Last modified on 1 February 2015, at 19:21

Commons:Village pump


  Welcome to Wikimedia Commons   Community Portal   Help Desk
Upload help
  Village Pump
copyright • proposals
  Administrators' noticeboard
vandalism • user problems • blocks and protections
 
↓ Skip to table of contents ↓       ↓ Skip to discussions ↓       ↓ Skip to the last discussion ↓
This project page in other languages:

বাংলা | Alemannisch | العربية | asturianu | авар | Boarisch | bosanski | български | català | čeština | dansk | Deutsch | Ελληνικά | English | Esperanto | español | فارسی | français | galego | עברית | hrvatski | magyar | íslenska | italiano | 日本語 |  | 한국어 | Lëtzebuergesch | македонски | मराठी | Nederlands | norsk bokmål | occitan | polski | português | русский | slovenčina | slovenščina | српски / srpski | suomi | svenska | ไทย | Türkçe | 中文(简体)‎ | 中文(繁體)‎ | Zazaki | українська | +/−

Welcome to the Village pump

This Wikimedia Commons page is used for discussions of the operations, technical issues, and policies of Wikimedia Commons. For old discussions, see the Archive. Recent sections with no replies for 3 days may be archived.

Please note


  1. If you want to ask why unfree/non-commercial material is not allowed at Wikimedia Commons or if you want to suggest that allowing it would be a good thing please do not comment here. It is a waste of your time. One of Wikimedia Commons' basic principles is: "Only free content is allowed." This is just a basic rule of the place, as inherent as the NPOV requirement on all Wikipedias.
  2. Have you read the FAQ?
  3. For changing the name of a file see Commons:File renaming.
  4. Any answers you receive here are not legal advice and the responder cannot be held liable for them. If you have legal questions, we can try to help but our answers cannot replace those of a qualified professional (i.e. a lawyer).
  5. Your question will be answered here; please check back regularly. Please do not leave your email address or other contact information, as this page is widely visible across the internet and you are liable to receive spam.

Purposes which do not meet the scope of this page


Search archives


 


The last town pump to be in use in Saint Helier, Jersey, until early 20th century [add]
Centralized discussion
See also: Village pump/Proposals • Archive

Template: View • Discuss • Edit • Watch




OldiesEdit

Christmas crosswordEdit

EFF Crossword Puzzle 2014: The Year in Copyright News

The Electronic Frontier Foundation has published this (copyright related) crossword which you can play online at http://thedod.github.io/eff-crossword-2014/. It makes a nice break from feeding yourself with Christmas treats. Happy holidays everyone. Smile fasdfdsfoiueire.svg (talk)   16:57, 25 December 2014‎ (UTC)

Something wrong with the UploadWizard?Edit

Here's how the problem is showing up: any file names are not accepted, no alternative given

At Commons:Upload help today, three users, among them myself, have reported being unable to upload any files with the wizard. The error messages are "file name is not allowed", "this type of file is banned", and "badupload_file" (that's what I got). Uploading with the "old" upload form worked fine, at least for me. One of the other two users sent me (per e-mail) one of the files that failed for him with "file name is not allowed"; for me, it failed with "badupload_file". It uploaded fine without the wizard (I have since deleted it). Anybody else have problems with the wizard? --Rosenzweig τ 19:00, 15 January 2015 (UTC)

+ another report from today at Commons:Upload Wizard feedback. --Rosenzweig τ 19:02, 15 January 2015 (UTC)
Yes, they do event tracking; if an adblocker blocks that, the UploadWizard crashes. See the linked bug report. Lupo 20:06, 15 January 2015 (UTC)
Here's a list of some known bugs. –Be..anyone (talk) 20:08, 15 January 2015 (UTC)
This is not about the wizard crashing. It displays an error, but you can then remove the file from the upload queue and continue. Disabling Adblock changed nothing. --Rosenzweig τ 22:48, 15 January 2015 (UTC)

I cannot upload any more. I can with the old upload (without the wizard), but this is cumbersome.. I do not have an adblocker installed, but maybe noscript might be the reason. It is a shame, that nobody seems to care to fix the problem. Not even saying thanks for reporting the problem. How many guys & gals out there will be rejected by failed uploads? regards --Herzi Pinki (talk) 12:04, 16 January 2015 (UTC)

Well, if this is related to extensions blocking certain scripts, then there is a patch waiting in phab:T86680... Please test with noscript disabled if the problem still appears. Thanks! --AKlapper (WMF) (talk) 14:38, 16 January 2015 (UTC)
  • Clearing the cache did not help, same behavior
  • some statistics (I've counted the uploads through UW per day / 24 hours), median is 5005
date (00:00 - 24:00 UTC) catscan call #new files (catscan result) % of max.
15.1. [1] 3986
68.3%
14.1. [2] 5334
91.4%
13.1. [3] 5343
91.5%
12.1. [4] 5005
85.7%
11.1. [5] 5837
100%
10.1. [6] 4812
82.4%
9.1. [7] 4422
75.8%
8.1. [8] 4708
80.7%
7.1. [9] 5114
87.6%

so I suspect that yesterday (15.1.) about 1000 uploads with the UW failed. Try no nocscript next. --Herzi Pinki (talk) 19:23, 16 January 2015 (UTC)

I tried again with noscript disabled, FF restarted, same effect (UW fails) --Herzi Pinki (talk) 20:14, 16 January 2015 (UTC)

Same issue here without any adblocks. This is definitely not the adblock issue. Schmelzle (talk) 13:08, 17 January 2015 (UTC)

I tried it again today, with a different computer located in another town and with another Internet provider. But now, the very same files I couldn't upload two days ago were accepted by the upload wizard without a problem. Firefox 34.0.5 (Linux), Adblock is not disabled, Noscript is not running, same on both computers. I'll be at the other computer again in a few days and will try there again if it is still helpful. Could it have something to do with the different Internet providers, or doesn't that make any difference at all? --Rosenzweig τ 14:20, 17 January 2015 (UTC)

And why am I suddenly denied access to https://phabricator.wikimedia.org/T87062? --Rosenzweig τ 14:48, 17 January 2015 (UTC)
Good question forwarded to mw:Project:Support_desk#Phabricator_down.3F_52984. –Be..anyone (talk) 15:01, 17 January 2015 (UTC)
(EC) Don't know. Possibly because someone posted the full contents of a POST request trying to upload an image. That POST-request contained tokens, cookie settings, and possibly even more sensitive data. BTW, the crucial thing to note in that POST request was that the request already was sent with an empty filename. (I saw it before it was protected.) Lupo 15:06, 17 January 2015 (UTC)
Apparently something like that was the reason, yes. But the result is that apparently no one (except perhaps the creator of the task) can access it anymore. So essentially it is useless. --Rosenzweig τ 18:38, 17 January 2015 (UTC)
Access to the task will be restored as soon as one of the Phabricator admins is back to work. Unlikely that a fix would pass before this happens anyway. -- Rillke(q?) 18:41, 17 January 2015 (UTC)
Generally, the best place for questions about Phabricator itself is mw:Phabricator/Help. In this specific case, someone (rightfully) restricted access to the task because private information was posted. For those interested, mw:Phabricator/Security covers the security policies and behavior of Phabricator. --AKlapper (WMF) (talk) 16:16, 18 January 2015 (UTC)

I'm back at my original computer again (see above), and again, the wizard gives me the "badupload_file" error message when trying to upload a file. As reported, the exact same file is accepted without any problems by the upload wizard when using my other computer with a different Internet provider. I can try to use the original computer in the other location with the other Internet provider to see what happens, but it will be several days before I'm there again. --Rosenzweig τ 22:41, 18 January 2015 (UTC)

I only have a single Internet provider, but as it is a big one, others should have similar troubles if it were for the provider. For me all files seem to fail (tried about 30 different). Acc. to Special:Version UW is of Jan. 7th (on the 12th everything was working well with the UW), but MW version changed (as I remember on the 14th and again on the 16th (current version 1.25wmf14 (d8660bb))). So if it is in the SW (and not my local config only), MW core could be the reason in combination with my local settings? --Herzi Pinki (talk) 00:10, 19 January 2015 (UTC)
some days gone. My assumption was that the WMF wants to be responsible for the software and configuration and … and the community should be focused more on content. Beside new fancy super features this will include the proper maintenance of the software / configuration and quick fixes in cases like this. UW is not a minor feature used by only some specialists. I did publish all information I could think of to be helpful (and even more). But nobody seems to care, the bug is not even assigned to someone of the SW team. Nor did anybody of the WMF staff try to propose a workaround (using Safari is not a general workaround!). Thanks for support from all the colleagues here and for providing additional diagnostic data. --Herzi Pinki (talk) 22:04, 23 January 2015 (UTC)

This error continues. The wizard reports an ogg audio file as "banned" but works fine when the file is converted to flac. It appears that recent changes need to be reverted until someone figures out the error. Khamar (talk) 16:52, 27 January 2015 (UTC)

Uncategorized pictures increasingly growingEdit

Categorization stats from 2015/01/23

Is this critical? Our backlog goes back to September 2012.--Kopiersperre (talk)

Yes, it is. No category is not much better than no license. Or JPGs of male body parts below 1MB. –Be..anyone (talk) 16:30, 23 January 2015 (UTC)

Not yet a million? That means it's far from a majority. I am pleasantly surprised. But of course an uncategorized pictures is usually worth nothing. Well, the majority of categorized ones are also useless, but use is very rare for uncatted ones. Perhaps the Upload Wizard should be adjusted to make this matter easier and more urgent. As for easier, a tiny improvement would be to make categories insensitive to case. Bigger improvement would be a tree-walking cat picker. Jim.henderson (talk) 16:41, 23 January 2015 (UTC)

I personally would enjoy a tree-walking cat picker much more than a media-viewer. Non-english speaking people sometimes categorize their uploads quite well, but sadly in their mother language. There should be something like the Template:Category redirect, for example a Template:Internationalized category.--Kopiersperre (talk) 18:11, 23 January 2015 (UTC)
There are so many things that could be improved with categories:
  • Category translations (built into MediaWiki, not with templates etc.)
  • Custom link titles in parent categories. (I.e. if I put the following in a theoretical category German great stuff: [[Category:Stuff in Germany|Great stuff|Great stuff]] the link to it will be displayed as "Great stuff" instead of "German great stuff" in that category. The "German" is irrelevant in that case and just makes it harder to recognize stuff.)
  • Category tree views and recursive views, including powerful filter mechanisms.
  • A decent category/files in category search.
But hey, we have a great Media Viewer now. --Sebari (talk) 19:24, 23 January 2015 (UTC)
How do I active or acces Media Viewer?Smiley.toerist (talk) 13:47, 28 January 2015 (UTC)
I agree uncategorized images are a problem. I disagree that uncategorized images are usually worth nothing- high volume automatic uploads (e.g. from Public Domain repositories, Flickr, etc.) often have valuable images that are only in hidden cats like "uploads by user". I previously brought up the issue of organizing a concerted effort to reduce the backlog, ideally with humans assisted by bots rather than the other way around, and with priority given to high quality sources or subcategories. see Commons:Village_pump/Archive/2014/11#Group_effort_to_clear_out_uncategorized_media.3F -Animalparty (talk) 19:02, 23 January 2015 (UTC)

The raw numbers do not appear statistically meaningful as they are not in relation to the sample space. Could someone rework these so that they show a proportion of total images on Commons on the dates they are measured? If the ratio "Uncategorized images / total number of images" is decreasing, then we would seem to be on top of things. I.e. if Commons had 90 images uncategorized, but only had 100 images in total (90%), that is a far worse problem than having 900 images uncategorized out of a total of 10,000 (9%). -- (talk) 19:29, 23 January 2015 (UTC)

@: Please give me the total number of images on Commons over time as CSV, then I can do.--Kopiersperre (talk) 20:08, 23 January 2015 (UTC)
Pictogram voting comment.svg Comment A couple months ago I've started to add categories to files in Category:Media needing category review by date. I realized than a large number of pictures of the 446,891 files in Category:Images from the Geograph British Isles project needing category review (I can't say "all" because I couldn't check) already had categories added with bot; however, "{{Check categories}}" remains. I told (because his bot add these cats) and he told me he has family problems and will check later, thing I respect and why I didn't continue pushing again about this point. Maybe run a bot to fix this problem could help. Thanks. --Ganímedes (talk) 20:32, 23 January 2015 (UTC)
I think it's correct to leave Check categories on bot-catgorized work, since the bot generally just adds a rough location category. However a lot of these geograph images are nondescript fields or hills and realistically will never be used for anything, so it's not really a big loss if they are uncategorised. I have categorised quite a few myself, since categorisation seems to be mildly addictive to me. --ghouston (talk) 21:38, 23 January 2015 (UTC)
It is not surprising as the number of uploads continue to grow, but the number of people doing maintenance does not seem to. That's just a feeling, and a real statistics would be useful. We need to expand the number of "permanent" people, not just people who upload and then go back to their home project. What about a recruitment campaign across Wikimedia with some precise and detailed objectives: 1. We need more people doing categorization, 2. We need more people doing license review, 3. We need more admins. Regards, Yann (talk) 20:35, 23 January 2015 (UTC)

@Kopiersperre:, okay here it is. Please have a go at factoring it in. Converting to CSV is easy in most text editors, just replace the pipe marks. January is incomplete of course. Smile fasdfdsfoiueire.svg -- (talk) 21:09, 23 January 2015 (UTC)

@: This is uploads/month and not the absolute amount. The uncategorized files are however a cumulative amount. I think this can't be stitched together.--Kopiersperre (talk) 21:22, 23 January 2015 (UTC)

Rather than the absolute data of uploads per month, it is looking at when existing images hosted by Commons were uploaded. As there cannot be an uncategorised image once it has been overwritten or deleted, this seems a good approximation as there seems little point in cumulatively counting stuff which no longer exists. If anything, if the uncategorized image data is counting images which are now deleted, this is probably more misleading, as it counts now deleted copyvios and out of scope material. I am not sure you will get much better data than this. -- (talk) 21:34, 23 January 2015 (UTC)
I've just failed. It's quite difficult to plot two timedata plots with different timeschemes (%Y%m%d%H%M / %Y%m) into one plot with gnuplot.--Kopiersperre (talk) 22:38, 23 January 2015 (UTC)
I have experience with combining data of different time scales. I use a spreadsheet to use a linear interpolation between values (other interpolations can also be used) to obtain values at the time needed (first convert the date to some sort of digital date format - either the date value used by your computer or some other convention - 1/31/2008 could either be 39478 or 2008.083333 for example). But in the data sets that you have, no interpolation is needed. You can just plot the monthly ratio, without worrying about any of the daily data that you have for one of the data sets, or the fact that the data is a few minutes apart from each other. Delphi234 (talk) 06:57, 25 January 2015 (UTC)

Want to share some thoughts on the problem (not on proper statistics).

  • In general there is an attitude on images nowadays and everywhere, where storage is free and internet access is flat, to upload anything you have and do no proper selection before. Let the viewer choose. This also includes images with wrong orientation and low quality. This is true especially for platforms like Flickr. But I also get a bunch of some thousands of images of my daughter's summer camp in that style, or from some outside school projects, etc. Commons seems to follow the same attitude when uploading images or importing them from some external source (like flickr), at least in some cases. This might be a tribute to the sheer number expected during competitions like WLM, a tribute to the own ego, or a tribute to time constraints that make it easier to upload a million images than to select the 100 usable. In the end I consider this to be one of the root causes for the huge number of uncategorized or badly categorized files or files with bad description. Many of them causing effort without generating much usefulness.
    • As an example consider images of political parties found on flickr, e.g. Category:Photos from European People's Party (I worked a lot with those). Once uploaded from Flickr, these images are not uncategorized, but still missing information about the individuals depicted on the images (semantically speaking, those are uncategorized although not marked as such). I doubt, that we need that many images of e.g. Angela Merkel (~1500 at the moment) and that we can use more than a few in articles at all. These images are already categorized per year and someone started to subcategorize per month. In the end after a lot of work this will hide all the images deep down in some category tree. Commons is not about storing every cramp. Make the selection before and upload only a selection. Can we make an appeal to all the power uploaders / bots to select first and upload later? We do not have to have everything here in copy.
  • There are a lot of photo-projects supported by various chapters. The precondition for getting money for photo tours is to upload the images with the proper Supported by Wikimedia … template. Proper categorization isn't a precondition. If someone refuses to categorize, categorization will not happen and will lead to uncategorized. Can we have a statistic of uncategorized per uploader? And start to handle the problem from the prominent end. I would change the rules for such photo projects with respect to categorization. At the moment there are 4100 images supported, but not categorized. These are - on the average - higher value images than normal uploads.
  • Some files are already categorized, but the {{Uncategorized}} wasn't removed. Cat-a-lot has an option to do so automatically, but this is not the default. Can we change the default to removing the template? Bots could remove the uncategorized.
  • I always found it difficult to look through thousands of uncategorized images to find the few I was able to categorize. The new search makes it possible to find images with keywords in name / description, allow a narrower scope and thus a higher percentage of images that can be categorized by an individual. E.g. this search will find most of the uncategorized images from Austria in 2014. If someone is working on Austria, this pre-selection is rather useful. Would it be an idea to offer searches like this to people / portals working on corresponding topics in the WPs?
  • Some images are categorized with a non-existing category. Since some weeks this will also be recognized as uncategorized. But it is a different case, creation of missing categories will solve more than one file.
  • Categorization is difficult. Not every category will fulfil the requirement. I found images in Categories Men looking right and Statues of sitting men, but without even the smallest hint where on earth this image was taken. Such categories are not especially useful to fulfil the requirement of categorization, better to leave the image uncategorized.
  • instead of putting effort to categorization, it should be easier to put low quality images to speedy deletion. --Herzi Pinki (talk) 23:55, 23 January 2015 (UTC)
    • I would suggest adding some additional queries to the upload process such as "is this a picture of a person/people", "is this a picture of a building", etc., with a mandatory yes/no input. Pictures with a "yes" could then at least be subcategorized into an "Uncategorized images of people" or "Uncategorized images of buildings" category, so there would be some beginning of an identification process. This would also instantly flag uploads raising common intellectual property issues beyond those inherent in who took the picture. BD2412 T 11:09, 24 January 2015 (UTC)
      • This last strikes me as a terrible suggestion. I probably upload as many photos as any but a handful of contributors here. If I started having to answer a bunch of mandatory questions to do so, you would probably see me stop entirely. - Jmabel ! talk 17:14, 24 January 2015 (UTC)
        • It is a suggestion directed to the problem at hand. Do you add categories to your images during the uploading process? BD2412 T 17:33, 24 January 2015 (UTC)
          • I think asking questions to try to guess a category would not help, but I have noticed that some wikis are more insistent than others about filling in the edit summary. Perhaps just going to a screen notifying the uploader that the category is blank and providing information about how to find an appropriate category - but in a manner that does not make you start over. I agree that just putting everything into an "image" or "photograph" category is not useful. Delphi234 (talk) 08:00, 25 January 2015 (UTC)

Question to the charts above: What happened in the end of 2012 to add just another 200000 images to uncategorized? This could not have been caused by uploads only. --Herzi Pinki (talk) 07:36, 24 January 2015 (UTC)

I guess in the end of 2012 British geophotios were donated, which were uploaded but not categorized.--Ymblanter (talk) 11:13, 24 January 2015 (UTC)
  • Pictogram voting comment.svg Comment Also some files with hidden cats are showing as "needing categories". Probably most of them should continue hidden, but some as this perfectly could be the main cat, since no other valid cat was added. Regards. --Ganímedes (talk) 20:36, 24 January 2015 (UTC)
  • Pictogram voting comment.svg Comment Working a lot with images in the categories “Media needing categories as of ......” I suggest that it would help when in the upload form is mentioned near the description something like “please mention where the photo was taken (country/city ...). This would help in categorizing. Descriptions as “Wow” or “Here we had dinner” or only “castle” are not very helpful. Wouter (talk) 10:20, 25 January 2015 (UTC)
    This is a very good suggestion which corresponds to my own experience.--Ymblanter (talk) 12:28, 25 January 2015 (UTC)

January 24Edit

Pngs and jpgs of the same imagesEdit

png
jpg

Do we really need to retain both a low resolution jpg as well as a high quality png file? PNGs display just fine in article space so I see no reason to keep both. Do other know or feel differently? In this category I happened to noticed many such duplicates. While server space may no longer be at a premium there seems little point in just using the space because it is available. Ww2censor (talk) 11:16, 24 January 2015 (UTC)

Also discussed at COM:Форум#Удаление файла, where I referred to Help:PNG. So, does that policy still hold true? YLSS (talk) 11:35, 24 January 2015 (UTC)
Deleting files does not save server space, as deleted files as kept hidden. However I agree that keeping lo quality JPGs when there is a PNG version is useless. It clutters the search at least. Regards, Yann (talk) 11:52, 24 January 2015 (UTC)
So should one bother to put such jpgs up for deletion as duplicates or not? Ww2censor (talk) 12:19, 24 January 2015 (UTC)
Yes, go ahead. Please check that the files are not used anywhere before creating a DR. Actually replacing small quality files by better ones is what takes the most time, the rest of the maintenance work is fast. BTW all PNG files had a double extension in this category. I renamed them. Regards, Yann (talk) 12:27, 24 January 2015 (UTC)
I think you shouldn't be that hasty. First of all, this concerns not only those Soviet stamps, but also 36,426 files in Category:PNGs with JPEG versions. So if there's any change of policies, such pages as Help:PNG and COM:MFC#PNG photos that require a JPEG version should be updated first. But then, are you sure that those policies do not hold any more? Compare the two pics to the right. At full size (1,902 × 2,157) they are the same; as thumbnail, jpg is better to my eyes. YLSS (talk) 14:24, 24 January 2015 (UTC)
It depends very much what is the content, and how the images were made. For scans of non-photographic documents (maps, drawings, books, engravings, etc.), PNG is usually better than JPEG, if the PNG is made from the same original source. But there may be exceptions. Obviously if the PNG was made from the JPEG, it can't be better. For pictures, JPEG is usually better. TIFF is another issue. It should be used as a archive format only, even if MediaWiki is able to show a thumbnail version. Regards, Yann (talk) 19:00, 24 January 2015 (UTC)

┌─────────────────────────────────┘

We have many thousands of pictures like File:New York City Boys organize tin club. These patriotic young men not only collected but also processed more than half... - NARA - 196374.tif in both TIFF and JPG format and links between the two. They clutter many topical categories but we have started decatting the TIFF version and leaving just the version link and a huge TIFF subcategory. Some word searches will still be cluttered unless we use a "-tif" to mask them. Good plan? Jim.henderson (talk) 13:21, 24 January 2015 (UTC)

Yep, JPEG is also better for me on thumbnail, same issue of TIFF, TIFF can hold more information, making easier to edit, but for the web, it sucks... for me keep both, xoxo -- RTA 18:27, 24 January 2015 (UTC)
I don't really see the issue here. Why are the two choices only a low resolution JPG and a high resolution PNG? Surely we can have a high resolution JPG rather than the low resolution JPG. PNGs might be slightly superior for archival use, but they are not at all practical for use on Wiki projects, because the thumbnail rendering is inferior and the file size is unnecessarily large for downloading. Personally, if I was faced with a choice of just one file, I would much prefer a JPG with very low compression than a PNG. The quality difference is very minimal (when low compression is used), but the file size difference is large. Diliff (talk) 19:01, 24 January 2015 (UTC)
I think JPEG is better also. I think that JPEG of 600 dpi is the standard for images of stamps. Let users decide, what format they will take for the needs, maybe every time different. And search strongly complicates not existence one more format, and existence of other images. Let's delete all other images, it will be easy to look for... --Matsievsky (talk) 19:12, 24 January 2015 (UTC)
Why should we delete all images? As Yann already said, this will not saves us space... and we will have a time consumption and a lot warnings, and troubles, furthermore, analysing some cases will be necessary, i.e., I remember to upload a image with a white background in JPEG and a similar one with a transparent background in PNG, for different reasons. JPEG stills losing quality in editions, and modify a image is one attribute required in a free license... -- RTA 19:54, 24 January 2015 (UTC)

Having uploaded hundreds of thousands of TIFFs and discussed whether to have jpegs or other formats instead, there are several learning points which are worth summarizing/repeating:

  1. As mentioned above, jpeg is not lossless. So if a lossless format such as TIFF, PNG or GIF is converted to jpeg, the quality of the image is going to be reduced.
  2. Jpeg is super handy for most reusers, especially for very large images such as archive/research quality scans of maps or high resolution panoramas from NASA, where the original file size is unrealistically big to either view in a browser or easily download to use. 95% of reusers are unlikely to make use of the full high resolution image, however the remaining 5% are researchers and volunteers that want to examine fine detail and these archive quality formats are precisely what they are looking for.
  3. TIFF is a rather special format, being a container where you can put images in various formats. Consequently chosing to 'unpack' the container or reformat the image has to be done with the detailed format in mind if conversions are to be lossless.
  4. Different formats will affect how well thumbnail generation works. For this reason having an image at the same resolution on Commons in TIFF and jpeg, means that it makes sense for Wikipedias to use the jpeg format as it is likely to render down to various thumbnail sizes rather better.
  5. Having an original version of an image on Commons that digitally matches a version held by organizational archives which are our sources, means that it is much easier to check that we have the image on our database already. If we only hold a compressed version, there is no currently agreed systematic way of checking that they are visually the same image.

Thanks -- (talk) 08:28, 25 January 2015 (UTC)

  • Don't confuse theoretical differences with real world differences though. In most cases, a low-compression JPG is functionally indistinguishable from a PNG or TIF. But as for your other points, I agree. Diliff (talk) 16:08, 25 January 2015 (UTC)

Deletion of bad picturesEdit

IMHO low quality pictures or pictures without a sufficient description are a bigger problem than non-categorized ones. What about deleting all unused pictures in Category:Blurred images?--Kopiersperre (talk) 19:14, 24 January 2015 (UTC)

Sometimes they are unique and irreplaceable, and until we do not have a better one, we could not delete. When we have a better one, that replaces, the blurred is speedy deleted. -- RTA 19:59, 24 January 2015 (UTC)
When a picture is unused and you don't even know who or what is shown, you can't replace it. Maybe we should tag these pictures with {{orphaned}}--Kopiersperre (talk) 22:07, 24 January 2015 (UTC)
I would agree you, but (as you can see on my user page "no admin", as only one reason) don`t wast your time with quality here on Commons. The most admins don`t see this. In the last time I tagged duplicates (different file types) a PNG copy of an SVG (my semi own created) and the admin User:INeverCry (this is only one, but personally I see him often keep crap. Sorry I don't have anything other trouble with him, in contrast to other). The very annoying boring primitive answer is always "In use".
Commons rule: If crap is in use don`t make a deletion request. User: Perhelion (Commons: = crap?)  09:30, 25 January 2015 (UTC)
Welcome to reality. A neighborhood may be an ugly pile of crap, but that doesn't mean you can just bulldoze it without permission from the inhabitants. Windows and Unix have huge issues because programs that worked 20 years ago can't just be broken, so they have to keep functions and systems around that have been long obsolete. So long as you're providing a system that other people work with, you can't just deleting things that they're using.--Prosfilaes (talk) 09:58, 25 January 2015 (UTC)
As I've seen on the EnWP, unused images get deleted there after 7 days.[10][11] (On the other side I see this only because I`ve tagged this images for deletion (NowCommons) which got removed (by User:Magog_the_Ogre as not same file type, otherwise a nice admin)
This seems all most incomprehensible for German user, because there images get very fast and easy deleted. User: Perhelion (Commons: = crap?)  09:42, 25 January 2015 (UTC)
@Perhelion: please read w:WP:CSD#F8: The Commons version [must be] in the same file format and [must be] of the same or higher quality/resolution. Magog the Ogre (talk) (contribs) 18:03, 25 January 2015 (UTC)
They delete images that are in use on the English Wikipedia on the German Wikipedia? How?--Prosfilaes (talk) 09:58, 25 January 2015 (UTC)
How about an experiment with "unused" and JPEG and "smaller than 100 KB" and "blurry"? Looking at the first page of Blurred images, maybe it's possible to rescue pictures with the proverbial (for QI) two megapixels. Very vague procedural idea:
  1. A bot collects candidates in a temporary "really bad" tracking category.
  2. Users verify and tag cruft with {{speedy|really bad}}
  3. An admin bot deletes really bad images in the "really bad" category.
  4. The first bot removes category "really bad" from all survivors.
  5. A human admin deletes the tracking category, because somebody created it.
  6. You write a "lessons learned" article for the Village Pump. ;-)
Be..anyone (talk) 10:26, 25 January 2015 (UTC)
Delete JPEG smaller than 100 kB is not even close to be acceptable, a lot of this < 100 kB are images like this File:K.c.kesavapillai.jpg, probably we will not find anything to replace that, and this works to illustrate.
And in five minutes a pass throw all 'blurred" photos, nothing to worry actually, 629 pictures is nothing, I do not see a use of ~100 of this, most dick photos, as we already have tons of this for educational purpose... (Obs: Is that hard to take a dick photo without blurring or miss the focus?)
So, why are you worrying about? -- RTA 18:39, 25 January 2015 (UTC)
unused doesn't match File:K.c.kesavapillai.jpg. –Be..anyone (talk) 19:04, 25 January 2015 (UTC)
How do you will know if that image is not important, and not gona be used in some day? -- RTA 05:28, 29 January 2015 (UTC)

Categorization through templatesEdit

Hi all, I've set the content of {{PD-La Moncloa}} to this text:

{{autotranslate|base=PD-La Moncloa}}{{#ifeq:{{NAMESPACE}}|File|{{{category|[[Category:Copyrighted free use files from La Moncloa|{{PAGENAME}}]]}}}}}<noinclude>
{{In category|Copyrighted free use files from La Moncloa}}
{{documentation}}
</noinclude>

The idea is to categorize all images with such a template in Category:Copyrighted free use files from La Moncloa. However, although all the files are correctly categorized (see this, for instance), when accessing the category, it's empty. Should each of the files be edited in order to get categorized? Best regards --Discasto talk | contr. | es.wiki analysis 16:32, 25 January 2015 (UTC)

  • Doesn't look empty to me. I seen 19 files there. Maybe a caching problem on your end? - Jmabel ! talk 18:47, 25 January 2015 (UTC)
These 19 were edited by me, so I think the files took the right category. Nowadays, there is one hundred. Maybe it's just a question of time (it takes some time to actually put the files into the right category). Best regards --Discasto talk | contr. | es.wiki analysis 22:31, 25 January 2015 (UTC)
It's probably just a matter of waiting for the category to populate. I know that used to take a while, though I have no idea how fast/slow it is nowadays. – Philosopher Let us reason together. 00:43, 27 January 2015 (UTC)
Never mind - looks like it's already categorized! (Next time I'll look at the date of the post before commenting...) – Philosopher Let us reason together. 00:45, 27 January 2015 (UTC)

Is there a replacement for derivativeFX?Edit

In the past, derivativeFX (see Commons:derivativeFX) by Luxo (who is inactive) was a very useful tool for uploading derivative works (e.g. cropped, retouched) of images already on Commons. It grabbed the original license, added a "derivative work" notice and the upload log of the original file. It was a Toolserver tool that didn't migrate to Labs when the Toolserver was shut down, see the author's note. So, I wonder: Has someone created something similar in the meantime, or what's your usual approach for derivative works now? Gestumblindi (talk) 17:04, 25 January 2015 (UTC)

For a simple crop, the CropTool will do - no down- and upload needed. In some cases this helps. On the other hand, UW supports upload of derivative work (it is not my own work), but you have to fill most fields manually. --Herzi Pinki (talk) 07:39, 26 January 2015 (UTC)

Looking for feedback on my funding proposal to work with UNESCOEdit

Hi all

I’m looking for feedback and endorsement for my Wikimedia Foundation PEG grant to be Wikimedian in Residence at UNESCO. I’d very much appreciate if you would have a look, the most relevant objective to Wikimedia Commons is:

2. Make content from the archives of UNESCO and its partners available on Wikimedia projects: This project will facilitate the upload of 30,000 images, audio files, videos, data and other content to Wikimedia projects from UNESCO archives (24,000 images), UNESCO Institute for Statistics (UIS) and other sources including 10 organisations changing their content license to be Wikimedia compatible.

I ran a pilot project that resulted in the images found in the category Images from the archive of UNESCO, here are a few examples:

If you think this is a worthwhile project please click this link and then click the endorse button.

Many thanks

Mrjohncummings (talk) 19:16, 25 January 2015 (UTC)

Checkmark This section is resolved and can be archived. If you disagree, replace this template with your comment. Dupe. –Be..anyone (talk) 20:25, 25 January 2015 (UTC)
Just to add that these are breathtakingly great photos, and my thanks and contratulations for them. -- Tuválkin 18:35, 27 January 2015 (UTC)

January 26Edit

Removal of a category so the file left uncategorizedEdit

User:Alan Liefting removed all the files from Category:Israel so they left uncategorized. I was failed to explain to him why it is better that files be in category instead of being uncategorized. I explained to him that files better to be categories rather then to be un categories. In a working place when all the time new files uploaded you will find all the time files in this category to be classified not to be removed from the category. I'll be glad if someone can help in this matter. -- Geagea (talk) 08:05, 26 January 2015 (UTC)

Personally, I prefer that people categorize files in a generic category, instead of not categorizing them at all. I often clean up Category:Berlin, for example. And since I do a lot of categorization work there, I know exactly where to put files, or at least where to look for categories. I do the same when I am not an expert in a subject matter. Just this morning, I categorized images without categories, and I put many in general categories, because I didn't know the exact sub-structure. And to be honest, trying to make sense of a such a structure takes a lot of time better spent - if I even get it right. TLDR: I agree with you. --Sebari (talk) 08:15, 26 January 2015 (UTC)
+1: I think it´s almost impossible to work the "uncategorized" heap and sort each file directly into the finest category. We should accept that it takes two or three steps as every level needs more expertise regarding the topic or the structure of the particular category tree. Country is a good first categorization step to increase the probability of someone taking further care of the file. --Rudolph Buch (talk) 10:36, 26 January 2015 (UTC)
It used to be that there was some kind of programming that prevented you from removing all visible categories, I guess now with cat-a-lot you can remove them in their hundreds and it is not so easy to undo, Maybe this should be fixed? The likely result of the files being removed from Category:Israel is that some editor with minimal knowledge of that country will come along and just add Category:Israel to the affected files Oxyman (talk) 10:38, 26 January 2015 (UTC)
Agree! --Jwh (talk) 10:43, 26 January 2015 (UTC) Country is a good first categorization step to increase the probability of someone taking further care of the file
I think the biggest problem is that the vast majority of uploaders are not familiar with how the category system works (espectially the fact that they are not tags) and are not familiar with the HUGE range of options of suitable categories. Alan Liefting (talk) 19:08, 26 January 2015 (UTC)

What about adding Category:Unidentified locations in Israel instead? And with this search you can make an educated guess about a lot of uncategorized images in Berlin per year (e.g. 2014). --Herzi Pinki (talk) 12:54, 26 January 2015 (UTC)

Not all of the images were of locations. Alan Liefting (talk) 19:08, 26 January 2015 (UTC)

Geagea, removing the Israel category from 1,157 files was not a decision I made lightly. I recategorised some, some were already in suitable subcategories, a bot added categories to others, and what remains will be picked up as uncategorised and eventually categorised correctly. Sure, removing the cat was not ideal but neither was having over a thousand files in a category that should have few or no files. I made a pragmatic decision mindful of the cost benefit analysis of my action. Alan Liefting (talk) 19:08, 26 January 2015 (UTC)

The categorisation of images on Commons is an utter mess. I don't know how viewers of Commons find images (and I don't know if any usability surveys have even been done) but if they use the category system they are confronted with one hell of a confusing, jumbled mess. Alan Liefting (talk) 19:16, 26 January 2015 (UTC)

I agree that, in most cases, a branch of Category:Unidentified locations is much to be preferred, I agree that our cat tree in general is a mess often of little value, and that the Mediawiki software, being designed for an encyclopedia and only slightly supplemented with specialized tools, deserves much of the blame. Jim.henderson (talk) 19:20, 26 January 2015 (UTC)
I agree with Geagea that it is wrong to remove generic category like Category:Israel from files which are not in the subcategories of Category:Israel. In most cases, whoever added that category felt like it belonged to one of the subcategories of Category:Israel. It might have been locations within Israel or people, works (books, movies), crafts (airplanes, ships), etc. related to Israel. Alan Liefting, I agree that the category system is not optimal, but it is better to have some general categories than no categories, and If you are not willing to put effort to improve categorization of those files, at least do not throw away work of others who caried the categorization that far. Of course some files might have nothing to do with any aspects of Israel, but that is hard to decide automatically for 1,157 files. --Jarekt (talk) 19:54, 26 January 2015 (UTC)
I also agree that Alan went the wrong way here. In the last 3 months, I've removed over 3000 photos from Category:New York City, but I've done it by moving them to a more appropriate category (typically a subcategory of Category:New York City, but not always: some were actually in New Jersey).
There is much more chance of a photo getting correctly categorized if it is in a too-vague but accurate category than if it is in no category at all. - Jmabel ! talk 01:32, 27 January 2015 (UTC)
In an ideal wikiworld yes it would be wrong to remove a category from a file that needed moving to a sub category. But we have a far from ideal wikiworld here at Commons. I am now curious to know how many files from which I removed the Israel category. If it is low enough that we can just move on, or is there actually an issue here? Alan Liefting (talk) 05:43, 27 January 2015 (UTC)
You removed it from 1,052 current images. See sandboxL. -- (talk) 11:03, 27 January 2015 (UTC)
But the question is how many were left uncategorised and on how many was information "lost". Alan Liefting (talk) 17:45, 27 January 2015 (UTC)
This is a question you should have been able to answer at the time. I can draw up a query based on the list of 1052 images on the above page and tell you which have no categories after your change, however my programming time is better used elsewhere and you could go through the list by yourself to put this right. -- (talk) 19:02, 27 January 2015 (UTC)
I made a judgement call at the time based on the contents of the cat, (the total number, the degree of categorisation of a sample of the files, etc), the type of category (high level), and based on past experience and concluded that on balance my action was justified. I agree the your programming time, and indeed the time of everyone here, is far better spent elsewhere. It is not as if there is nothing else to do around here!! Alan Liefting (talk) 03:51, 28 January 2015 (UTC)

Geagea, it is not true that I "removed all the files from Category:Israel so they left uncategorized." Yes, I removed that category but only some files were left uncategorised. Alan Liefting (talk) 05:55, 27 January 2015 (UTC)

@Alan Liefting I don't know how many files are affected, I'm just gonna take some recent examples here and this one, both (formerly included in Category:Art) are now without any category and also they are NOT in Category:Media needing categories; with the result that there is no chance anymore that anyone finds and properly categorizes it someday. OK, I don't know if that particular files are in scope for Commons, but that are just two examples out of a very big lot, I guess. So, what you did is a blatant misuse of the Cat-a-lot tool and by no means helpful for the project. --A.Savin 07:06, 27 January 2015 (UTC)
The two images you mention, which are almost as far removed as possible from art, are far better off uncategorised than being in such a high level category and, as you say, they may not even be appropriate for Commons. Also, it is not true "that there is no chance anymore that anyone finds and properly categorizes it someday". YaCBot comes around (sometimes after about ten days) to mark files as uncategorised, and I am sure a database report could be made for uncategorised files. Alan Liefting (talk) 07:52, 27 January 2015 (UTC)
Wrong. I occasionally do cleanup categories like Russia, Moscow and similar, so I know what I'm saying. Several times I found images having been without any categories (including maintenance ones) for two, three years. Here on Commons, you cannot entirely rely on bots. --A.Savin 08:12, 27 January 2015 (UTC)
Point taken on the bots and that looks like yet another problem that needs addressing. I am pretty sure that a database report can be made on uncategorised files. Anyway, given the poor quality and poor documentation of some of the files added to Commons lack of categories is no great loss. I sometimes come across files that makes me think why am I bothering with spending time editing such utter rubbish. Alan Liefting (talk) 08:21, 27 January 2015 (UTC)
Concerning two points addressed above, my humble opinion:
  • Removing all categories from a file and leaving none is something better dealt with in a specific page, and that isn’t this one. (Simply — whoever worries about files being found in a too-general, or even wrong, category should also took the trouble of finding a better one.)
  • Commons’ categories are a mess they say — this is heard a lot: But never followed by a description of a better system. Please come forward with a proposal for better categorization (something that can cope with millions of disparate formats and subjects); of course proposing better tools for the current system, and/or new features for the current system, and/or better/continued curation within the the current system — all that means that the current system is good enough, just needs some work. (On the other hand, everybody can think of much worse ideas, such as Flickr tags).
-- Tuválkin 18:35, 27 January 2015 (UTC)
I don't see why removing all categories would be seen as vandalism. There are sometimes files that are up for deletion and the uploader has spammed the image into lots up the upper level categories. Also, since categories are sometimes added by automated means there is nothing wrong with removing them by semi-automated means. And another thing, because of the HUGE workload that we have there is a need for expediency at the expense of thoroughness (but accuracy should never be compromised of course).
The categories are a mess because Commons is under-staffed and it is way to easy for uploaders to add files. Commons:Statistics suggests that it will get worse. Alan Liefting (talk) 19:05, 27 January 2015 (UTC)
Someone could probably make an abuse filter rule to find edits which remove the last (non-hidden) category. That would make it easier to track this. As for (un)categorization statistics, see User:Multichill/Categorization stats. Multichill (talk) 18:41, 27 January 2015 (UTC)
That is a good resource. Can you move it over to Commons namespace? I want to link it from Commons:Statistics. The page confirms what I already know - there are too few editors for the amount of work that needs doing. Removing that last non-hidden category will not always be "abuse". Alan Liefting (talk) 19:05, 27 January 2015 (UTC)

@Alan Liefting: even though you disagree you must acknowledge that there is a vast majority that categories shouldn't be removed so the file left uncategorized. So please don't do that. There is more chance that file will be used when he is in general category rather then uncategorized. The only people that work on uncategorized files are Commons users and many times they categorize them in a general category.

I opend this discussion here and did not notify you on AN/U as I saw that your intention is to help and not to harm. I really want that you understand. If you think that it is incorrect please made new proposal, but in my opinion, before you do so try to work more with our understanding. -- Geagea (talk) 07:56, 28 January 2015 (UTC)

The situation here is that I think what I did was an overall improvement to Commons and you are saying that it was not. In actual fact, in the absence of in-depth usability surveys, we really don't know what the best course of action is. But I made a judgement call and you disagreed with my judgement.
I don't agree with you that a file should be left uncategorised. As I have already stated ideally all files should be categorised but there are situations where it is ok to remove categories and leave it uncategorised. One situation where I am sure we can all agree on is a file that has been placed in numerous high usage categories but it is a candidate for speedy deletion. The situation with my edits to files in Category:Israel is a bit more subtle and complex. There were over a thousand assorted files sitting in Category:Israel for a long time. That is of no use to those who use Commons as a resource. Having all the files in one place to be recategorised is of some use to Commons editors but that is not the function of a content category. We use maintenance categories for that sort of thing. If there was a Category:Israel files needing categorisation as a subcat of Category:WikiProject Israel I would have placed them there. But there is no such infrastructure. There is actually a lot of infrastructure missing on Commons but that is another story.
You say that Commons users (editors) add files to general categories. That is the wrong thing to do. Editors should always add files to the most specific category as per COM:CAT. Unfortunately that can be difficult to do but it is made easier with HotCat and systematic category naming.
As for a proposal the problems as I see them go way beyond that. But that is another story as well.
Finally, why would you think I am here to harm Commons? Do my tens of thousands of edits look harmful? Or is it only a few? If the tens of thousands of edits are constructive why would I then do unconstructive edits? Alan Liefting (talk) 10:00, 28 January 2015 (UTC)
Looking this and this I can tell only that you have a lot to learn here. All your suggestions may be good and you may propose them, but please understand that the vast majority, of users working here for years, says don't remove category and left the file without category.
Please read again. I did not say you are her to harm. I said the opposite ot that. -- Geagea (talk) 10:14, 28 January 2015 (UTC)
@Alan Liefting:, please stop removing generic categories without replacing them with better ones. Last June user:Martin H. and user:Jmabel requested that you stop the same activity. After all those discussion and the discussion here, you are still doing edits like this, this or this. That is bordering on vandalism. Based on your edits and your proposal to delete template maintenance categories and your request to stop categorizing Institution templates, it is obvious that your ideas about categorization are quite different than of most other users, but you still need to follow the consensus of the community, or get blocked for vandalism. --Jarekt (talk) 20:04, 28 January 2015 (UTC)
Alan, I don't think your intentions are bad, but I do think you are editing against consensus. - Jmabel ! talk 23:54, 28 January 2015 (UTC)
Yes, that does appear to be the case. I am off on holiday and I have an ailing father so I probably wont be doing much editing over the next three weeks. Alan Liefting (talk) 06:31, 29 January 2015 (UTC)
I curious to know why my editing would give the impression that they are "bad". Alan Liefting (talk) 06:37, 29 January 2015 (UTC)
If user contribute against the community consensus it considered as a problematic behavior. Call it whatever you want - vandalism, disrespect or harassment. All the users mentioned: Martin H., Jmabel, Jarekt, A.Savin and others including me are experianced users that working with categorys for years, so please take thiere advices. The comunity have the ability to inforce consensus. It starts with warnings in the talk page and then notification in COM:AN/U which may bring to yourblock if necessary. Willingness to help is not enough. Following community consensus and experienced users are necessary as well. So take the opportunity, read the comments above and try to work with the same rational. -- Geagea (talk) 01:59, 30 January 2015 (UTC)
Yes, I know all that. I have been working on wikis for long enough to know how it works. I like to think my edits are based on sound rationale. With all due respect, I could argue that everyone else here is not being rational. Where is the evidence that my edits are problematic? My editing rationale is based on what is best for those who use Commons and it appears all the other editors commenting in this thread see Commons as something that is used by editors alone.
I have been editing wikis for over ten year but that is not relevant. It is the quality of editing that is important. Alan Liefting (talk) 06:16, 1 February 2015 (UTC)

I am slowly sticking the files without useful categories in Category:Israel_files_needing_categorization. If you follow the link cat-a-lot will work even if the category has not been created. It may not be a perfect search for these, but it'll do. -- (talk) 00:16, 29 January 2015 (UTC)

Thank you very much Fae, your work is helpful like always. -- Geagea (talk) 07:45, 29 January 2015 (UTC)
Thanks. The outcome was that 703 files out of 1052 checked were identified as needing 're-categorization' after removal of their last category. These have now been repaired.
This has been a long discussion over not much, however the resolution seems to be that it is a bad thing to remove the last meaningful category from an image, even if this is a category that needs diffusion. As @Alan Liefting: has kept this going, perhaps you could take the collegiate step of requesting a tick-box/preference to be added to cat-a-lot to prompt the user when removing the last visible category, it seems a realistic improvement? -- (talk) 10:27, 29 January 2015 (UTC)
What about the 350 files? They should be also in this category. If they have extra category it have to check manually. If they have category like plants it should moved to plants in Israel. If they have correct category then it's ok but it have to checked manually not by cat a lot. -- Geagea (talk) 11:02, 29 January 2015 (UTC)
Okay, adding the remainder for manual check. This will probably take about an hour. -- (talk) 11:14, 29 January 2015 (UTC)
Thanks. I'll see about cat-a-lot (not today). It should be a button to undo cases like this. -- Geagea (talk) 11:25, 29 January 2015 (UTC)
✓ Done At this point in time the non-created category Israel_files_needing_categorization has 346 files in it to check. -- (talk) 14:51, 29 January 2015 (UTC)
Thanks Fae. -- Geagea (talk) 01:59, 30 January 2015 (UTC)
That looks like an eminently suitable solution. Alan Liefting (talk) 06:16, 1 February 2015 (UTC)
@Geagea: why have you made the retrograde step of moving 800 odd files back into Category:Israel. Fæ has put forward a solution yet you choose to ignore it? It is good to see that you are recategorising them but it SHOULD' be done from out of the Category:Israel_files_needing_categorization. You are wasting your editing time (your choice) and messing things up for viewers (a concern for all editors). Alan Liefting (talk) 08:03, 1 February 2015 (UTC)
Fæ, the removal of the last category is just as important as removing any one of the other categories. It is not the absolute number of categories a file has but it is whether the category is appropriate or not. Removing a completely relevant category is wrong. Removing any category instead of sub-categorising is not the best (which is what I did) but can be a means to an end. Removing a completely irrelevant category is right. Removing all categories can can be right. Etc. Alan Liefting (talk) 06:16, 1 February 2015 (UTC)
All the files should process manually. they should not removed from the category:Israel. I have no more arguments to explain you more then it discussed her. Even that you made a lot of category edits in other projects it is not the same her. Fell free report me in COM/AN. Please do not remove general categoriys from files in cat a lot. It should be manually and only if the file have an apropiet category include. -- Geagea (talk) 08:26, 1 February 2015 (UTC)
I don't agree that all of the files should be processed manually. That is ideal but things are far from ideal here. If there are millions (or more) files that are automatically added along with inappropriate categories I see nothing wrong in using automated methods in removing those categories. We have a huge Commons:Backlog of work, increasing numbers of files being added, and an almost statice number of active editors. This all adds up to the need for expediency in editing.
I also disagree that none of the current files in Category:Israel should be removed. That seems to be a contradiction on your part since you are actually recategorising them at present. (There may be language barrier issues here?) Country categories should have very few, if any files in them.
I also don't fully agree with you when you say that other projects have different methods of categorisation. All wikis are essentially the same in how categorisation works. What does differ is the policies and guidelines. Here on Commons these are poorly developed (along with many others).
I see no reason to report you to COM/AN. I would do but only if it was really necessary but that is not the case. Alan Liefting (talk) 08:56, 1 February 2015 (UTC)
The category of Israel contains files to be processed by somebody that knows where to put them. Sometimes somebody adding a specific category but did not removed the category Israel. In that case it is o.k. to remove it. But many of the files have extra category which category Israel shouldn't be removed. For example. If the extra category is plants than the category Israel shouldn't be removed but both of the categories should move to category plants in Israel. If the file contain extra category December 2012 in Israel the category Israel shouldn't be removed only to move to specific one. The fact that you dont know what is the correct category that not means that you should remove it.
We are adding uncategorized files a general category like Israel and we expect that somebody from Israel will know to which specific category it belongs. It can be general category like maps. In that case we expect that somebody that maps is in his scope will categories the file with specific category.
Files in category Israel should be processed manually in any way. Uncategorised or not. You always find files in this category. That's the way it is working her. It is clear to me that you are not work here regularly. The category system here is definitely different than the other wikis because the aim of the projects is different and I already advised you to work with it before you come to conclusion. So please do not mass remove any general category from files with cat-a-lot or in another way. -- Geagea (talk) 10:42, 1 February 2015 (UTC)

January 27Edit

Layout problemEdit

At least for me, the layout of the Category:Lewis County Historic Courthouse page appears messed up somewhere around the inclusion of {{NRHP}}. Can anyone work out what's going on & fix it? - Jmabel ! talk 04:55, 27 January 2015 (UTC)

Works for me, as a quick test I added {{-}} in the category, does that have any effect for you? If yes and it's good, it could be added in the template. –Be..anyone (talk) 06:37, 27 January 2015 (UTC)
That fixes it for me. - Jmabel ! talk 17:08, 27 January 2015 (UTC)
Quick fix added to {{NRHP}}, please revert it when it's not more needed. –Be..anyone (talk) 21:20, 27 January 2015 (UTC)

Auschwitz AlbumEdit

Today, 70 years ago, Auschwitz was liberated. The Auschwitz Album created in 1944 is now in PD. I uploaded 22 pictures of 192 from Yad Vashem (labeled as PD on their website) in the Category:Auschwitz Album. Feel free to upload the other pictures, add them to articles and to help with identify persons --Hannolans (talk) 10:35, 27 January 2015 (UTC)

Thank you Hannolans. It's great that you started to do this and thank you for informing other editors here. I would help you but I think this is probably better done by just one editor since it's difficult to tell where one person started and stopped (also I'm not sure what copyright license you are using) and it's not a huge project. I hope it isn't too much work for you but please know that your efforts are sincerely appreciated. CheersMonopoly31121993 (talk) 17:57, 27 January 2015 (UTC)
Yes I can do this coming week. --Hannolans (talk) 11:33, 29 January 2015 (UTC)
Those images would be "PD" only if their photographers died the same year the photos were taken (1944), and then only since January 1, 2015. --Rosenzweig τ 00:53, 31 January 2015 (UTC)

چطور مقاله ثبت کنم؟Edit

سلام به همگی لطفا برای ثبت مقاله منو راهنمایی کنید مرسی — Preceding unsigned comment added by Zahrashirazi66 (talk • contribs) 13:34, 27 January 2015 (UTC)

  • If Google translate understands correctly, this is a question in Farsi asking how to save paper. I have no idea what it might have to do with Wikimedia Commons. Can a Farsi-speaker confirm that this is totally off-topic? Thanks. - Jmabel ! talk 17:14, 27 January 2015 (UTC)

No coordinatesEdit

When I click on coordinates I get from wfmlabs.org, "No webservice" and "This URI is part of the geohack tool, maintained by Dispenser, Magnus Manske, and Kolossos." Someone please repair. Jim.henderson (talk) 23:43, 27 January 2015 (UTC)

Working now; thanks. Jim.henderson (talk) 10:29, 28 January 2015 (UTC)

January 28Edit

Don't panicEdit

To deploy a patch for an urgent security vulnerability, the operations team at the Wikimedia Foundation are rolling out restarts of all servers. If something is down, from the wiki to labs to thumbnail renders/image scalers to whatever, never fear! they shall be back up very shortly. If the outage of something lasts for a really long time, do let me know. Keegan (WMF) (talk) 00:08, 28 January 2015 (UTC)

Surely this is done because a perfectly logical reason for our own safety. Thanks for keeping us informed about things like that, I wish they could inform other situations that are also made for our safety, as global locks --The_Photographer (talk) 15:24, 28 January 2015 (UTC)
This emergency patch was carried out by tech ops, which does not have anything to do with global locks. I believe you should get in touch with wikimedia legal if you're worried about abuses/breaches of the local terms of service. -FASTILY 05:45, 29 January 2015 (UTC)
The technical problem is this, nothing related to Global locks. — Revi 05:48, 29 January 2015 (UTC)

January 29Edit

Category:Garden apartmentsEdit

Apparently, several years ago Category:Garden apartments was deleted on the basis that it was "Not really that clearly defined as a category, moved contents to apartment buildings." This was apparently done without discussion. I agree that the term has more than one meaning, but is there some other term for the single-story, townhouse-like buildings, often with numerous units facing a common courtyard that are very common in the older cities of the western U.S. and certainly found in quite a few other places in North America? My concern is, there is something here that it seems deserves a category, and I don't have a different term for it. - Jmabel ! talk 02:46, 29 January 2015 (UTC)

How about creating the category but this time add a description similar to that you just gave to define the category? Oxyman (talk) 16:55, 29 January 2015 (UTC)
This makes me think on Hofje. Wouter (talk) 15:03, 30 January 2015 (UTC)
Certainly physically similar structures. In any case, I will follow Oxyman's very reasonable suggestion. - Jmabel ! talk 07:28, 31 January 2015 (UTC)

January 30Edit

Suggestion: create redirects to categoriesEdit

Welcome.because thr theimportance of categories for Commons.I suggest Create redirects for all categories So:

  • the redirect name:Category name

#REDIRECT [[:Category:Category name]]

(Except galleries names)

The process needs Bot, who can run a bot to do it? --ديفيد عادل وهبة خليل 2 (talk) 09:21, 30 January 2015 (UTC)

My English is not good, so maybe I do not understand your suggestion. But there is a well established policy for category redirects at Commons:Rename a category. One of the relevant points there (as far as I understand them), is to use {{category redirect|target}} for category pages instead of #REDIRECT, as this will allow a bot to automatically move files from the old to the new category and is recognized by the HOTCAT-Tool as well. --Rudolph Buch (talk) 11:51, 30 January 2015 (UTC)
My English is good enough to spot redirects for all categories and add a speedy Symbol oppose vote.svg Oppose. Replicating a complete namespace foo: with its foo:bar pages by #REDIRECT [[foo:bar]] on bar pages in namespace 0 (articles, here galleries) would be a surprisingly bad idea. Many important cases of category:bar actually have a gallery bar, therefore the page is anyway not available for a bogus redirect. If you really MUST have a shortcut use page CAT:bar for the redirect, this is documented on Help:Namespaces and an established vintage 2005 workaround. –Be..anyone (talk) 14:52, 30 January 2015 (UTC)

My English is not good also.People are looking at the main namespace first and then categories.Why do not we help them to quickly access to the categories?! --ديفيد عادل وهبة خليل 2 (talk) 07:37, 31 January 2015 (UTC)

It seems that the software already does what you are suggesting. If you enter a name of something to the search slot, where there is a category, but not a page, you will get redirected to the corresponding cat. E.g. try Spitzerberg. --Herzi Pinki (talk) 00:26, 1 February 2015 (UTC)

closed deletion request of teaching materialsEdit

Hey everyone once there was a deletion request to delete the files uploaded by user:dapeldo. It was closed by User:krd with the statement that the files will be kept. Now the request was reopened by User:ellin Beltz and some files have already been deleted. I would like to ask these files to be restored. I am holding the MOOC on Web Science on Wikiversity v:Web_Science. We upload videos and exercise sheets on wikimedia commons. All content is original by me and my coworkers and we are fully aware of what we are doing by putting a creative commons license to them (especially there is no copy right violation since we also pay close attention on what graphics we can use (most of them are in public domain)). As far as I understand these files are Open educational resources which is one of the scopes of WMF. Since they are used on Wikiversity (currently not all of the files a linked in wikiversity, since some of the content is not completly produced yet) I do not see how they are out of scope of wikimedia commons which is the media archive of Wikimedia projects. Since we also use them in our lecture it is highly unconventient if these files get delted all the time. Since we link them from our university homepage and students cannot access them. --Renepick (talk) 11:14, 30 January 2015 (UTC)

All the files are uncategorised, the description is not very helpful and some of the files are not in use. So I can fully understand that they were deleted. I´d suggest to categorise them and describe their content and relation to the Wikiversity project in the description field of each file. Then all misunderstandings are avoided and they will surely be kept in the future. --Rudolph Buch (talk) 11:32, 30 January 2015 (UTC)
We can certainly do that. What category and discription would you suggest so that admins will understand the problem. And in order to be able to give descriptions the files have the ber restored first --Renepick (talk) 13:58, 30 January 2015 (UTC)
A Category like Category:Web science MOOC at Koblenz University would either fit in Category:Wikiversity projects‎ or in Category:Wikiversity courses (the two a not very distinctive against each other in their contents). I think files in this Wikiversity category tree are rarely deleted for scope, so this category is a bit like an DR-thwarting amulett :-) I can´t really help you with the file descriptions as I don´t know your project, but at least the fact that it is a Wikiversity project should be mentioned along with the institution that supported it, and perhaps a rough outline of the specific file´s content. Most files at Commons have just a few words of description, but nobody will accuse you of being chatty even if you donate your files four or five full sentences... --Rudolph Buch (talk) 15:07, 30 January 2015 (UTC)
Hi: Agreed that the lack of description and category is part of the consideration here, but I'm still concerned about the statement "pay close attention on what graphics we can use (most of them are in public domain" and point out that all of them have to be public domain to be hosted on Commons. Sources for any open permission graphics which appear on the image upload page are required. For example File:Web Science MOOC Exercises Week 5.pdf please provide the source of the graphics in use on the page. The compilation may be your own work, but each item in it needs a source and a citation. Cheers! Ellin Beltz (talk) 16:45, 30 January 2015 (UTC)
Especially the graphics on the seet of week 5 (with all the icons and computers) are completely from scratch made by us in SVG format with inkscape in 2013 by User:Rob-nowman... have a look at all the videos that we created Category:Videos_for_Web_Science_MOOC_on_Wikiversity as you can see in there either the used graphics are public domain (most of them) OR we have created them (some of them) OR at the very end of the videos we put references to the original files (which where CC-BY or CC-BY-SA). In the sense of creative commons we should probably quote our videos if we reuse the screenshot with the computer network for exercise sheet 5 but since we are the ones who created the files we can reuse them at any point since we are the creator. Otherwise in all the videos that we have created and used the graphic we would have to quote the first video since the computers have already been used in there. Our university spends 3 full positions to employ people to create original content for this course. Especially in order to not commit any copyright violations (which by the way makes the process of content creation much harder) We do this in order to be able to share the teaching resources under an open licence... What is happening here (not for the first time) is really frustrating. Asking if everything is cool is ok. Deleting stuff if you detect a copyright violation and can proof it perfect. But assuming that our material are copyright violations and deleting without asking and trial is really frustrating. --Renepick (talk) 22:08, 30 January 2015 (UTC)
Let me see if I am understanding this... You write above that User:Rob-nowman created the graphics on images that User:Dapeldo uploaded and claimed as own work. Also, User_talk:Dapeldo has been fully noticed for the proposed removal of the files. Each DN one is given at least a week for discussion. I'm sorry you don't feel this is sufficient "asking and trial" but the concern at Commons is that the material be within scope, which includes full sourcing and open licenses. Cheers! Ellin Beltz (talk) 01:36, 1 February 2015 (UTC)

Flickr accounts and License launderingEdit

Regarding to this DR. Can someone make a list of files uploaded from Flickr account and the total uploades of the same Flickr account to Flickr is less then 10 (or 20). We can find possible COM:LL.
And one more - a list of of files uploaded from Flickr account which uploaded to Flickr in less then one hour before it uploaded to Commons. It could indicate of possibility to COM:LL. -- Geagea (talk) 12:27, 30 January 2015 (UTC)

I was slightly confused by the DR referred to, as there is only the one file on Commons loaded from that particular Flickrstream. Here's a general rephrasing:
  1. List all files uploaded from Flickr where there are fewer than 10 images in the Flickrstream
  2. List all files uploaded from Flickr where the upload to Flickr time is less that one hour before upload to Commons
Both can be done, "fairly" easily. I would do this by calling the FlickrAPI using flickr_people_getInfo. I have run through two examples below using a small script, the first is the account in the DR you mentioned:
username Chris Baugher
photosurl https://www.flickr.com/photos/82271859@N00/
firstdatetaken 2013-08-19 11:40:02
firstdate 1376937739 19 Aug 2013
count 1
username
photosurl https://www.flickr.com/photos/50398299@N08/
firstdatetaken 2000-07-20 14:16:22
firstdate 1274272808 19 May 2010
count 6460
Checking "count" solves your first question, while "firstdate" seems to solve the second. "firstdate" appears to be the date in Unix timestamp format that the Flickr account was created first photo was uploaded to Flickr, while "firstdatetaken" is the date of the earliest photo in the stream deduced from its EXIF data.
If we were running a large report, we may have to throttle to 3,000 queries per hour or fewer (if from WMFlabs, this might not be the only app using the same IP address to Flickr).[12]
P.S. I'm overstretched right now, but I can imagine the solution and can add this to my backlog to do it later, if nobody else experiments with a report.
Here's an example of what can be done by first searching Category:Deletion requests January 2015 (4,816 files) for images with flickr.com in the image page text, then searching the resulting 190 images for related distinct Flickrstreams with fewer than 20 images. This search takes about 2 minutes in total (running from my desktop, not labs).
Suspect Flickrstreams in Deletion requests January 2015
NSID Date Username Count
41599103@N05 2009-08-19 Ciko7 15
-- (talk) 02:29, 31 January 2015 (UTC)
Thanks Fæ. It is good to hear that it can be done easily. The two different cases above can help us to find possible license laundering. But the query should work for the existing files not those that already in DR. The EXIF is not important but only the dates of the uploads to Flickr and Commons. If file uploaded to Flickr and shortly to Commons there is good chance that it is license laundering. The files should be checked manually and carefully of course as it is not necessarily means it is license laundering. -- Geagea (talk) 04:02, 31 January 2015 (UTC)
This may be less of an issue for Commons than you think, in terms of quantities, though may be serious when it does happen. I have run a test last night which watched RecentChanges for uploads from distinct Flickr accounts and found about 1 per hour. All file edits were checked, so this does find accounts from old uploads with recent minor edits, though this may be useful if an old upload is now from a deleted Flickrstream. See Report, feedback or suggestions welcome. I'm imagining this can become a live report on Commons, updated every hour (only with suspect sources), with uploads dropping off the report after a week.
Based on the 2 highlighted suspect streams, I have created Commons:Deletion requests/File:The 2011 Film Independent Spirit Awards 2.jpg for one of them. So even now the process is adding some value. Smile fasdfdsfoiueire.svg -- (talk) 10:13, 1 February 2015 (UTC)
Thank Fae. Maybe we can try less than one day instead of one hour? -- Geagea (talk) 12:05, 1 February 2015 (UTC)
I'm running another test, just showing suspects. As I was getting false matches like this, I have changed the criteria to "Flickr account created within the last 30 days" OR "fewer than 30 photos in the Flickrstream". I'll link to results here when a few hours are done, then a full 24 hours. After that I'll think about putting it on WMFlabs and having it as a wiki page of suspected Flickrwashing for investigation rather than relying on my desktop. -- (talk) 12:11, 1 February 2015 (UTC)
Thanks very much. It can be very helpful. -- Geagea (talk) 12:18, 1 February 2015 (UTC)
A test "live" report is at User:Faebot/Flickrstreams of concern. This "live" in that new entries will be added from recentchanges in realtime. However improvement such as checking the log for past NSIDs being reported has yet to be added. Based on the last soak test, the criteria are now:
  1. Flickr account created in <= 30 days or photos in flickrstream <= 30.
  2. Image on RecentChanges has a first edit < 90 days ago (taken as equivalent to image upload date).
By all means watch the test report, however this may be overwritten by better formatting or stop being updated while in a beta version.
Keep in mind this is "Flickrstreams of concern" rather than "Flickrwashers". Uploaders should be checked with a presumption of good faith, especially if they just need some advice on copyright before continuing to contribute to the project. -- (talk) 18:34, 1 February 2015 (UTC)

GLAM-WIKI 2015 conference, April 9-12 2015, The NetherlandsEdit

Wikimedia Nederland welcomes interested Wikimedians and GLAM enthusiasts to join us at the GLAM-WIKI 2015 conference, from 9 - 12 April 2015 in The Hague, The Netherlands. The call for proposals and application for scholarships are now open!
Ter-burg (talk) 14:57, 30 January 2015 (UTC)

Geograph upload requestEdit

A long-standing bug means I cannot use my TUSC login.

Pleaase could someone who can use geograph2commons to upload:

and place them in Category:Chad Valley toys? Andy Mabbett (talk) 16:19, 30 January 2015 (UTC)

Using image nominated for deletion, because its unusedEdit

I want to use a picture in a wiki article. But its currently nominated for deletion (because the picture itself is unused). So should I wait for nomination to close? Or should I go ahead & put it in the wiki? And if its being used in the wiki, that means there's no need for deletion. right? thanks Emphatik (talk) 20:42, 30 January 2015 (UTC)

If you use it and it stays, than it should not be deleted as "out of scope" because it is unusable by any wiki project. Although it is hard to imagine how such an image might be useful for anything. --Jarekt (talk) 21:14, 30 January 2015 (UTC)
Yes, it is hard to imagine how such trivial lens flare images could be tout as a proof for the existence of a mythical brown dwarf companion to the Sun (as opposed to, say, data from orbital infrared telescopy), but indeed this happens (as reported by User:Emphatik) and that’s why they need to be kept and described as such in Commons. -- Tuválkin 21:28, 30 January 2015 (UTC)
You can use a picture here as you like, INUSE is relevant for deletion debates. If you don't mind the red link and the broken image link error tracking you can also link to it after it was (hopefully, in this case) deleted. –Be..anyone (talk) 08:28, 31 January 2015 (UTC)
Alright. I've attached the pic to the wiki as per your suggestions. cheers Emphatik (talk) 16:49, 31 January 2015 (UTC)

Wilfredor → The PhotographerEdit

Wilfredor changed his account name to User:The Photographer, and that needed to be fixed in references to the former name, namely in talk pages and file pages. While I’d expect to be taken care of by means of a redirect in the user pages of the former name (as, f.i., was done for the name change Wiebevl → Vunz), apparently User:Lucia Bot has been in the past few days editing most (all?) uses of the string "Wilfredor" to "The Photographer", not only in links but also in text, including archived deletion requests, admin noticeboars, etc., and affecting other people’s comments (see one example). This strikes me as technically unnecessary and prone to cause more confusion than it solves. -- Tuválkin 20:44, 30 January 2015 (UTC)

I have corrected the above comment and section name so that it doesn't look as weird as it did. I am reverting that bot edits on my talk pages and archive pages, for that sole reason, they are unnecesarily editing historical pages. Diego Grez return fire 04:23, 31 January 2015 (UTC)
This bot was not approved for such a task. --Dschwen (talk) 00:49, 1 February 2015 (UTC)

January 31Edit

POTY bannerEdit

At top right says "Image credit (2007 winner)." That seems pretty odd, in that it's hardly an image credit, but even beyond that the image shown appears to be a portion of the 2010 POTY winner, not the 2007 winner. I have nothing to do with POTY, but would someone who does please sort this out? - Jmabel ! talk 07:22, 31 January 2015 (UTC)

Discussed on the POTY talk page and in an archived thread on en:VP/T. –Be..anyone (talk) 08:03, 31 January 2015 (UTC)
I was unaware of those discussions, but the upshot there seems to be "Yes, someone on Commons should fix that." Or do you read them differently? - Jmabel ! talk 01:49, 1 February 2015 (UTC)

{{BArch-image}} and WikidataEdit

I'm not sure if this should be a proposal or a bot request, but I was browsing through the Bundesarchive pictures and I noticed they reference people and places by links to de.wp (which is already very nice). It would be even nicer if we could use wikidata to allow for the internationalization of links. Any ideas on how to do that?--Strainu (talk) 17:15, 31 January 2015 (UTC)

What do you mean by "the internationalization of links"? Ruslik (talk) 17:34, 31 January 2015 (UTC)
On Commons most of the links to countries and big cities are internationalized. Try Paris, Warsaw or Bactria I think that what Strainu meant. Our links were created through templates created out of old style interwiki links. I assume that at some point we will do it automatically based on wikidata, but that is not doable at this stage. --Jarekt (talk) 17:42, 31 January 2015 (UTC)
Yeas, that's what I meant, but applicable to any article. Something like {{Wikiarticle|wikidata_item=00000|default_lang=de}}. Sorry to hear this is not feasible at this point.--Strainu (talk) 21:30, 31 January 2015 (UTC)

Can an administrator review his/her own uploads?Edit

According to Commons:License review, Please note that as of 21 February 2012, image-reviewers may not review their own uploads unless the account is an approved bot. However, I don't know whether such a statement applies to any reviewer or just to non-admin reviewers. Any clarification? Best regards --Discasto talk | contr. | es.wiki analysis 18:01, 31 January 2015 (UTC)

It applies to ANY reviewers, so administrators AND license reviewers cannot review their own uploads. — Revi 18:05, 31 January 2015 (UTC)
That's what I thought, but IMHO the redaction can be seen as ambiguous as "image-reviewers" is a explicit flag (implicit to admins). --Discasto talk | contr. | es.wiki analysis 18:12, 31 January 2015 (UTC)

February 01Edit

No image summary at smart phonesEdit

Hello Village pump, by chance i passed by a displaying problem on smart phones (Iphone 5s with Safari/Chrome and Android with Chrome). It seems that photos that contain location information such as this do break the displaying of the summary which then consists only of the location but no other data such as description, author etc. This does not happened when looking at this mobile view on desktop browsers (at least not in Firefox/Chromium on Ubuntu 14.04). So it must be connected with rendering on smart phones. Can anyone fix that? Thank u in advance, --Arnd (talk) 18:19, 1 February 2015 (UTC)