 6 September 2013 (UTC)
Ok, then keep them. But this problem of huge amount of the files in cultural monument category should be resolved. Although we would make category Cultural heritage buildings in the city, the quantity of the files is still too big. --Kulmalukko (talk) 15:27, 6 September 2013 (UTC)
Monuments and memorials are commemorating items. Cultural heritage monuments are memorable items Smile. --Havang(nl) (talk) 16:12, 6 September 2013 (UTC)
hello there. WLM Ukraine team is speaking^^ as soon as we are a bit free and able to take a breath, we shall put the files into smaller categories. do not worry please at this stage at least. we shall sort things out --antanana 19:54, 6 September 2013 (UTC)
Good. --Kulmalukko (talk) 20:36, 6 September 2013 (UTC)
Good luck with that. In Category:Lisbon we’re still cleaning up the mess left by WLM 2011. They love to take pretty snaps (well, sort of), but cannot be bothered to sort them into categories. Seems it is too late for the Ukraine now, but could at least the London WMF photography training above dedicate 1% of its time to try hammering in that Commons is not a hosting cloud for their vanity albums, but a tool for sharing that needs some additional work to make their snaps actually useful? Please, WLM people and any other “campaigned” and “outreached” volonteers — add geolocation, proper timestamps, description, and minimally useful categorization. Thank you! -- Tuválkin 23:30, 6 September 2013 (UTC)
we are going to sort images using IDs and bots. and we haven't been able to do it with WLM Ukraine 2012 images yet, as there were some crucial convertation-related mistakes in the lists, so even the IDs for the whole regions were to be changed... it was like starting the work over((( a great deal of time and efforts are to be invested into creating a category tree, where the ID=the category. actually we were thinking of maintaining the lists on wikidata, but this idea (though a great one, we believe) will need even more time to implement... next year, perhaps --アンタナナ 08:33, 7 September 2013 (UTC)
Why aren’t you just populating (and expanding, and maybe improving) the current category tree? I count 90 subcats under Category:Cultural heritage monuments in Kiev, and over 3000 uncategorized images needing diffusion. Why wont WLM volonteers fire up their Cat-a-lots and have that work done? I tackled categorization of around 1000 images under Category:Lisbon (needs more work) in the past 6 months — it can be done. -- Tuválkin 09:11, 7 September 2013 (UTC)
My understanding is that this is what they are planning to do starting in November, as they are understaffed. But it is not a bad idea, I can have a look myself and see whether there are easy things to categorize (I last visited Kiev in 1993 and remember pretty much nothing).--Ymblanter (talk) 09:22, 7 September 2013 (UTC)
the current category tree is a mess(( and we are working on the lists at the moment. there is no single database in Ukraine, thus letters to a town/city, raion, oblast and state are to be written. and the lists are to be converted. often they look like this: http://wlm.org.ua/tryvaje-onovlennya-spyskiv-pamyatok (created long-long time ago, using a typewriter, then they were scanned). we have approximately 120-140K of the monuments in the country. and we have managed to get lists of 50-60K of them, 44-45K are converted already, but are to be checked, as the IDs are not unique. and the contest is running. people are uploading now, and they ask why they can't find this or that object now. they send us the proof (the monuments can have a plate stating that they are monuments) and we are to look into the list we got, to check, to look into the books or to write additional letters. we simply do not have enough manpower to do everything at the same time, thus we are doing it in the way we are. now - lists and current work. after the contest - cleaning up --アンタナナ 09:26, 7 September 2013 (UTC)

(break)

I just noticed this topic.
We use "Cultural heritage monuments" to make the distinction between, like Havang put it, "Monuments and memorials are commemorating items. Cultural heritage monuments are memorable items" (we should keep that around, it's an excellent description). If local well defined names are available, we use that. See for example Rijksmonumenten. The definition is different per country and it can be very blurry, but one thing all country have in common is that we're trying to build a tree for easy navigation and to not get overflowed categories. Structure of the tree:
<cultural heritage monuments in some country> -> <cultural heritage monuments in some subdivision> -> <cultural heritage monuments in another subdivision> -> <category for a single cultural heritage monument>
Category:Rijksmonumenten -> Category:Rijksmonumenten in North Holland -> Category:Rijksmonumenten in Haarlem -> Category:Grote Kerk, Haarlem
The tree should be expanded based on available content and content should be pushed down the tree so images can be found. To help with that, a bot runs several times a day, please see Commons:Monuments database/Categorization. The focus is to sort out the main categories like Category:Cultural heritage monuments in Ukraine. It already moved over 100.000 images to lower categories. It's possible for the bot to move images to lower categories than Category:Cultural heritage monuments in Kiev. I'm not sure if it makes sense to make raion subcategories and link the different lists to them. Would be a one time run to move everything in Category:Cultural heritage monuments in Kiev down and new images go to lower categories right away. Antanana, would this make sense? Multichill (talk) 11:23, 7 September 2013 (UTC)
thanks, i believe that will make sense (we have splitted the lists in WP by raions anyway). this idea will help as well: moving the category from the file to the category itself, thogh, without raions it will create a mess eventually. our ID 80-361-9014 (f.ex.) consists of Kyiv code (80), Kyiv raion (361) and a "serial" number (9014) --アンタナナ 13:29, 8 September 2013 (UTC)
As another WLM organiser in Ukraine, I tried to sort these photos in one single raion... it took a couple of hours to sort out everything even with Cat-a-lot, and yes, everything from appartment buildings (for example, one where a notable poet lived before his arrest) to statues and parks is recognized as a monument indeed. I do not think it will be possible to work with this category until the end of September as it does take a lot of time, but we will try to do our best afterwards — NickK (talk) 03:12, 12 September 2013 (UTC)
Would it be a good idea to send all photographs from the big category to the raion categories, and then to categorize them more precisely? I am asking this, because for Russia we adopted the opposite scheme: nobody follows Oblasts categories, and if a file is sent there it is lost for many months, so that we categorize everything properly directly in the root category. But if you follow the raions it is definitely easier to categorize everything into raions first.--Ymblanter (talk) 08:23, 12 September 2013 (UTC)
We did follow raions in our IDs from the very beginning, but for another reason: lists in two different raions may be in completely different formats, and monuments in two different raions may have completely identical "governmental" IDs. Thus we generated our own IDs strictly following raion subdivisions. It is especially helpful to find pictures of, say, "all pictured monuments in raion": a research for a raion code (say, 80-361) gives the result directly, thus such actions can be automated. Yesterday I tried to find necessary monuments in the city category, but it definitely took lot more time than for raion categories — NickK (talk) 10:02, 12 September 2013 (UTC)
I see. Then I will try to depopulate the category by re-categorizing all images into raions.--Ymblanter (talk) 15:39, 12 September 2013 (UTC)
I mean like this. I am under understanding you guys can take care of this file now: creating a special category, or otherwise categorizing it by streets, type etc. if needed--Ymblanter (talk) 15:43, 12 September 2013 (UTC)
I tested and the bot seems to run fine on Category:Cultural heritage monuments in Kiev, just some images ending up in Category:Baikove cemetery and that's not part of the category tree. To you want me to run on the whole category? Multichill (talk) 16:34, 12 September 2013 (UTC)
I am not in a position to answer this; let us wait for NickK or antanana. (I guess the answer is yes).--Ymblanter (talk) 16:52, 12 September 2013 (UTC)
I have looked at the bot edits, this looks fine for me. We would be happy if you could move all photos from Category:Cultural heritage monuments in Kiev to raion categories. Concerning categories of cemeteries, there is a problem with them, as we have several cemeteries where many (but not all) graves are monuments, but, on the other hand, separate graves hardly deserve individual categories. Thus these files may seem to be "lost" to the category tree, but it's OK for now, we will figure out a solution later — NickK (talk) 19:02, 12 September 2013 (UTC)
agreed! thank you very much, everyone! --アンタナナ 23:52, 12 September 2013 (UTC)

File:75%.svg modified without consensus

Hi everybody, I would like to point out that the file File:75%.svg, whose colour was recently changed from orange to blue, is massively used in a lot of projects (lot of Wikisources and others), as stated in the talk page, and it should absolutely not be changed before gaining a broad consensus from all the projects who use it. This as far as I can tell has not be done, the user who performed the change only asked in an en.wikibooks page (after already doing the same 1 year ago, and being rollbacked). I ask for this file to be restored as soon as possible (and maybe protect it to avoid further changes). Thanks Candalua (talk) 10:52, 11 September 2013 (UTC)

I'd say if the file was created for use in Wikibooks, then Wikibooks should be able to change it. Other projects should either accept whatever changes Wikibooks makes, or make their own copies of the file. It could have been given a better name, but it does state on the description that it's intended for use in Wikibooks. ghouston (talk) 23:28, 11 September 2013 (UTC)
I disagree. Commons isn't set up to handle projects calling dibs on certain images, but to encourage wide reuse of images across multiple WMF projects and beyond. This file is used tens of thousands of times, on many projects - not just on English Wikibooks, nor even on other language Wikibooks, or even the various Wikisources. For example, it has several hundred uses on French Wikipedia. I don't think we should allow two people on en Wikibooks to disrupt all those other users.
By uploading an image here, you accept our policies and guidelines (such as Commons:Overwriting existing files) will determine how we host it. A project can always upload a local version at the same filename if they want something different there from what we host. --Avenue (talk) 02:13, 12 September 2013 (UTC)
I've full upload-protected this file to avoid further overwrites without discussion. INeverCry 02:34, 12 September 2013 (UTC)
I see, so Wikibooks should consider making copies of the files onto their own servers and use a bot to update the pages that use them.ghouston (talk) 07:05, 12 September 2013 (UTC)
Not necessarily on their own servers. They can use another Commons filename. darkweasel94 08:17, 12 September 2013 (UTC)
But if they host it on their own servers, they can change it again any time they like, without needing to modify every page that uses it. ghouston (talk) 11:44, 12 September 2013 (UTC)
Commons:Overwriting existing files seems not to deal with "consensus" in such a case, so should I try to get a consensus or should I rather upload right now with another filename? Doing so, I have to do the same with all the percent images as it is used with an automatic process. ftiercel (talk) 12:49, 12 September 2013 (UTC)
I do not think you can get consensus of all the users as they are scattered all over. I would suggest uploading a new file under more meaningful name. --Jarekt (talk) 13:24, 12 September 2013 (UTC)
I have uploaded new versions with another name: 00 percents.svg 25 percent.svg 50 percents.svg 75 percent.svg 100 percents.svg ftiercel (talk) 13:01, 13 September 2013 (UTC)

Commons:Requests for comment/Technical needs

Dear community, I would like to invite you proofreading, discussing and copyediting Commons:Requests for comment/Technical needs (despite a lot seems to happen, recently). Thanks in advance. -- Rillke(q?) 20:10, 12 September 2013 (UTC)

Wie stellst du dir denn dabei das Zusammenspiel mit der deutschen Umfrage vor? Sind diese unabhängig und werden auch unabhängig bewertet und an die WMF weitergeleitet? Oder werden alle vorschläge multilingual aufbereitet und dann gemeinsam zu Abstimmung angeboten? Wenn ich Vorschläge in der deutschen Version eingebracht habe, die Commons betreffen, soll ich diese dann hier wiederholen oder nicht? Danke und Gruß, --Patrick87 (talk) 21:26, 12 September 2013 (UTC)
Es werden zwei getrennte Listen mit unterschiedlichen Prioritäten erstellt. Es ist also sinnvoll, die dort eingebrachten Vorschläge auch hier zu listen, auch da diese auf de.wp nicht übersetzt werden, während ich hier versuchen werde vor der Abstimmung/dem Ranking, möglichst viele Übersetzungen der Vorschläge einzuholen. Außerdem wird es so für uns und die WMF übersichtlicher. -- Rillke(q?) 22:18, 12 September 2013 (UTC)

Images still not in category

Hi. I re-added a category to Template:PD-South-Africa days ago (someone removed it a few years ago and no one ever added it back). Now I did, but all files who should belong in the category Category:PD South-Africa (see Special:WhatLinksHere/Template:PD-South-Africa) are still not added. At the moment 43 files are in it, while the template is included on at least hundreds of files. How can I add all those files to this category without having to purge each file manually? Trijnsteltalk 21:22, 12 September 2013 (UTC)

My bot's doing it (won't show up in the contribs page, though). --Zhuyifei1999 (talk) 13:34, 13 September 2013 (UTC)
Thank you very much for the help! Trijnsteltalk 15:58, 13 September 2013 (UTC)

September 13

Descriptive Subtitles

Hi I'm working with a university to release some content related to deep sea exploration, they have a video they would like to release which has descriptive captions. Is there a way of adding these to the video as a subtitles track that is on by default so that they can be translated into other languages? --Mrjohncummings (talk) 17:23, 13 September 2013 (UTC)

Commons:Timed Text... -- AnonMoos (talk) 20:18, 13 September 2013 (UTC)

Public Domain Automobile Images

I'm not an expert at navigating this site and I'm trying to find a list of public domain automobile images. I'd greatly appreciate it if someone could point me to such a place, if there is one. AutomaticStrikeout (talk) 17:24, 13 September 2013 (UTC)

You can use CatScan, here's an example. Kw0 (talk) 22:27, 13 September 2013 (UTC)

Obtaining interwiki data to copy to a Commons category

https://toolserver.org/~legoktm/cgi-bin/wikidata/copypaste.py is currently failing. - Jmabel ! talk 05:17, 14 September 2013 (UTC)

You can try InterwikiList. --Ricordisamoa 06:11, 14 September 2013 (UTC)
This looks like just what I need, but it does not have any effect for me, see wikidata:User:84user/common.js, what may I be missing? I've had to manually add links to Category:Miles Glacier Bridge despite my attempts here. I also looked in vain for any new tool link at [4]. -84user (talk) 13:27, 14 September 2013 (UTC)

Overview pages for quality processes/reviews

Hi all. I'm surprised not to find an overview page of Commons' quality processes. I'm aware of the prominent links from the Main page to featured/quality/valued, but it seems that the focus on quality and peer review is a laudable aspect of Commons that could do with more awareness among potential partners or donors. As part of the outreach work that I do, it would be useful for there to be a single bookmarkable page which gives a plain, succinct explanation of the different quality processes and how they differ. Maybe Help:Quality or Help:Reviews?

It also seems to me that for regular contributors it would be useful to have a separate overview page of reviews and other opportunities to discuss the quality of contributions, including indication of any backlogs. Maybe this could even be achieved just by a applying a category to pages such as Commons:Quality images. I welcome reactions to these two suggestions. If what I've proposed already exists and I've missed it, then apologies. MartinPoulter (talk) 15:22, 14 September 2013 (UTC)

NSFW - unexpected inclusion of images in a user's sexuality gallery?

Subsection 1

A week back I uploaded some photos of a model of Zheng He's treasure ship. They were edited yesterday to correct the spelling of Zheng (from Zeng), so I revisited them. When I looked at the image usage I saw, to my surprise, that they had been included in User:Mattbuck/Sexuality/2013 August 11-20 - a user gallery of sexual images uploaded to Commons. Apparently this is due to a bot tool that auto-creates the galleries based on images added to specified categories, although I can't figure out which category they were in that caused this.

This was a rather unpleasant surprise to be presented with, particularly as there was no explanation of what was going on. There's a long archive of this user gallery at User:Mattbuck/Sexuality, so false positives will continued to be included in these galleries for a long time. I'm also worried that this might be quite a regular occurrence as the bot seems to include such categories as Category:Snails, Category:Men and Category:Virgin Mary.

I understand the need to catch sexuality images and deal with them separately/quickly (which I understand is why these user galleries exist), but is there a way of doing this that doesn't mean that false positives are included in galleries of sexual images? Could lists be used, or could they be tagged using Commons:Abuse filter? Or if that's not possible, perhaps galleries like this should be required to have clear descriptions about what's going on, to inform users that stumble across them? Thanks. Mike Peel (talk) 16:22, 26 August 2013 (UTC)

Basically it's possible to exclude certain categories at User:OgreBot/gallery (it's protected though). I think "nudity" and "sexuality" are sufficiently descriptive names, so you can actually know what to expect when you click them. ;) I don't know if OgreBot can generate lists but I think they would make them less useful because people who patrol them would need to click on each image to check it - even on false positives. darkweasel94 16:38, 26 August 2013 (UTC)
Feel free to remove those images from the gallery. -mattbuck (Talk) 17:30, 26 August 2013 (UTC)
That doesn't address the issues Mike raises at all. — Scott talk 17:51, 26 August 2013 (UTC)
@Matt, as the listing of an image in your service-cat is not relevant from the side of the image (so to say), would it be possible to "suppress" the listing of its inclusion on the image page (i.e., so that this inclusion/use is not shown on the image page)? --Túrelio (talk) 18:17, 26 August 2013 (UTC)
I don't see how, it's a gallery page. -mattbuck (Talk) 21:46, 26 August 2013 (UTC)
From Category:Zheng He, if you go up in the category tree, after five steps you arrive at Category:Sexuality. -- Asclepias (talk) 17:50, 26 August 2013 (UTC)
Thanks for figuring that out. Five category levels sounds like far too many to me for something like this. :-/ Mike Peel (talk) 18:31, 26 August 2013 (UTC)
(ec with Fæ below) Category levels say nothing about how related the subcategory is to the parent category: you'd expect something like Sexuality -> Sex in humans -> People having sex in art -> Homosexual sex in art -> Gay sex in art -> Anal sex (gay) in art to end up in that gallery. ;) darkweasel94 18:52, 26 August 2013 (UTC)
Good point. I wonder what the probability of related vs. non-related subcategories at the various category levels is here, though. Thanks. Mike Peel (talk) 19:01, 26 August 2013 (UTC)
Magog the Ogre operates OgreBot. Personally, rather than relying solely on regular expression matches, I would add an exception list of uploader names to be filtered on, it could even be left as an open editable list. No harm in openness and it then caters for the operator not being around if there is a complaint like this. It is easy to pick up the name of the original uploader of a file from the version history, and test against a checklist, before adding to the gallery. Mike is a well known and valued contributor on Commons, please take his exception to having his uploads in this (somewhat notorious) gallery as one to take action on. Thanks -- (talk) 18:16, 26 August 2013 (UTC)
Thanks Fae, but I'm hoping that there's a solution here that works for all, rather than specific usernames being excluded. Imagine how offputting this could be if you were a new uploader and your photographs were unexpectedly included in these galleries as false positives... Thanks. Mike Peel (talk) 18:29, 26 August 2013 (UTC)
Sure, though a complaint coming from you does give this more weight. I'm actually thinking that different dimensions to the match algorithm is what would work best to counter false positives for a fuzzy topic like "sexuality" (goodness it's like living the 1990s again, thinking about fuzzy logic). Basically, we know it is an impossibility to eradicate false positives using a simple category search or even with a more complex regex match. So, perhaps the answer here is a couple of minor changes to OgreBot to allow more optional parameters of something like:
  1. List of <(title | image text page) regex (must haves | must not haves)>
  2. List of <uploader name exceptions | limited to>
I can think of other useful options, all of which I have created reports for in the past, such as EXIF data fields, file mime-type or image size, but maybe this is beyond what OgreBot was designed to do. I am sure we could create a better bot for gallery generation (though I don't really want to create something that would only haphazardly run from my old macmini). What OgreBot does is not at heart that complex, so it should be easy to improve and offer refinements for all users. -- (talk) 18:46, 26 August 2013 (UTC)
(Edit conflict) One way would be to add a big text to the top of each gallery that explains what these galleries do and why there may be some false positives in them. I was actually somewhat confused too when an image I uploaded ended up in another OgreBot gallery (the image was of Vienna and the gallery was of South Tyrol) - perhaps have a template in OgreBot's userspace (which can be translated) that it adds on top of every gallery it creates? darkweasel94 18:52, 26 August 2013 (UTC)
  • I don't get the whole fuss about this issue. It's a maintenance gallery, nothing more, nothing less. A false positive isn't the end of the world. I second the idea by Darkweasel94 to simply put a short description there, this should be sufficient in most cases.
Actually I think a real solution to the underlying problem is not to exclude images, but to reorganize categorization. A false positive is nothing else, than a hint for bad category structure (as long as categories for the bot are chosen correctly, which I'm quite confident of). --Patrick87 (talk) 20:22, 26 August 2013 (UTC)
False positives are useful to determine both bad category structures as well as find images which have had inappropriate categories added (by vandals or by mistake). I was using OgreBot for a similar but much more focused purpose (see User:Delicious carbuncle/Newly added penises/2013 March 11-20 for example. After an image of a (fully clothed) person was added to one of the pages, Magog removed my task from OgreBot's list, despite my offer to rename the pages, in case the subject complained and he was held to blame. The same situation regularly occurs on Mattbuck's "sexuality" and "nudity" pages, but the solution is improving categorization, not making Mattbuck stop collecting images. Delicious carbuncle (talk) 21:01, 26 August 2013 (UTC)
Your edit summary was "False positives are helpful, not hurtful." Consider that they can actually be hurtful for those who uploaded images that are unexpectedly added to this sort of gallery. My aim here is to see the system improved to avoid this, rather than seeing it abolished. Maybe that solution is to improve categorisation, or maybe it's to improve tagging abilities, or maybe it's another approach. Thanks. Mike Peel (talk) 21:20, 26 August 2013 (UTC)
There is nothing wrong (i.e., nothing to “correct”) with the categorization in this case:
Zheng HeChinese eunuchsEunuchsCastrationSexuality
This is correct. What is incorrect is accepting that 5 steps away from any Category:Sexuality we’re still in “NSFW” zone (for people working in Saudi Arabia, the United States, and such places). The example above about how five is still good, concerning the 5-step linkage from Category:Sexuality to Category:Anal sex (gay) in art is IMO bogus, as any of those five in itself should raise a regexp flag, as they include the word "sex" and or "gay", which should be kept away from sensitive eyes don’t you know. Reducing that 5 to 3 or even to 1 would reduce false positives. -- Tuválkin 21:28, 26 August 2013 (UTC)
But we don't have a regex bot, we have a bot which does it by category, and considering this is all to do with removing useless penis uploads, something everyone keeps complaining about, 1 is clearly not even close to enough. -mattbuck (Talk) 21:46, 26 August 2013 (UTC)
Oh, and I count 7 levels up to sexuality from Category:Human penises, erect length 50-75 mm. -mattbuck (Talk) 21:49, 26 August 2013 (UTC)
Mike Peel, I think it is a question of how the uploader interprets the appearance of their image in such a maintenance gallery. Anyone who uses Commons should know that it isn't censored and they could see explicit images at any time. In Mattbuck's case I believe he uses the galleries in an attempt to monitor new uploads. You would have to confirm this with him, but if that is true, his older galleries could be deleted after, say, monthly. (It was my intention to continue tracking the images that OgreBot was placing in galleries for me, but I could easily have renamed the galleries to something innocuous like "DcMaintenance". Perhaps I will ask Magog to reconsider following this discussion.) Delicious carbuncle (talk) 21:59, 26 August 2013 (UTC)
Yeah, it's for new uploads only really. -mattbuck (Talk) 22:02, 26 August 2013 (UTC)
Any reason why you couldn't delete the galleries older than a month, then? Delicious carbuncle (talk) 14:37, 27 August 2013 (UTC)
Laziness, pretty much. -mattbuck (Talk) 15:37, 27 August 2013 (UTC)
Can you try to overcome your laziness and delete the old ones? It will probably take less effort than posting another reply, but the choice is, of course, yours. Thanks. Delicious carbuncle (talk) 15:58, 27 August 2013 (UTC)
I've started a request to delete the old galleries. Delicious carbuncle (talk) 15:51, 28 August 2013 (UTC)

I'm still wondering why several of my uploads that had nothing to do with watercraft were included in the User:OgreBot/Watercraft galleries... AnonMoos (talk) 02:44, 27 August 2013 (UTC)

Which pages were they? Magog the Ogre (talk) (contribs) 17:53, 2 September 2013 (UTC)
File:15crossings-decorative-knot.svg, File:Flag of the Union Démocratique Bretonne.svg, etc. AnonMoos (talk) 23:24, 2 September 2013 (UTC)
Category:Knots being a subcat of Category:Sailing is probably a good bet. Carl Lindberg (talk) 23:36, 2 September 2013 (UTC)
File:15crossings-decorative-knot.svg is no longer in the category tree for Watercraft because I removed it a year ago.[5] The other file is: Category:Watercraft => Category:Boats => Category:Boating => Category:Sailing => Category:Nautical flags => Category:Gwenn ha Du => Category:Variations on flags of the Gwenn ha Du => File:Flag of the Union Démocratique Bretonne.svg. As such, I've removed a few categories from the process to avoid a repeat of the error. [6] Magog the Ogre (talk) (contribs) 23:49, 2 September 2013 (UTC)

Also curious how File:600px Taronja Valencià.png landed in User:Marcus_Cyron/OgreBotCeramics... -- AnonMoos (talk) 06:42, 4 September 2013 (UTC)

Category:Ceramics => Category:Terracotta => Category:Terracotta in architecture => Category:Brick constructions => Category:Brick structures => Category:Brick buildings => Category:Brick buildings by country => Category:Brick buildings in Spain => Category:Mudéjar in Spain => Category:Mudéjar architecture in the Land of Valencia => Category:Route of the Borgias => Category:Valencia => Category:Sports in Valencia => Category:Valencia Basket Club => File:600px Taronja Valencià.png. I removed Category:Brick constructions from the tree. Magog the Ogre (talk) (contribs) 02:07, 6 September 2013 (UTC)

Proposal

I've created a template at User:OgreBot/gallery/notice. This template can be affixed to any galleries which might cause problems. What does the community think? Template is immediately below:

Warning

Notice


This page is automatically generated by a bot from the files found in designated categories (and most of their subcategories) at the time it is run. Because subcategories (and their subcategories, and so on) are not always relevant to the designated category, and because some files or subcategories may be miscategorized, you may see a false positive. Feel free to remove any file which appears to be defamatory or slanderous in this context from the gallery.

For more information, see here.

Magog the Ogre (talk) (contribs) 17:53, 2 September 2013 (UTC)

Good idea. russavia (talk) 18:35, 2 September 2013 (UTC)
Yes, that's a good idea, although I'd quibble with some of the wording. Removed from what - the gallery, the category, Commons? And part of the problem is with files that aren't miscategorised. You could replace the last sentence with something like "Feel free to remove any file which appears to be defamatory or slanderous in this context from the gallery."
I think it would also be clearer to replace "from the categories available" (in the first sentence) with "from the files available in designated categories (and most of their subcategories)". Even better would be to change "designated categories" to the specific top-level category/categories used, if that's not too much work. --Avenue (talk) 00:14, 3 September 2013 (UTC)
Please make the edit yourself, for CC-BY-SA attribution purposes, and because I can't write worth a damn and would just copypaste your text. Magog the Ogre (talk) (contribs) 00:17, 3 September 2013 (UTC)
Okay, done. I've wikilinked to the configuration page instead of trying to specify which categories are used for each particular gallery. --Avenue (talk) 04:04, 3 September 2013 (UTC)
Magog, once this is in place, can you please re-add the job of mine that you removed some months ago? Thanks. Delicious carbuncle (talk) 03:34, 3 September 2013 (UTC)
Can you agree to the terms at User:OgreBot/gallery/notice? Magog the Ogre (talk) (contribs) 21:59, 4 September 2013 (UTC)
Yes, so long as it applies to everyone else with similar galleries. Delicious carbuncle (talk) 21:26, 8 September 2013 (UTC)
This is all good but, as mentioned many times in this discussion, it is not about miscategorization, not necessarily, so keep it off the warning box. The Zheng He case, that started this (this time), was not miscategorized, it doesn’t need any “fix” — and probably many of the “false positives” are likewise not false at all, just unexpected by people who think sex is something separate (or separatable) from human experience, be it medieval sailors or electric toothbrushes. So, please, change the wording in the warning box — as it is now, it shifts the blame on people doing the categorization, and that’s both unfair and inaccurate. -- Tuválkin 07:00, 7 September 2013 (UTC)
Please provide a reasonable alternative. Magog the Ogre (talk) (contribs) 14:19, 7 September 2013 (UTC)
How about changing the middle sentence to read "Because some files or subcategories may be miscategorized, and because subcategories (and sub-subcategories, etc) are not always entirely relevant, you may see a false positive." The new bit is underlined. --Avenue (talk) 14:56, 7 September 2013 (UTC)
I went ahead and added something along those lines, although I switched it round for better readability. --Avenue (talk) 21:27, 9 September 2013 (UTC)
Implementing this template would be a good step forward - thanks for doing this! It would be good practice to have this (or possibly one that doesn't specifically say "defamatory or slanderous") at the top of non-problematic galleries just so that it's obvious how and why the page was created to anyone that visits that page. Thanks. Mike Peel (talk) 19:59, 14 September 2013 (UTC)

August 27

Essay: Language guide

Does Commons have the equivalent of an "Essay" which is a suggestion in a written form from Wikipedia and not a policy or guideline? An idea I have is a "language guide" which gives general advice on what languages to add to media or to post to the translation request depending on the country/countries involved in the media. That way it can function as a suggested guide for internationalization for the the Wikimedia Commons can internationalize.

Here is a preliminary, very general list of several major languages:

  • English: Worldwide (Every media file on the Wikimedia Commons should have an English description)
  • Spanish: Spain, Mexico, Central America, Spanish-speaking Caribbean and South America, United States (especially Puerto Rico, U.S. states bordering Mexico, Florida, and major cities as well as some state and federal-related topics), Equatorial Guinea
  • French: France, Canada (especially Quebec, New Brunswick, and Ontario), Belgium, Switzerland, Francophone Africa, the Maghreb (Morocco, Algeria, and Tunisia), the French Caribbean, and the French Pacific
  • Russian: Russia and the former Soviet Union
  • Modern Standard Arabic: Arab League, Israel
  • Mandarin Chinese: Mainland China, Taiwan, Hong Kong, Macau, Malaysia, Singapore
  • German: Germany, Austria, Switzerland
  • Italian: Italy, Switzerland
  • Japanese: Japan, Taiwan, Guam, Northern Mariana Islands, Federated States of Micronesia, Marshall Islands
  • Korean: South Korea and North Korea
  • Portuguese: Portugal, Brazil, Portuguese Africa

WhisperToMe (talk) 17:49, 13 September 2013 (UTC)

Yes, you can just use {{essay}}. darkweasel94 19:17, 13 September 2013 (UTC)
Thank you WhisperToMe (talk) 20:07, 13 September 2013 (UTC)
I started Commons:Language guide as an essay. If anyone wants I can add more popular languages and write country suggestions. WhisperToMe (talk) 20:11, 13 September 2013 (UTC)
It is of course good to add descriptions in English and local/regional languages, but I very much wonder weather translations should be requested. Just look at any local category, such as Category:People of the Marshall Islands or Culture of Macau and you will find half the files having a description only in one language (local, English or native language of the visitor).
Anybody who want to translate can just pick any local category and start working on translations. If you add translations requests for such images you happen to come by, the translation request list will just be a random sample of files needing translation work.
Such "random" requests will not help translators, but rather make some of them concentrate on this sample instead of thinking for themselves about priorities or use the list for files the translations of which are especially important.
--LPfi (talk) 07:56, 14 September 2013 (UTC)
I have some doubts about translation descriptions:
  • Is there actually any significant amount of people translating descriptions? If it is, I never noticed, although my watchlist is long.
  • Ideally, any media, category and page should have translation to every language, but that isn't feasible. Directing efforts to a geographical area won't solve it. If I had to suggest media to describe, I would say that the most useful descriptions are those of very used media where description is not self evident.--Pere prlpz (talk) 14:26, 15 September 2013 (UTC)
I suggest that you discuss that on Commons talk:Language guide. darkweasel94 14:34, 15 September 2013 (UTC)

Merge multiple deletion requests in to one

Hello again all,

Today a user has nominated 55 files I had previously uploaded for deletion. They have all been done as separate deletion requests. As I unfortunately don't have sufficient time to reply to each individual deletion request, is there anyone here who can merge them all in to one single "multiple deletion request"? That way, the discussion regarding these files can be centralised in one place.

They are all found on this page. It is deletion requests 14 through to 68 inclusive on that page (all the ones which were created today) - from Commons:Deletion requests/File:Bolton Art Gallery, Library and Museum (2).JPG to Commons:Deletion requests/File:Tranmere Cross (4).JPG. Of course if it's not appropriate to merge them all into one then please let me know. It's just that I'll be unlikely to make my viewpoint known on each of the separate pages due to the sheer number of them. Thanks again for any possible help. Rept0n1x (talk) 18:00, 13 September 2013 (UTC)

It's OK please disregard the above request, I've contacted the nominator and he has reminded me that it's not possible to merge the deletion requests this time, because each request has a separate link in it. So I gather it can't be done in this case, thanks anyway Rept0n1x (talk) 18:22, 13 September 2013 (UTC)
This is ridiculous. These images should never have been nominated for deletion in the first place. Multichill (talk) 10:09, 14 September 2013 (UTC)
Please notice that User:Kulmalukko has been starting this kind of deletion requests for months with the same "similar images" rationale (even views of the same city taken months apart) and doing individual deletion requests instead of one multiple request for "similar" images with same uploader and subject (just making more difficult for uploader to follow and answer deletion requests, especially for inexperienced users). I think this way of proceeding needs a review and probably some guidance to Kulmalukko would be good.--Pere prlpz (talk) 15:36, 15 September 2013 (UTC)

September 14

Copyright status Macduff Railway station

This image File:Macduff Railway station.jpg is scanned from a postcard, believed to be circa 1900. Because it is from a postcard, I think that qualifies as published. If I am reading Commons:Copyright_rules_by_territory#United_Kingdom correctly,

Photographs created before 30 June 1957: 70 years after creation if unpublished, 70 years after publication if published within 70 years of creation

as long as it was published prior to 1943, it should now be in the public domain.

I think this means the appropriate template is {{PD-UK-unknown}}. Does this sound right?--Sphilbrick (talk) 17:53, 14 September 2013 (UTC)

Yes, and also the second template should be {{PD-1923}}. Ruslik (talk) 18:57, 14 September 2013 (UTC)
Thanks.--Sphilbrick (talk) 21:18, 14 September 2013 (UTC)
Btw, the uncropped version is shown there. -- Asclepias (talk) 01:59, 15 September 2013 (UTC)

September 15

Cannot upload File to Wikimedia!

I need to use this File, but cannot upload to Wikimedia by anyway! Was I blocked? Or the files upload of Wikimedia Commons have bug? -–ArikamaI (talk) 03:00, 14 September, 2013 (UTC)

✓ Done --Zhuyifei1999 (talk) 05:00, 14 September 2013 (UTC)
Also, this File cannot either. -–ArikamaI (talk) 10:18, 14 September, 2013 (UTC)
✓ Done I'm wondering, what error message did you get when uploading? --Zhuyifei1999 (talk) 15:46, 14 September 2013 (UTC)
Error message and steps to reproduce highly welcome. Also see https://www.mediawiki.org/wiki/How_to_report_a_bug for general information. --AKlapper (WMF) (talk) 16:13, 16 September 2013 (UTC)

PD art with copyright notice in EXIF

How do we deal with EXIF on images of PD art works, containing copyright notices, like that in File:The Ringers of Launcells Tower - Frederick Smallfield.jpg? Andy Mabbett (talk) 15:48, 15 September 2013 (UTC)

I asked Dcoetzee about a similar image (File:Ebenezer Elliott 2.jpg) a while back, and he tagged it with PD-Art, which seems to give enough detail to explain the copyright situation to potential re-users. I've tagged yours with PD-Art. INeverCry 18:40, 15 September 2013 (UTC)
I just noticed he also put a photography credit in the author field as well. INeverCry 18:43, 15 September 2013 (UTC)
We make sure the copyright information on the description page faithfully preserves and reflects the copyright notice from the source or contained in the EXIF. The potential reusers must be made aware of any copyright restrictions that may exist on the image, and we must notify them of all the relevant information so they can take their decisions in full knowledge. For example, if the description page linked above has only a "PD-old-70" tag, a distracted reader might not notice the copyright restrictions and might be mislead into believing that the image is fully freely usable anywhere. Thus, it may be a good idea to replace the "PD-old-70" tag it with a tag "PD-Art-two|PD-old-90|PD-1923", or the equivalent as the case requires, as the PD-Art wrap provides both an explanation of why the file is on Commons and, more importantly in relation to your question, a warning to the potential reusers that the use of the image is restricted in some jurisdictions and a link to the help page Commons:Reuse of PD-Art photographs. The generic "PD-Art" wrap, or more likely the "PD-Art-two" wrap, as those cases typically involve reproductions from non-U.S. sources, often British, can be used when the reproduction is fully restricted, or when it is licensed with a licence for which no license tag exists on Commons. It can be supplemented, if useful, in the permission field of the information template or the equivalent in the Artwork template, with a more complete explanation or with a link to the copyright notice of the source. If the reproduction is licensed with a license for which a license tag exists on Commons, then the wrap "Licensed-PD-Art", or more likely the wrap "Licensed-PD-Art-two", may be preferred. -- Asclepias (talk) 19:06, 15 September 2013 (UTC)
The photographer should be listed in the author field, if available. EXIF data like this that claims full copyright to a PD-Art work is misleading and contradicts the more correct and thorough advice presented in the PD-Art template (which links Commons:Reuse of PD-Art photographs, which presents info on local jurisdictions). On the other hand, the EXIF data is also small and way down at the bottom, so it's not a huge deal. I'd say remove it if you're so inclined, and otherwise just ignore it. However, if you remove it, you should preserve the information in the file description (e.g. in the Notes field put "The Royal Institution of Cornwall claims rights to this reproduction in the UK; Supplied by The Public Catalogue Foundation"). Note that these images also contain invisible watermarks used to track use of the image, and these cannot be removed without substantial image degradation - we should assume they will issue complaints to Commons at some point and be prepared to respond. Dcoetzee (talk) 02:11, 16 September 2013 (UTC)

September 16

Proposed revision to PD-Art text

See Template_talk:PD-Art#Proposed_revision_for_brevity_and_clarity. Please respond there. Dcoetzee (talk) 02:28, 16 September 2013 (UTC)

derivativeFX

I cannot access http://toolserver.org/~luxo/derivativeFX/deri1.php today. (Indeed, toolserver.org seems to be down.)

I’d like also to give a heads up about this tool — although it has been used to upload over 100 thousand derivative images, assisting the uploaders with categorizing, with linking to and from the files, and with filepage contents in general, it has been severely neglected by WMF (big surprise), especially shocking in view of the invenstemnt made in several other upload tools.

Users of derivativeFX were over one year ago forced to repeat a step in the upload process (which could be partly avoided by judicous use of the browser’s back function). More recently, the intervening upload form stopped inheriting the file page contents generated by derivativeFX, forcing its users to manually paste it. (At the same time users were forced to add a hack to their .js page in order to avoid the “improved” upload form.)

So, the question is: What is the reccomended way to upload derivatives of Commons’ files, according to the WMF, if derivativeFX was so bad it had to be sabotaged out of the water? -- Tuválkin 05:44, 16 September 2013 (UTC)

Seems to be up again. But since it looks like the toolserver will be shut down sooner or later, we'll probably need something else. Doesn't seem like anyone is working on porting derivativeFX to wmflabs? Can't believe that our glorious *cough* UploadWizard is still not able to handle derivative works of files from Commons. --El Grafo (talk) 13:15, 16 September 2013 (UTC)
Feel free to vote for the request as soon as voting is open. ( Rfc technical needs ) --McZusatz (talk) 20:10, 16 September 2013 (UTC)
Back when it was announced that toolserver.org was going to be pulled the rug under, I perused the list of its tools and tried to vote for derivativeFX (or something like it) to be ported; I remember that that Mediawiki (or was it Meta?) page was not very user friendly… I’ll certainly remember all these issues when voting next time. -- Tuválkin 07:27, 17 September 2013 (UTC)

Giving a copyvio notice on a YouTube piece

Anyone know how I can complain about the copyvio (my photo used uncredited in a YouTube video) I mention at https://commons.wikimedia.org/wiki/File_talk:Ned_Raggett_02.jpg? I can't see any obvious way there to file a complaint. - Jmabel ! talk 14:52, 16 September 2013 (UTC)

Have you tried the little flag-like button below the number of views (labeled "Report" on mouseover)? --El Grafo (talk) 15:04, 16 September 2013 (UTC)
http://www.youtube.com/yt/copyright/copyright-complaint.html JKadavoor Jee 15:13, 16 September 2013 (UTC)
  • Jeez. The only remedy they let me ask for there is removal. I don't want it removed, I just want to be credited. I'm going to need to mull this over. - Jmabel ! talk 22:24, 16 September 2013 (UTC)
    • Isn't there a way to add a comment about the film or contact the uploader if you create an account and log in to Youtube? --Stefan4 (talk) 22:36, 16 September 2013 (UTC)
  • There is a related discussion at w:User_talk:Jimbo_Wales#New_way_to_profit_from_Wikipedia.... JKadavoor Jee 02:35, 17 September 2013 (UTC)
    • Thanks, Jkadavoor. I posted a remark there. - Jmabel ! talk 03:31, 17 September 2013 (UTC)
  • Ew. That youtube account is violating the copyrights of Commons users at an industrial pace. He now has over 29000 of those "videos" automatically generated from Wikipedia articles. Any regular Commons user here probably has some of his works illegally used there. I searched his youtube list for the first of my photos that came to mind, and there it was, without attribution and without proper license. He doesn't even give correct links to Wikipedia. I might have spared an ounce of sympathy if the video wasn't complete garbage. But it's just a computer trying to read the infobox of a Wikipedia article with an incomprehensible result while the whole time my photo is being displayed on the screen with wrong proportions. Bleh. -- Asclepias (talk) 04:24, 17 September 2013 (UTC)
  • If you find videos by this author using your content, use the form at [7] and YouTube should take it down quickly. If many of us do this, then eventually the account will be reviewed and banned. I'm not myself eligible to participate because my works are CC0, but anyone else may do so if they wish. Dcoetzee (talk) 07:35, 17 September 2013 (UTC)

Request for review

Hi, I've created a UploadWizard campaign version specifically for the art and design students that will take part in the wikiArS initiative: Campaign:wikiArS. The aim is to facilitate them to upload their artworks, to choose the appropriate options and to hide the fields their don't need. I would appreciate if you can review (and improve, if necessary) the texts and if you can participate translating it to other languages (now some of them are translated into Spanish and Catalan, but not all). The used templates are linked in this page: Commons:WikiArS/Upload campaign and from there you can access to English text or click a button to add a new language. Comments and feedback are welcome. --Dvdgmz (talk) 10:24, 17 September 2013 (UTC)

September 18

Interwiki links via Wikidata coming soon

Commons Interwiki links before implementation of wikidata
Proposed Commons Interwiki links using wikidata: alternative 1.
This model can be refined by creating items for Commons Galleries (sample: d:Q14916142) and linking them to the items for the article (d:WD:Property proposal/Generic#Main Commons gallery topic).
Proposed Commons Interwiki links using wikidata: alternative 2

Heya folks :)

I just wanted to let you know that the Wikidata development team has been working on enabling interwiki links via Wikidata for Commons. We're now planning to enable this on 23rd of September.

What does this mean for you exactly? The links connecting pages here to Wikipedia (Category:China -> en:Category:China and China -> en:China for example) are currently in the wikitext. They can then be removed from there and moved to Wikidata. There they are maintained only once in one central place and are easier to maintain. To see how this looks like for Wikipedia and Wikivoyage have a look at the bottom of d:Q148.

Some more tips:

  • It'll no longer be needed to keep them in the wikitext like it is currently.
  • It'll still be possible to do so however but this will overwrite the links coming from Wikidata.
  • With the magicword noexternallanglinks links from Wikidata can be turned off on an article either for all languages or only specific ones.

Be aware that this is only about the interwiki links. Access to data on Wikidata is not yet possible.

Please let me know if you have any questions. --Lydia Pintscher (WMDE) (talk) 12:26, 8 September 2013 (UTC)


\o/ And there was much rejoicing :)
Question, to be entirely clear: in order for the system to "know" that Category:China is linked to d:Q148, we will have to add it to d:Q148 as a sitelink ("Page of Wikimedia Commons linked to the element"), effectively deprecating d:P:P373?
Jean-Fred (talk) 13:11, 8 September 2013 (UTC)
Is d:P:P373 already included through {{#property:P373|of=item}} in a Wikipedia? -- Rillke(q?) 13:41, 8 September 2013 (UTC)
Yes you'll have to add it there. If or if not you want to deprecate the property is up to the community :) --Lydia Pintscher (WMDE) (talk) 13:15, 8 September 2013 (UTC)
Uhm wait - the interwiki links from Commons categories are going to lead to Wikipedia categories? Sorry but that doesn't make much sense, currently they usually lead to Wikipedia articles, given that few topics even have galleries. Or will it be possible to link a Commons category to a Wikipedia article? darkweasel94 13:33, 8 September 2013 (UTC)
I second that query. We need cross-namespace interwiki links (e.g. Category:China <-> en:China). Going ahead without them would do more harm than good. --Avenue (talk) 14:35, 8 September 2013 (UTC)
It's true that Wikipedia usually has an article where we have a category and that Wikipedia often does not have a category where we have one and that Commons often does not have a gallery where Wikipedia has an Article. -- Rillke(q?) 14:51, 8 September 2013 (UTC)
I am very much with Avenue & Rillke on this, especially the remark about "more harm than good." Please be careful how you do this. - Jmabel ! talk 18:30, 8 September 2013 (UTC)
Linking it this way will also be possible. Up to the community to decide how to do it really. --Lydia Pintscher (WMDE) (talk) 10:03, 9 September 2013 (UTC)
Thanks, I'm glad to hear that. I'd say that in many cases the communities here and on en WP have already decided what the best links are, and it's just a matter of transferring them across to Wikidata. --Avenue (talk) 10:18, 9 September 2013 (UTC)
That's great news because Commons categories are often moved without even being able to update all the commons-cat-links. -- Rillke(q?) 13:41, 8 September 2013 (UTC)
I can see this will basically be a great improvement. A small question: Can two pages on Commons be linked to the same Wikidata item? I remember seeing Category:X-Y-Z and X-Y-Z linked to the same list of Wikipedia interwikis, so some (probably very small number of) things may not be converted well to the Wikidata way. It might not be a huge problem as long as we can resort to interwiki-ing by wikitext for these exceptionsl cases, though. --whym (talk) 10:45, 9 September 2013 (UTC)
No linking two will not be possible. For those cases interwiki links will have to stay in the wikitext for now. Hopefully this is only a small number of pages though as you say. --Lydia Pintscher (WMDE) (talk) 11:02, 9 September 2013 (UTC)
Yes — this will probably not be news to anyone, but indeed Commons’ pages/articles such as X-Y-Z are, compared to Category:X-Y-Z, very scarce and, in most of those seldom cases (I’d say >99%), completely useless — either abbandoned or created/mantained by less astute editors (or created by those and then abbandoned). Useful Commons’ pages/articles are truely exceptional, and much of its intended contents (manual galleries, explanations on categorizing criteria and terminology, etc.) is also/instead found in category pages’ intro sections. In short, ignoring Commons’ pages/articles in Wikidata would not be a great loss, since manually added wikitext langlinks can be created. -- Tuválkin 18:59, 9 September 2013 (UTC)
Ever looked up biological species in Common? For these it is common practice to have an article/gallery in parallel with the corresponding species category, in particular if the cat contains a large number of files. These articles are definitely not rare - and they neither seem exceptional nor useless to me. --Burkhard (talk) 18:11, 12 September 2013 (UTC)
Yes, I did, many times (I’m a biologist by train, even if not by trade). Some are terrific (and probably should be moved to Wikispecies), a few suffer from the same kind issues diagnosed above (see this recent example: Commons:Deletion requests/Zophobas morio). However, the matter is: What should be the default target for interwiki links to and from Commons — categories or pages? Seems that categories win, regardless of some really good pages around. Those can be manually linked. -- Tuválkin 06:15, 18 September 2013 (UTC)
  • Very exciting news! Will the interwiki links be shown also on Wikipedias or will this be unidirectional for now? If it's not unidirectional, can we see a mockup of how it will look on Wikipedias? --Nemo 11:36, 9 September 2013 (UTC)
Yey, that is great. A lot of good points were raised here (I guess many people had a LOT of time to think about them Clin ). I agree with darkweasel, Rillke, Avenue and others that the most crucial issue is how are we going to work with the limitation of connecting one Commons page to wikidata and wikipedia articles. Currently we do have interwiki links connecting the usual pages like templates to templates, users to users, categories to categories, galleries to articles and most frequently categories to articles (see images on the right). Ideally all wikipedia articles should have a category associated with it. However since we are limited to only one commons page we probably should concentrate on connecting our categories to wikipedia articles whenever possible. Interwiki links from gallery pages should be maintained manually, or replaced with a template pointing to wikidata, and interwiki links between our categories and wikidata category pages would only be used if there is no article on the subject, (see second diagram on the right). For example both Category:Pablo Picasso and gallery Pablo Picasso have interwiki links to articles in d:Q2628198, however only Category:Pablo Picasso links will be maintained by wikidata and Pablo Picasso links will have to be maintained manually. --Jarekt (talk) 16:35, 9 September 2013 (UTC)
I wrote an email about this some time ago. In short:
Haarlem -> d:Q9920 -> w:Haarlem
Category:Haarlem -> d:Q7427769 -> w:Category:Haarlem
d:Q9920 are d:Q7427769 connected so you can use some logic to grab the data. Multichill (talk) 17:50, 9 September 2013 (UTC)
Proposed the property to get the magic working at d:Wikidata:Property proposal/Generic#Topic main category. Multichill (talk) 18:23, 9 September 2013 (UTC)
That would be the cleanest solution for storing the data, and I would be OK with it as long as there is some plan to implement some fallback logic so the approach works for majority of the cases where Commons only have a category and wikipedias only have an articles. In such case, if wikipedia does not have a category we would link to an article. So we and up with
Eugeniusz Lokajski -> d:Q735173 -> w:Eugeniusz Lokajski
Category:Eugeniusz Lokajski -> [[d:???????]] -> w:Category:Eugeniusz Lokajski
We would have to create a lot of redundant wikidata pages for our categories. In case of Haarlem, d:Q9920 and d:Q7427769 should also share properties so lets say its population or country can be accessed from either category or article namespace.--Jarekt (talk) 18:31, 9 September 2013 (UTC)
Hi Jarek, to answer your questions:
Yes we have to create Wikidata items for our categories if they don't exist yet, but these are not redundant, we're using them! Who cares about a million more items, we have bots for that and this way we have a clear data model.
d:Q9920 and d:Q7427769 wouldn't share properties. d:Q9920 contains all the relevant information and from Category:Haarlem we would access this information using d:Q7427769 with d:Property:P301 as an intermediate step. Multichill (talk) 18:55, 9 September 2013 (UTC)
What worries me is that until there is some smarter wikidata software associating category and article records, our categories will loose interwiki links we learned to rely on. For example, Category:Eugeniusz Lokajski now has interwiki links to w:Eugeniusz Lokajski, etc. Under new system it will be associated with new wikidata record which will not be associated with any pages on any wikipedia. Eventually this new record will get associated with d:Q735173 somehow, but are we ever going to see our regular interwiki links to w:Eugeniusz Lokajski? --Jarekt (talk) 13:08, 10 September 2013 (UTC)
Hi Jarek, we will only loose interwiki links if we remove them. Let's agree on not doing that until we have something that replaces it.
This first step will be of most benefit for the Wikipedia -> Commons links, not for Commons -> Wikipedia.
If we look a bit further than we'll have access to all the linked records, so in this case we can use lua to go from Category:Eugeniusz Lokajski -> Q for the category -> d:Property:P301 -> d:Q735173 -> give us labels, descriptions and interwiki's. This is just a matter of time. We could make a template that would be a bit like sum-it-up, but not static, no fetch the data from Wikidata. With that feature we'll also solve the whole creator template problem. Multichill (talk) 19:30, 10 September 2013 (UTC)
  • On en.wiki categories are normally named after articles in some fashion and they are moved as a matter of housekeeping when the article is moved. Commons categories typically follow en.wiki names for English-language subjects. With this formal linking, will those changes be reflected automatically as name changes in commons? (i.e. has a bot been developed?) or will the amount of housekeeping involved in a move increase? 130.195.179.107 01:23, 11 September 2013 (UTC)
    • I am not sure I understand it correctly but here is what is currently happening already for Wikipedia and Wikivoyage: If an article is moved on a Wikipedia then the link will also be changed in the Wikidata item automatically. Thereby all other language versions of Wikipedia automatically get the right link (if they don't have them defined locally in the wikitext). The same thing would happen if a page is moved here. What's not happening automatically is a move of the page here when the enwp article is moved. A bot could be developed for that but doesn't exist yet. If there is general agreement that this is what should always be done then we can maybe also talk about implementing this in the Wikidata software directly. But from my understanding of things right now this doesn't seem to be a good idea. Please ask more questions if this doesn't answer your initial question. --Lydia Pintscher (WMDE) (talk) 11:56, 11 September 2013 (UTC)
      • A move on English Wikipedia is not in itself a reason to move a Commons page. The most obvious example is where we use a proper name in a local language and English Wikipedia uses a (less known) English name. I strongly object giving the English Wikipedia such power over the multilingual Commons project. The reasons for a move on a Wikipedia may or may not apply to Commons. --LPfi (talk) 08:17, 16 September 2013 (UTC)
        We are not talking about the name of the Commons page (which is going to stay the same), we are talking about where the links on that page are pointing out.--Ymblanter (talk) 09:05, 16 September 2013 (UTC)
Is there a new template in w:Category:Wikimedia_Commons_templates which uses some of these capabilities (like confirms that there linked-to cat exists)? 130.195.179.107 01:36, 11 September 2013 (UTC)
Not sure I understand the question ; is en:Category:Commons category Wikidata tracking categories what you are looking for? Jean-Fred (talk) 08:54, 11 September 2013 (UTC)
Related to this, I wrote down my ideas at User:Multichill/Commons Wikidata roadmap. Multichill (talk) 13:33, 14 September 2013 (UTC)

The resulting scheme would be:

Sample item Wikipedia categories
(linked through WD interwikis)
Articles/Commons categories
(linked through WD interwikis)
Commons galleries
(linked through WD interwikis)
Commons -/- Category:Pablo Picasso Pablo Picasso
Wikidata (WD) d:Q9062435 d:Q5593 d:Q14916142
Wikipedias (e.g. en) en:Category:Pablo Picasso en:Pablo Picasso -/-

On Wikidata, properties would link d:Q9062435 + d:Q5593 + d:Q14916142 together.

Stats Wikipedia categories
(linked through WD interwikis)
Articles/Commons categories
(linked through WD interwikis,
Commons currently: 800'000* with P373)
Commons galleries
(linked through WD interwikis)
Commons -/- 2,950,000 categories 110,000 galleries
Wikidata 1,050,000* 4,390,000* to create: 110,000
Wikipedias (e.g. en) 1,050,000 categories 4,390,000 articles -/-

(*) assumptions

Hope this helps. --  Docu  at 23:23, 15 September 2013 (UTC)

Student interested in working with sound files

(Copied from Commons talk:Village pump:)

I run a program at my school to allow students to work with Wikimedia to complete community service requirements at both the high school (International Baccalaureate) and the undergrad level (required of all Mexican candidates). For the latter, the idea is to relate this community service to their major, as to get some job experience as well. I have a student who works with sound engineering and I thought of adding pronunciation files to many of the WP articles on Mexican people and places, which lack them. However, I dont see a sound editing community or help pages specifically for sound files. Did I miss something? Anyone willing/able to help us get started? Thelmadatter (talk) 14:54, 13 September 2013 (UTC)

I've just created Commons:Audio. --McZusatz (talk) 16:31, 13 September 2013 (UTC)
Also note that work is going on to create a Prononciation Recording Tool which might make things way easier for everybody who wants to contribute a prononciation. --AKlapper (WMF) (talk) 16:12, 16 September 2013 (UTC)
For adding pronunciations to the English Wikipedia, please see en:Wikipedia:WikiProject Spoken Wikipedia/Pronunciation task force which I recently updated. Dcoetzee (talk) 03:01, 18 September 2013 (UTC)

Squashed and squeezed thumbs

Кингисепп, Екатерининский собор.JPG
Tower of Pisa and Clouds.JPG

(Ouch!) I found these two (see discussion): They look like anamorphically distorded images, good for {{DistordedAspectRatio}}, but when you follow the direct image link, or when you open a downloaded copy, they are okay. Is this a known bug? What causes this? Can it be fixed? -- Tuválkin 06:27, 18 September 2013 (UTC) Typo: {{DistortedAspectRatio}}. -- Tuválkin 11:55, 9 November 2013 (UTC)

Purging the thumbnails fixed the problem. Both images have a 404 as a first version; something has gone wrong with image scaler in thumbnail generation at some point. MKFI (talk) 06:34, 18 September 2013 (UTC)
Yay, it’s fixed. Thanks! -- Tuválkin 07:35, 18 September 2013 (UTC)
re 404- if by 404 you mean 500, there is an ongoing issue where thumbnails being scaled by mw1154 seem to blow up, which is bugzilla:54045. This is probably unrelated to the original squishing issue caused by varnish cahing of thumbnails suckinh. Bawolff (talk) 23:14, 18 September 2013 (UTC)


A bot for this?

Is there a bot that could add Category:Photographs by Lee Paxton to all of the images listed User:GrapedApe/LeePaxton? --GrapedApe (talk) 12:05, 18 September 2013 (UTC)

You don't need a bot, you can do that from your own user account with COM:VFC. darkweasel94 12:08, 18 September 2013 (UTC)
He can, you can, I can too. ;-) ✓ Done -- Tuválkin 12:22, 18 September 2013 (UTC)
Sure, but you know the old saying: give a man a fish and you'll feed him for one day, teach him to fish and you'll feed him for all his life. :) darkweasel94 12:51, 18 September 2013 (UTC)
Thanks for the fish, and also thanks for teaching me how to fish!--GrapedApe (talk) 00:15, 19 September 2013 (UTC)

September 19

Where's the upload button?

...didn't there used to be an "Upload" link in the toolbox on the left side? Where is it now? The only way I can find a way to upload something through Commons now is by going via "Special pages"...there's no link at all for either the Upload Wizard or Special:Upload anywhere else that I can find... - The Bushranger (talk) 00:30, 19 September 2013 (UTC)

Toolbox refers to "What links here, Related changes, etc"? I cannot see a link to UploadWizard either, wondering if this is a regression (if that link was there before). --AKlapper (WMF) (talk) 11:33, 19 September 2013 (UTC)
Participate>Upload file? JKadavoor Jee 11:46, 19 September 2013 (UTC)
Its still in toolbox, but being hidden by css/js (you can see it in source code). However, the larger issue is the wrong translation is being used for MediaWiki:Sidebar for some people (or at least me). I'm seeing the default sidebar, instead of what's on mediawiki:Sidebar, which adds the participate section. This is a bug (And if it happens on non-english wikis, possibly a rather serious one). Bawolff (talk) 14:50, 19 September 2013 (UTC)
Filed as bugzilla:54326 Bawolff (talk) 14:55, 19 September 2013 (UTC)
Ah, the 'participate' section with the 'Upload file' link (as well as the 'Village pump' link in 'Navigate' which was gone before!) is now there again., thanks. - The Bushranger (talk) 17:51, 19 September 2013 (UTC)

Templates failed

User:HK Arun/Wikimania 2013 should probably be a cat and not a page. Rename and move templates both failed.--Canoe1967 (talk) 06:44, 19 September 2013 (UTC)

Moving from other namespaces to the category namespace is impossible. -- Rillke(q?) 15:29, 19 September 2013 (UTC)

Questions regarding Flickr user Gcdgraphics

Hi, something has been bothering me over the last few days, the user mentioned above has released all his images on flickr under a "free licence" and two years ago this was discussed here on Village pump regarding the OTRS release of his images, what i want to ask is, Does the user really own the rights to those images?. Around 400 of his images have been uploaded to wikimedia and most of them are used across all the wikis. I ask this question because all his images are a bit suspicious, none have EXIF data which i know is not important but all are of good quality and yet very very random. There is usually only ONE image of a celebrity, whereas other flickr images uploaded from the same event by other users on flickr will have more than a few...some are also "edited" (photoshopped) such as this pic of actor, James Gandofini whose metadata shows that the image was "Photo shopped". My questions is, do we really know if that person actually takes those pictures or does he take them from somewhere else, edits them heavily and then posts them to his flickr?....If you look at the set from where that pic was taken (http://www.flickr.com/photos/gdcgraphics/6189999947/in/set-72157627506260955/), its very random (like someone uploading random pics to a set) and seemingly perfect shots.....either way, does he actually own the rights for those images?--Stemoc (talk) 06:25, 19 September 2013 (UTC)

I think they are OK. They seem to be from Toronto International Film Festival 2011, and many images have the same background. --Jarekt (talk) 15:46, 19 September 2013 (UTC)
To have any real doubt, you'd need to show that the pictures couldn't have been taken by one person in the way that is claimed (e.g., a particular subject didn't attend the Toronto Film Festival), or that the copyright is claimed by somebody else too. ghouston (talk) 23:02, 19 September 2013 (UTC)
That would be hard to prove as those images are old and have been used by many websites once uploaded on wikimedia, the "graphics" in his name does mean he manipulates images as i don't think any camera in the world can take a pic like >this...whatever you guys think is appropriate, I'll avoid using his images on any articles I link to on any wikis though...--Stemoc (talk) 00:51, 20 September 2013 (UTC)

My flickr account has a profile and in that profile is my email. Contact me and I will be happy to email thumbs of over 40 images of James Gandolfini from that event in 2011. And then please clear any confusion on here. Thanks gdcgraphics.

I've been in correspondence with gdcgraphics, and I can confirm he has sent me via email a contact sheet showing photos from the event, so there is no reason to doubt authorship of the images. Thanks gdcgraphics for making your photos under a CC licence, and I hope you will continue to do so in the future. Cheers, russavia (talk) 05:22, 20 September 2013 (UTC)
Checkmark This section is resolved and can be archived. If you disagree, replace this template with your comment. Jarekt (talk) 12:08, 20 September 2013 (UTC)

Preload Special:Upload with custom upload description

Hi all,

can someone tell me how to preload Special:Upload&uploadformstyle=basic with a customized upload description (or "summary" as the input is called)?

E.g. I want to add == {{int:filedesc}} == and == {{int:license-header}} == by default.

Regards, --Patrick87 (talk) 01:49, 19 September 2013 (UTC)

Use wpUploadDescription parameter - aka https://commons.wikimedia.org/w/index.php?title=Special:Upload&uploadformstyle=basic&wpUploadDescription=YoHo! Bawolff (talk) 14:57, 19 September 2013 (UTC)
Use Commons:User scripts/UploadLink. -- Rillke(q?) 15:19, 19 September 2013 (UTC)
Less flexible than the one provided by Rillke, but Commons:Upload Wizard/Fields prefilling might be useful too. Jean-Fred (talk) 15:53, 19 September 2013 (UTC)
Thank you both. Is there also an easy possibility to do this with JavaScript, so I can continue to use the default upload links of the MediaWiki UI? Or should I rather change all links using JavaScript (which sounds a bit hacky to me). --Patrick87 (talk) 15:56, 19 September 2013 (UTC)
Actually I already tried to use the following code in my common.js:
addOnloadHook(function () {
   if (document.getElementById('wpUploadDescription'))
      document.getElementById('wpUploadDescription').value  = "test";
});
but it does only work for the "plain" style of the upload form, not the "basic" style. --Patrick87 (talk) 16:03, 19 September 2013 (UTC)
// For the simplest of all forms:
window.JSconfig.keys['loadAutoInformationTemplate'] = false;
window.JSconfig.keys['UploadForm_newlayout'] = false;
var prefill = [
        '{{Information',
        '|Description={{en|1=test}}',
        '|Source=',
        '|Date=',
        '|Author=',
        '|Permission=',
        '|other_versions=',
        '}}'].join('\n');
if (!mw.util.getParamValue('wpUploadDescription') && !mw.util.getParamValue('wpForReUpload')) $('#wpUploadDescription').val(prefill);
// For improved from:
window.UploadForm_ownwork_author = "Someone";
window.UploadForm_ownwork_date = "{{subst:{{#time:r}}}}";
window.UploadForm_own_language_first = true;
// ... see [[MediaWiki:UploadForm.js/Documentation]]
-- Rillke(q?) 16:45, 19 September 2013 (UTC)
Thanks, but what is window.JSconfig.keys['loadAutoInformationTemplate'] = false; supposed to do? I solved it now with the help of window.UploadForm_autofill = false;, so the changed content of #wpUploadDescription does not get overwritten:
window.UploadForm_autofill = false;
var prefill = ...;
if (!mw.util.getParamValue('wpUploadDescription') && !mw.util.getParamValue('wpForReUpload')) $('#wpUploadDescription').val(prefill);
--Patrick87 (talk) 17:20, 19 September 2013 (UTC)
window.JSconfig.keys['loadAutoInformationTemplate'] = false; prevents that MediaWiki:Upload.js replaces the contents with the information template in the simplest of all forms (the MediaWiki-default-form). If you intend adding configuration variables to the global (window) scope - so scripts can recognize them - please write explicitly window.config_variable = value;. var still works but it may not in future if user scripts are also wrapped in a closure by ResourceLoader. -- Rillke(q?) 10:03, 21 September 2013 (UTC)

Bulk download

Hi everyone. I have a list of about 1.8 million images which I have to download from commons. Is there any simple way to do this which doesn't involve an individual HTTP hit for each image? Many thanks in advance.

This can be done with rsync and developer assistance, if you can provide a list of filenames. A fee may be involved. I'm not sure who the appropriate contact is, but someone else should know. Note that if the files are relatively large, then doing an HTTP hit for each one is actually a relatively efficient way of downloading them, so consider that as well. Dcoetzee (talk) 22:49, 19 September 2013 (UTC)
Hey. Thanks for the answer. I would prefer to take the rsync approach. Does anyone know whom I should contact for that? Also, if I were to consider the individual HTTP hit approach, how many queries per second might I send? -- 06:30, 20 September 2013‎ 89.238.224.126
For the http approach, usual advice is if you make requests in serial, its usually fine (mw:API:Etiquette#Request_limit). I have no idea about the rsync thing (The best I could find is m:Wikimedia_update_feed_service, which is very old and probably no longer accurate). I would recommend sending an email to wikitech-l asking about possible options. Bawolff (talk) 15:10, 21 September 2013 (UTC)
Thank you. :)

Lost my way in the category structure of French hiking trails

This isn't the most pressing problem, I suppose, but I'm a bit puzzled by the category tree for French hiking trails. There is a category Category:GR footpaths, which is a subcat of several cats, among them Category:European long-distance paths. Cat Category:Hiking trails in France is a subcat of Category:Trails in France and Category:Hiking trails by country. Shouldn't these two trees be connected? Is there a difference between "Trails" and "Hiking trails"? Best regards, 82.169.125.23 20:03, 19 September 2013 (UTC)

The above question was from me, didn't notice that the system had logged me off. MartinD (talk) 11:14, 20 September 2013 (UTC)
It is not unlikely that they should be merged. These kinds of forks happen when two people make categories and do not know what the other is doing. The way that you sort this ought is that you yourself think about which name is best, then follow the instructions at Commons:Categories for discussion to make a proposal for a merger of the categories. Present a thoughtful reason why you think they should be merged, and other people will be signaled to give feedback. If consensus is that the categories should be merged then there is then a permanent record of the consensus and anyone would be free to close one category and only use the other. Blue Rasberry (talk) 14:24, 20 September 2013 (UTC)
However, the "Categories for Discussion" process is somewhat broken (people who have many images in the categories on their watchlists, but don't have the category pages themselves on their watchlists, will have no way of knowing that the discussion is even taking place, to start with...). AnonMoos (talk) 02:33, 21 September 2013 (UTC)
Can't blame France, there's a lot of overlap in Category:Hiking trails, Category:Hiking, Category:Walking paths, Category:Trails, Category:Public footpaths and probably others. It's much easier to create such categories than to merge them. Category discussions often stagnate from lack of interest or stalemate with no consensus. ghouston (talk) 04:17, 21 September 2013 (UTC)

September 21

Error generating thumbnails

I'm suddenly getting "Error generating thumbnails" errors for every file when I try to click on the various thumbnail sizes - and for anything for a file I just uploaded except for "full resolution", even when I added it to a page on en.wiki. Is the server having a hiccup? - The Bushranger (talk) 01:46, 21 September 2013 (UTC)

The specific error is "The requested page title was invalid, empty, or an incorrectly linked inter-language or inter-wiki title. It may contain one or more characters that cannot be used in titles."; this is occuring even on this file's thumbnail which I uploaded over a year ago (although its other thumbnails appear to be working...) - The Bushranger (talk) 01:52, 21 September 2013 (UTC)
Downloading the file "medium" sized and uploading it that way made it display on the page...but still no thumbnails, and it doesn't show at all on the en.wiki page I put it on...and uploading a different file also produces "no pictures at all"...what's going on? - The Bushranger (talk) 02:03, 21 September 2013 (UTC) - The Bushranger (talk) 02:00, 21 September 2013 (UTC)
Looks like it's not just me this is happening to; tons of thumbnails appear to be borked, even ones from 2008 in the revision histories at the bottom of pages. - The Bushranger (talk) 02:06, 21 September 2013 (UTC)
I'm getting this bug too. Fry1989 eh? 02:17, 21 September 2013 (UTC)
Looks fine to me. --Zhuyifei1999 (talk) 03:25, 21 September 2013 (UTC)
It's working again now, yeah. - The Bushranger (talk) 03:30, 21 September 2013 (UTC)
For the technically curious, looks like (Based on looking at code changes around these timestamps) it was accidentally caused by gerrit:85393, and fixed about 3 hours later by gerrit:85397. See also the server admin log. Bawolff (talk) 14:53, 21 September 2013 (UTC)

Hello community,
this is to inform you about the (re)start of a discussion in which you might be interested. In short, myself and a few other Wikimedia editors decided to oppose the registration of the community logo as a trademark of the Wikimedia Foundation.

The history of the logo, the intents behind our action and our hopes for the future are described in detail on this page; to keep the discussion in one place, please leave your comments the talk page. (And if you speak a language other than English, perhaps you can translate the page and bring it to the attention of your local Wikimedia community?) I’m looking forward to hearing from you! odder (talk) 10:00, 21 September 2013 (UTC) P.s.: You can check whether the WMF protects the logo of your project by seeing if it's listed as "registered trademark" on wmf:Wikimedia trademarks. - — Preceding unsigned comment added by Nemo bis (talk • contribs)

September 22

Bundesarchiv lack of license evidence

I have nominated File:Bundesarchiv Bild 102-00418, Josephine Baker, Gemälde.jpg for discussion because I saw no evidence for the claimed CC license. If I have missed something, please could someone clarify on the page description. My comment was "How is this freely licensed? The painter died in 1980, therefore derivative works should be copyrighted until 2051. I see no evidence that the Bundesarchiv has the required permissions from the artist's estate to release this under a CC license." -84user (talk) 15:52, 22 September 2013 (UTC)

September 23

Kofi Awoonor

A Google image search show that File:Prof- Kofi Awoonor 2013-09-23 14-01.png was used by many online news services in the last couple of days before being uploaded here today. Can someone check the copyright status, please? Andy Mabbett (talk) 18:23, 23 September 2013 (UTC)

Typical "mobile upload". Therefore deleted. --McZusatz (talk) 19:01, 23 September 2013 (UTC)

Upload wizard broken?

When I launch the upload wizard (either from the Upload file link or via the WLM listings in the uk) I get the page but just a spinning thingy instead of the first page of the wizard. Does anyone else get this. I've tried two browsers. Colin (talk) 18:48, 23 September 2013 (UTC)

Yes, I can reproduce this. I suggest just uploading through Commons:Upload, which is more powerful anyway (except for not allowing batch uploads). darkweasel94 19:16, 23 September 2013 (UTC)
That's not a good enough solution, especially while WLM is running. If people use the old form, they may well not get their uploads counting, and won't benefit from the automatic categorisation/identification of the subject. How do we alert some tech person at WMF that things are broken. Colin (talk) 20:32, 23 September 2013 (UTC)
I've asked on #wikimedia-tech and some people there seem to have seen it. darkweasel94 20:41, 23 September 2013 (UTC)
It seems to be working again. darkweasel94 20:48, 23 September 2013 (UTC)
Good. Thanks. Colin (talk) 21:16, 23 September 2013 (UTC)
Always sad to see that such an important tool is broken just because someone tried adding user action analysis (e.g. clicktracking). That's what "Event logging" is actually used for. -- Rillke(q?) 23:05, 23 September 2013 (UTC)


Looking for a classification

In File:Gare de Valenciennes Faubourg de Paris stairs.jpg there is the ridiculous text, that the passenger has to ask the train conductor for a ticket or validation of the ticket. The platform is long gone, the stairs are closed of and buying a ticket on the train is no longer allowed. Are their any classification for this kind of nonsense text? Smiley.toerist (talk) 22:17, 24 September 2013 (UTC)

If you search for "obsolete signs" you will find some others, including a few categories for obsolete road signs. Maybe category:obsolete signs could be created if nothing else exists. ghouston (talk) 23:31, 24 September 2013 (UTC)
Category:Historic road signs in France ? Okki (talk) 08:35, 25 September 2013 (UTC)
It has nothing to do with the road, but the railways.Smiley.toerist (talk) 11:28, 25 September 2013 (UTC)

I have created Category (obsolete signs), but maybe their should be a supercategory Unusual (under topics) for subjects outside the normal classification. Humor is another example. Smiley.toerist (talk) 11:57, 25 September 2013 (UTC)

September 25

Musopen in FLAC

Now that MediaWiki finally support FLAC, can someone convert and import the previous Musopen records? I found this link, but no idea if it contains all the public domain recordings. And if you want to make new donations for future recordings, there is a kickstarter campaign. Okki (talk) 17:27, 16 September 2013 (UTC)

I'd be happy to help import the old recordings systematically - I'm now becoming a member on the Musopen website which will allow me unlimited downloads of lossless materials, including those made available after the 2012 DVD. (Note: MediaWiki has supported the FLAC codec for many years, just with a .ogg extension, in an OGG wrapper. Many people are not aware of this.) Dcoetzee (talk) 22:51, 19 September 2013 (UTC)
Follow-up: there is a serious issue with importing Musopen files, naming that they accept user-contributed files with no independent verification of their copyright status (see FAQ #5), and in some cases (e.g. [8]) these files do not even list the performer, making the copyright status impossible to verify. Currently these files do not seem to be distinguished from their own commissioned works, as far as I can tell. I followed up with them about this by email. Until such time as this is addressed, no files can be imported from Musopen at all, and all the existing files are called into question. Dcoetzee (talk) 04:45, 20 September 2013 (UTC)
More follow-up: I got email from Musopen basically saying that only performers that actually have "Musopen" in their name represent commissioned works, and so for now these are the only ones I plan to import. Dcoetzee (talk) 17:27, 27 September 2013 (UTC)

September 17

http://www.flickr.com/photos/paparazzistudio/tags/%EA%BD%83%EB%AF%B8%EB%82%A8%EB%9D%BC%EB%A9%B4%EA%B0%80%EA%B2%8C/

Can you help me? I find these mass freely-licensed Korean celebrity images from this freelance photographer and journalist (KIYOUNG KIM, 김기영)'s flickr account, and 6 similar images of the celebrity images at this internet news article. However, I don't think the flickr version images are copies of the news version images and believe that the flickr versions are original, because the news version images has lower quality and lower resolution than flickr versions and lacks EXIF data but flickr versions has that. Does news version images' copyright status affect flickr version images' copyright status? Does the news website's copyright only apply the news version images? (the writer of the news article and the owner of the flickr account seem to be same.) --Puramyun31 (talk) 12:52, 19 September 2013 (UTC)

If there has been no transfer of copyright (which doesn't seem to be the case), and the original author licenses the files under a free license somewhere (which is the case), then they are released under a free license and usable for Commons. There doesn't appear to be a reason to believe that this isn't so. darkweasel94 13:14, 19 September 2013 (UTC)
But if someone claims the photographer has no right to license the images under a free license but the news website does due to his contribution of the news article despite his flickr images are original files, then how can I defend myself? --Puramyun31 (talk) 03:02, 20 September 2013 (UTC)
Do the copyright notices on that news site refer to the author or the news site? I do not understand Korean. darkweasel94 05:17, 20 September 2013 (UTC)
The Copyright notices of each image used in the news article:ⓒ 뉴스타운 (ⓒ newstown, the name of the news site). --Puramyun31 (talk) 07:43, 20 September 2013 (UTC)
  • Symbol keep vote.svg Keep. I think the Flickr licenses are valid and the news outlet is trying a form of w:Copyfraud. The Flickr ones we should assume good faith with because they are the original sizes with exif.--Canoe1967 (talk) 20:17, 20 September 2013 (UTC)

Well, thanks for your answers, Canoe1967, Darkweasel94. But I want to hear more Commons users' opinions, are there any more comments else? --Puramyun31 (talk) 06:53, 22 September 2013 (UTC)

Commons:Village pump/Copyright is the place to ask but they will probably give the same answers.--Canoe1967 (talk) 01:20, 23 September 2013 (UTC)
  • What I find troublesome is the copyright notice at the bottom of Newtown's article: 저작권자 © 뉴스타운 무단전재 및 재배포금지 = Copyright holder © Newstown Unauthorised copying prohibited. Maybe it means that he transferred the copyright to Newstown, at least for the photos published by them.
Did you notice that Newtown lists his e-mail address? At the top of the page, it says "김기영 기자 | evasky@naver.com". Have you tried contacting the photographer?
By the way, I'm having trouble with Newtown's website: when I go to it directly, I can only download 1-2 kB per second (sometimes less), but it loads a lot faster when I use a proxy server to access it. Is anyone else having the same problem? --Stefan4 (talk) 21:41, 23 September 2013 (UTC)
Not all of KIYOUNG KIM's flickr images are similar with the images used in www.newstown.co.kr. If he really transferred his copyright to Newstown, I think the transferred copyright only applies the news site version images which have lower resolution and quality than flickr versions, or the news versions and their respective closely similar flickr version images (same but except resolution and quality and existance of EXIF data). --Puramyun31 (talk) 08:18, 24 September 2013 (UTC)
Free lance photographers rarely transfer copyright. If they are employees of media outlets then the copyright usually originates with their employers. I think in the UK the copyright on personal portraits normally goes to the subject by law, to make it easier for them to give copies away. Media that use free lance photographers would not spend the extra money and bother with complex contracts. When they license usage of the images they are pretty free to do what they want with them without going the extra mile to own them outright for a story that may only run a day or two.--Canoe1967 (talk) 17:25, 26 September 2013 (UTC)

September 20

Should the file descriptions of fictitious flags explain that they are fictitious?

File:Flag of Cantonia.svg was uploaded today, and I highly suspect that the author, en:User:John Cruel, is using it to soapbox on his enwiki userpage. Ignoring this issue for now, given that it isn't the main pressing issue at the moment, let's move on to the main concern: This flag is 100% fictitious. According to the file description, it is the "flag of Cantonia", and his enwiki userpage reads "My hometown is Kong Moon in Cantonia, and now I live in Hong Kong. I can speak Cantonese and Mandarin... Fight for Independence of Cantonia". For those who aren't familiar with the antics of this user on enwiki, he has a history of soapboxing and shoehorning in a fictitious "super secret underground government" (in his own words) throughout the encyclopedia, claiming that there is a rebellious "Government of Cantonia" that exists (google gives zero reliable hits, apart from this user's youtube channel and blog entries), essentially something that he's made up. My concern that someone might come along this file on Commons, read the description, and think that this might be a real flag. Would it be reasonable for the file description to clearly explain that this is a fictitious flag for a non-existent political entity made up by someone and their school friends in their blog? -- 李博杰  | Talk contribs 14:32, 23 September 2013 (UTC)

See also: en:Wikipedia:Articles for deletion/Provisional Government of the Republic of Cantonia, en:Wikipedia:Sockpuppet_investigations/Isuzu1001/Archive. -- 李博杰  | Talk contribs 14:34, 23 September 2013 (UTC)
(Edit conflict) Be bold and add that it's fictitious. File descriptions should be as accurate as possible. darkweasel94 14:35, 23 September 2013 (UTC)

Benlisquare -- We have template {{Fictitious flag}} especially for that purpose. There are hundreds of "Special or fictional flags" on Commons, and they aren't usually deleted just for being special or fictional alone, but they definitely can be deleted for being hoaxing or hatemongering, and it looks like this particular image definitely could be both... AnonMoos (talk) 16:39, 23 September 2013 (UTC)

  • Comment: At this stage I think the talk page at File talk:Flag of Cantonia.svg should be speedy deleted. It's the same kind of soapboxing that has plagued the English Wikipedia for more than a year (forcing the made-up term "Cantonia" with zero google hits as if it were fact, and claiming that there is an ethnic uprising). Also, in case this was missed, File:廣東獨立旗.png is an identical image in PNG format, uploaded by the same person. -- 李博杰  | Talk contribs 00:44, 24 September 2013 (UTC)
  • Comment: I'm not a sockpuppet of Isuzu1001. Don't say that when you have no evidence.John Cruel (talk) 14:11, 25 September 2013 (UTC)
    • I didn't say that you were a sockpuppet. I said that if anyone wanted to know more about what's going on, they can find the full story by following two links. Find me an exact quote where I called you a sockpuppet on the 23rd of September, 2013. -- 李博杰  | Talk contribs 01:11, 26 September 2013 (UTC)

Category tree

Hi all, i'm looking for the full category tree used in commons wikimedia. In other words, starting from http://commons.wikimedia.org/wiki/Category:CommonsRoot, I wish to retrive all subcategories with right indentation.

Is it available for the download?

Regards, Luca

It contains loops, so these must be detected and broken to make a 'descendants' report. There are standard ways of doing this inductively, but I have yet to find time to fiddle with this as a report. -- (talk) 08:18, 25 September 2013 (UTC)
I can place to my bot's sandbox2 (User:YiFeiBot/sandbox2tools.wmflabs.org/yifeibot/sandbox2) if you want to. --Zhuyifei1999 (talk) 12:07, 25 September 2013 (UTC)
I have a dump of this on the Toolserver. It's a long list of parent<->child relations. You can use it to crawl the tree, for example to find loops. The tree is quite big and at the moment very outdated. It should probably be rebuild on labs so we can filter for overcategorization again. Multichill (talk) 18:19, 25 September 2013 (UTC)
Probably not the easiest to work with, but the entire categorylinks table is available for download from http://dumps.wikimedia.org/commonswiki/latest/commonswiki-latest-categorylinks.sql.gz (Warning 2 GB file). It would require a bit of processing to get the sort of thing you want Bawolff (talk) 01:48, 26 September 2013 (UTC)
Isn't it better just query the table from tool labs (commonswiki.labsdb)? --Zhuyifei1999 (talk) 12:24, 26 September 2013 (UTC)

Performance statistics for handling Commons deletion requests

Closed DRs for August 2013

I wanted to do a quick analysis to show whether the deletion request process was well under control by the admin community, or whether there was a backlog problem. The results appear to show a good level of control with a healthy majority of DRs closing precisely on the ideal 7 day target.

  • DRs closed after more than 8 days = 1037 (18%)
  • DRs closed within 8 days = 4495 (81%)
  • DRs closed within 1 day of being created = 207 (3%)

I chose these measures as it makes sense to count to the 8th day as admins will wait until the 7 day target is elapsed, in practice a large number close precisely on the 7th day. It seemed worth counting the early closes too, as this indicates the number of DRs being raised that probably should have been raised as speedy deletions. Detailed numbers below. I'm happy to consider suggestions for other ways of analysing the DR process, however as the process seems well managed I'm not currently planning to turn this into a regular report for every month unless there is a specific need.

Thanks -- (talk) 21:26, 25 September 2013 (UTC)

I presume that DRs that were still open at the end of August don't show up in these figures. They do seem relevant to the question of how well the process is working, however. Would you be able to shed any light on them too? In particular, I wonder:
  1. How many DRs remained open at the end of the month?
  2. How long had they been open for?
I realise that we can't expect recently opened DRs to have been closed, but a similarly detailed answer to (2) would let us break out those that we could expect to be closed (e.g. those opened before halfway through the month). --Avenue (talk) 01:48, 26 September 2013 (UTC)
It would be a harder question to answer, I'll put it on the pot boiler to think about as it seems a new report that pulls in all open DRs today, and check how old they are—in fact a category that is populated automatically with "old" DRs could be quite useful. However the number of "old" DRs should be small, and is effectively answered (in a statistical sense) by measuring only the closed monthly figures on the assumption that every DR does eventually get closed, i.e. if you look in the table at DRs closed after 30 days, that will be the figure for DRs that overran from previous months. I think I'll run June and July and we can see if there are patterns to the 30 day+ figures (the script was inefficient and takes several hours to run through a month).
Note, I would dismiss the 51 DRs counted as more than 99 days old, I think these can be ignored as re-opened DRs and similar, for example Commons:Deletion requests/File:A-S Mutter--Fischer-Dieskau--Abado.jpg has been counted as 241 days old because it was re-opened and quickly closed for a second time. Though my script checks the DR edit history to try to identify which edit was the closure (by matching edit comments to ignore category housekeeping and using the timestamp of the most likely closing edit), it is not super intelligent and only goes back through the final 3 edits. A lack of standardization of DR process/format is an issue when trying to automate a report and building something that compensates for human variation, runs into a law of diminishing returns in terms of programming time. :-) -- (talk) 07:06, 26 September 2013 (UTC)

Closed DRs for July 2013

  • Closed after 8 days = 803 (18%)
  • Closed within 8 days = 3548 (81%)
  • Closed within 1 day = 229 (5%)

Out of interest, I include a dump of links to DRs that were open for more than 10 days. This is a repaste from a terminal window (I don't want to spend time sorting this out) so fancy unicode has been swapped to a hex regex form, but you can probably work these out if you want to track the link down.

-- (talk) 09:42, 26 September 2013 (UTC)

Closed DRs for June 2013

  • Closed after more than 8 = days 937 (22%)
  • Closed within 8 days = 3270 (77%)
  • Closed within 1 day = 255 (6%)

For June I increased the limit to a minimum of 21 days as there were significantly more than appeared in July and August (a good thing as this shows increasingly better control). -- (talk) 22:13, 26 September 2013 (UTC)

September 26

Cleaning out Category:Royal coat of arms of the Netherlands

A list of versions of the Royal Coat of Arms, 1907 version, looking for some feedback

  1. File:Nederlands Rijkswapen van 1907.jpg is the official weapon. I've added the identity of the creator.
  2. File:Nederlands Rijkswapen van (1907).jpg Choppy retouch of the above.
  3. File:NL - COA.png and File:Nl-arms.gif are identical copies of eachother. I've tagged the GIF version as a duplicate, because it had the larger filsesize. The file itself however is copied from the Monarchy's website (previous design I assume?), and might be a copyright violation.
  4. File:Coat of arms of the Netherlands.png was overwritten in 2009 by a rather distorted file, seemingly from 1982 (thus not in the PD and a copyright violation). The original was a low resolution file from Vector Images, however.
  5. File:Coat of arms of the Netherlands.svg (derivative File:Coat of arms of Dutch East Indies.svg) was overwritten by a seemingly sloppy vectorisation of the above file. The original was a peculiar but proper SVG file.
  6. File:Grandes armes Pays-Bas.svg OK
  7. File:Royal Coat of Arms of the Netherlands.svg Based on the first file. OK.
  8. File:Greater Coat of arms of the Netherlands.svg Completely redudundant because of the above. As the 'creator' (though I did very little), I'd rather see it deleted.
  9. File:Great Coat of Arms of the Netherlands.svg OK, though incomplete
  10. File:Coat of Arms of the Netherlands.svg and File:Nederland wapen 2.svg (derivative File:Nederland wapen.svg) are both copyright violations of the work of Piets Bulsma, per Commons:Deletion requests/File:Coat of Arms of the Netherlands.png. A shame, because the second one was vectorised really well. The original revisions of #4 are copyright violations as well.

So, some questions. Is there a use for #2, if so, could someone redo it? #5 probably needs revision deletions, a copyvio tag might risk losing previous revisions. Is there any basis on which I can get #8 deleted? I hope this sounds reasonable, and I hope I'm not forgetting policies. Lemmens, Tom (talk) 10:37, 26 September 2013 (UTC)

Making False Statements

An August 31, 2001 a post by Filipino postage stamp collage artist Nux Suzara claims "she is the first and only arist that creates collages made entirely of both rare and common postage". Either a bold lie or the artist lives in a cave. Postage stamp collage art has been created by many artists and hobbyist since the 1850's when glue backed perforated stamps were first printed. See December 2012 "American Philatelist" feature article "Postage Stamp Collage Art" by Russ Hahn. A history of the art form since 1850. stampsartist@aol.com — Preceding unsigned comment added by 65.8.179.126 (talk • contribs)

  • I'm missing the relevance of this to Wikimedia Commons, can you explain? - Jmabel ! talk 19:40, 27 September 2013 (UTC)
File:Unique Postage Stamps Collage by Filipino artist Nux Suzara.jpg has the statement.--Canoe1967 (talk) 19:47, 27 September 2013 (UTC)
Well, as the image was uploaded by person obviously different from the alleged creator, it needs OTRS-permission. --Túrelio (talk) 21:01, 27 September 2013 (UTC)

FOP Thailand

http://www.panoramio.com/photo/84303994 seems like a nice building. Commons:FOP#Thailand says we can host images of it. Does anyone know a user in Thailand that has a camera?--Canoe1967 (talk) 23:52, 27 September 2013 (UTC)

I don't know, but I suggest trying at Commons:สภากาแฟ (Commons Village Pump) or th:วิกิพีเดีย:สถานทูตวิกิมีเดีย (Thai wikipedia embassy).--Pere prlpz (talk) 11:20, 28 September 2013 (UTC)

September 28

Statistics on galleries

Do we have any statistics on gallery namespace? Of the 110k pages, how many have just one or two pictures? How many have been edit in the last 3 months by non-bots/hotcat? --  Docu  at 13:10, 28 September 2013 (UTC)

34627 main namespace non-redirect pages have exactly 1 or 2 images on them (but not counting pages with 0 images. For reference there are 1019 such pages with 0 images). There are 7574 non-redirect gallery pages that have a non-hot cat edit to them since June 28 (Note, I'm not aware of a way of filtering out bot edits, so this includes bot edits made in that period). At the time of when I did the query, there was 110031 pages in the gallery namespace that weren't redirects (This would include the main page) Bawolff (talk) 14:12, 28 September 2013 (UTC)
Thanks. BTW, the 7574 edits would include bot edits, I suppose. --  Docu  at 14:23, 28 September 2013 (UTC)
My preceding comment seems to be simultaneous to an update to the previous comment. A way to filter the bot edits could be to check the user group for a bot (or the username for "*bot*"). --  Docu  at 14:34, 28 September 2013 (UTC)
Filtering out people who are currently in the "bot" group, brought the number down to 7428. If on top of that I remove people with names ending in "Bot" or "bot", its 7427. Bawolff (talk) 16:07, 28 September 2013 (UTC)
Some galleries are bound to be static. E.g. Paintings of deceased artists will be filled with all paintings at some point. --Hedwig in Washington (mail?) 20:23, 28 September 2013 (UTC)
Hm, it seems that I forgot about "CommonsDelinker". It doesn't have "bot" in its name, but it did 1500 automated edits in the last three month. The high part of delinker edits in Gallery namespace suggests that these are primarily maintained by CommonsDelinker. --  Docu  at 08:40, 29 September 2013 (UTC)

Category:Coats of arms of users

While looking through the disputed coats of arms category, I stumbled upon this file: File:Rey Fiscal Armas Personales.svg. It is tagged as disputed because of the following reason: "very dubbious (and pretentious) use of a royal crown. Unless this user is the hidden identity of a king, it has nothing to do here." There are a lot of user coats of arms that assume elements they have no rights to, for example rank crowns, supporters, and even orders. Now, if this tag is to remain, a lot of files should thus be tagged. Should we have a policy on this? Should we disallow unrealistic assumed arms, or differentiate them from ordinary assumed burgher arms? Should there be a limit on how many personal coats of arms one can keep? And what if they aren't used at all?

A related issue is categorization. Should these coats of arms be categorized in other categories than Category:Coats of arms of users? Some of them are, some of them aren't. Lemmens, Tom (talk) 13:22, 28 September 2013 (UTC)

If those images are non notable coats of arms without any educational value, they are out of scope unless used in userpages. Anyway, even if used, it would be good to mark them as fictional, and that tag seems a tentative way to do it.--Pere prlpz (talk) 14:05, 28 September 2013 (UTC)
Well, here in Belgium one can assume arms, even without registering them. The situation differs in each country, but self-assumed arms are not per se fictional. It only becomes problematic when one uses elements and ornaments that one is not entitled to, e.g. crowns. But in any case, I'll mock up a template. Lemmens, Tom (talk) 14:24, 28 September 2013 (UTC)
Right now, "Category:Coats of arms of users" is a subcategory of "Category:Special or fictional coats of arms", so all the images in the category are already indirectly labeled. I don't think it makes too much sense to try to police exuberant imaginative use of vague generalized noble/royal symbolism, which for some people seems to be a large part of the fun in playing with heraldry to come up with a personalized symbol (just as people who try to write their names in Egyptian hieroglyphics often put them in a royal cartouche). It might make sense to try to single out user page images which are not actually used on a user page (though this has only rather sporadically been done), or to police user coats of arms which closely resemble the arms of a specific real-life kingdom or noble family or heraldic order... AnonMoos (talk) 15:53, 28 September 2013 (UTC)
I was thinking of a template like this. It would replace the user image tag (correspondingly, "coats of arms of users" would become a subcategory of user images). The tag supports a parameter which seperates "serious" from "non-serious" heraldry. It would give a clearer warning (since the user image tag doesn't stand out), without simply dismissing the coat of arms with a giant fictional sign. Lemmens, Tom (talk) 16:49, 28 September 2013 (UTC)
In the first template, why not say "which the user LIKELY has no traditional rights to", since most of the time we can't know for sure based just on the username. And why does the second template start out by mentioning problems which don't occur in the image? Anyway, I know you consider fantasy noble/royal heraldic elements to be in poor taste (and I didn't use them in my own self-assumed coat of arms, but I'm not sure there's much point in attempting to crack down on them unless they closely resemble real noble/royal symbols... -- AnonMoos (talk) 18:03, 28 September 2013 (UTC)
I was really struggling with the wording (there's still somewhat of a language barrier), it can be changed. Concerning the second template, I thought it would provide some context. Of course, a good template description would do that job as well. I do indeed consider it both poor taste and something that can possibly be very confusing. A template like the one I constructed could alleviate at least the last problem. Really cracking down on them isn't my preferred course of action, though I did want to raise to issue. I believe my template would be a decent solution, but I can really use feedback on the wording. Lemmens, Tom (talk) 18:18, 28 September 2013 (UTC)

Lombards Museum?

Greetings, friends,

I'm looking for some info about "Lombards Museum". Wiki Commons has over a hundred beautiful photos from this "Lombards Museum" - mostly pre-Columbian art - such as this,

https://commons.wikimedia.org/wiki/File:Lombards_Museum_038.JPG

But where is it? I've looked already for a while, but can't find anything. (Small museum in Illinois doesn't count.)

Can someone help please? Y-barton (talk) 14:41, 28 September 2013 (UTC)

Working on the principle that the uploader is a French wikipedian (from the users global contributions), one possibility is fr:Musée Lombart in France which is described as the Lombard Museum here. However before booking flights I think its best to check with the uploader, as although the French wiki article mentions an ethnological collection, it does not say anything specific about pre-Columbian art.--KTo288 (talk) 15:32, 28 September 2013 (UTC)
I guess that it's not that museum. We do need better descriptions for the Special:Search/intitle:Lombards Museum files. --  Docu  at 15:53, 28 September 2013 (UTC)
Asked uploader of the files at User_talk:Paname-IV#Lombards_Museum.--KTo288 (talk) 20:17, 28 September 2013 (UTC)
Thank you for your replies. Situation somewhat strange... A nice museum that's lost somewhere out there in the space-time continuum? :)
I guess this might be some small private museum somewhere? I went through the files; there seem to be 41 files altogether uploaded to Commons, all of them but one pre-Columbian art. Museum might be somewhere in S America... Maybe some semi-private museum belonging to a bank (lombard)? Best wishes, Y-barton (talk) 19:11, 29 September 2013 (UTC)
I'm guessing private collector with own museum.--KTo288 (talk) 20:03, 29 September 2013 (UTC)

September 29

Non-copyright free photos

Can anyone help me to delete File:GLN国际互动视频中心2.jpg, File:GLN国际互动视频中心3.jpg and File:GLN国际互动视频中心1.jpg? I've found that they are come from there. I am a Wikipedian and I don't know how to submit them to CSD. Can anyone help to delete them?--Carrotkit (talk) 05:29, 29 September 2013 (UTC)

Tag the page with a {{speedy}}. See COM:D --Glaisher (talk) 06:41, 29 September 2013 (UTC)

Collapsible gallery?

Hi, is there anyway to make a gallery with the <gallery> tag collapsible? // WikiPhoenix [Talk] 14:13, 29 September 2013 (UTC)

You can make anything collapsible with {{cot}} and {{cob}}. darkweasel94 15:21, 29 September 2013 (UTC)
If you prefer custom styling, checkout MediaWiki talk:Gadget-CollapsibleTemplates.js -- Rillke(q?) 16:57, 29 September 2013 (UTC)
Thanks! Face-smile.svg // WikiPhoenix [Talk] 11:04, 30 September 2013 (UTC)

Input in essay Commons:Language guide

A Wikipedian's been helping me out with the essay Commons:Language guide. He suggested that there be a list by country, so I added that in addition to the list by language. Editors are welcome to make changes or leave feedback WhisperToMe (talk) 09:15, 30 September 2013 (UTC)