Commons:Village pump

From Wikimedia Commons, the free media repository
(Redirected from Commons:Village Pump)
Jump to: navigation, search

Shortcut: COM:VP

  Welcome   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 ↓
Village pumps for 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 | українська | 中文(简体)‎ | 中文(繁體)‎ | Zazaki | +/−

Welcome to the Village pump

This page is used for discussions of the operations, technical issues, and policies of Wikimedia Commons. Recent sections with no replies for 7 days and sections tagged with {{section resolved|1=~~~~}} may be archived; for old discussions, see the archives.

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


Cast iron pump with handle dated 1875 in the form of a fluted column with Corinthian capital on a profiled, square stone base [add]
Centralized discussion
See also: Village pump/Proposals • Archive

Template: View • Discuss • Edit • Watch



Conversion of still GIF images using color or transparency to PNG[edit]

So I have been running User:GifTagger for some time until it was blocked due to "Problem with the editing and the enforcement of an opinion without conversation with the relevant community" in January. I had to clean up the code and was too busy with other things that time anyway. Though, now I feel like getting a discussion up on this topic. I had some users on my talk page asking if I could fire the bot up again but I'd rather have any sort of feedback before getting blocked out of the blue again. --McZusatz (talk) 20:20, 17 June 2015 (UTC)

Just to start the discussion, I will present what I think are the advantages of using PNG over GIF:

The thumbnail of this gif is useless as it only shows some random black pixels. Sadly it was used in two articles.
Much smoother thumb
  1. Better thumbnails
    C.f. thumb on the right; GIF thumbnails are often pixelated due to transparency (Only a single transparency index is supported). Also PNG thumbnails support 24-bit RGB in contrast to a maximum of 256 colors in GIF (thumbnail) files.
  2. Smaller size
    GIF files tend to be larger than PNG
  3. Easier editing/"Lossless" editing
    Someone may want to edit those files. Surely the editor prefers a 24-bit palette over the 8-bit GIF-palette without uploading a new PNG derivative file before or after his edit by fiddling with (partly) broken tools such as UpWiz or DerivativeFX. Also you can not apply most filters to indexed palettes which results in a lossy process: Palette -> 24 bit -> Apply filter -> another? Palette
    However, the conversion from GIF to PNG is lossless and also the transparency, if present, does not get lost. (PNG supports 8-bit transparency in contrast to GIF's 1-bit transparency.)

--McZusatz (talk) 20:20, 17 June 2015 (UTC)

My opinion is to leave the GIFs as is and only update them on a needed basis. A large number of GIFs should be replaced by SVG (vector data) to begin with, so a extra PNG is unneeded in a large number of cases. Thumbnail point you made could be dealt with by generating PNG thumbnails with a anti-alias option instead of generating GIF thumbnails (i.e. similar to how we convert SVG to PNG for thumbnails/re-sizing). This is an issue that could be fixed through the wikimedia software end without having to re-upload a new version of every file. Offnfopt(talk) 22:27, 17 June 2015 (UTC)
@McZusatz: One rationale I find hard to understand is (for example at File:Enrico cialdini.png): "This GIF was problematic due to non-greyscale color table." Isn't it correct for a colour image to have a non-greyscale colour table⁇ I'm not sure which of your other rationales may also apply, but the filesize was bigger than that of File:Enrico cialdini.gif. I'm also a little concerned that (non-admins) lose the edit history of the deleted files, so it's harder to check the licensing etc. --99of9 (talk) 06:25, 18 June 2015 (UTC)
99of9 After reading your comment, I went ahead and ran a small test case of some of the files in the contrib list for the bot. I don't have access to see the pages for the deleted files, so I converted the PNG files to GIFs using the exact same indexed palette found in the PNG, so there is no loss or change in the image from the conversion I did. The original GIF file sizes may be different from mine depending on what options they used to save the files, I didn't do any special optimizations or tweaks to the files. The below table is the results of this small test case. You can see from the table that the claim of PNG having a smaller file size did not hold up to this test case. Also as I stated before, the bulk of these images are better served being converted to vector graphics due to the nature the pictures. Offnfopt(talk) 08:50, 18 June 2015 (UTC)
File GIF size (in bytes) PNG size (in bytes) Smaller file size PNGopt
File:Enis logo2.png 1,997 2,150 GIF
File:Enlarged lateral ventricles in schizophrenia.png 45,747 58,353 GIF 39.227
File:Enlarged spleen.png 10,106 10,960 GIF
File:Enlluita-logo.png 5,500 6,009 GIF
File:Enmascaramiento temporal.png 4,032 5,391 GIF
File:Enneagram - 2.png 30,934 31,843 GIF
File:Ennea triangle.png 1,139 2,161 GIF
File:Enrico cialdini.png 19,980 23,409 GIF 17.314
File:Enrin u0.png 13,577 20,241 GIF 11.227
File:Ensea-fonctions.png 7,854 7,815 PNG
I can agree Offnfopt and 99of9. The file size is only one point, as I mentioned to McZusatz before, the bot need a simple PNG optimizing function (I used the very fast PNGGauntlet). So I only vote for the bot if he has such functionaltiy. User: Perhelion (Commons: = crap?)  09:29, 18 June 2015 (UTC)
I replaced the poor PNG optimization with some better one in the meantime, so this should not be an issue any more. --McZusatz (talk) 10:11, 18 June 2015 (UTC)
@99of9, Offnfopt: Your proposal to use PNG for thumbnails instead of GIF, indeed solves the issues that can be observed in GIF thubnails today. Nonetheless, I'd still prefer to have the source image in PNG as well to make it easier for users to upload an enhanced version without going through quantization of the color values. This quantization degrades an image slightly if the color values don't match the quantization levels. --McZusatz (talk) 10:11, 18 June 2015 (UTC)
I want to add that I'm not against the bot per se, when I first posted I was in the mind set thinking the bot was going to leave the original GIF, so I was thinking it was going to cause a large number of duplicate files. But being that is not the case and letting the pros and cons rattle around in my head for a while, I'm warming up to the idea. Though I do worry about the bot encountering a file configuration it may not be able to handle and end up outputting a garbage PNG file and removing the original GIF. I wonder if the bot should generate a paged gallery log of files converted so others could keep an eye on the changes being made without having to scroll through the contrib log of the bot. I also support the good suggestion by Perhelion. Offnfopt(talk) 10:29, 18 June 2015 (UTC)
There is but I can also create a custom log page if you want me to. --McZusatz (talk) 14:42, 21 June 2015 (UTC)
@Billinghurst: If you have any issues with the Bot, please raise them. --McZusatz (talk) 14:42, 21 June 2015 (UTC)
@McZusatz: My issue was that the bot was deleting images, not that it was running an image generator. The bot and this community should not be overriding how the communities have placed and are managing their inclusions. Commons is a central repository, not a determinator. If a community has and chooses to use a gif, then leave their image alone, and just do your conversion. If you wish to discuss with the communities about your bot and its actions then start an RFC and invite these communities to join the conversation. The means exists.  — billinghurst sDrewth 13:57, 27 June 2015 (UTC)

The "non-greyscale color table" bot comment is due to giving non-transparent greyscale GIFs lower priority for conversion (i.e. not to be replaced without a further round of discussion at a future date), because of the fact that PNG thumbnailing for many years was significantly worse at resizing opaque grayscale PNGs than GIF thumbnailing is at resizing opaque grayscale GIFs -- and also various developer comments which have led to suspicions that improving PNG resizing in this area is not a WMF priority, and in fact that the very belated partial progress which was finally made in such PNG resizing could actually be reversed if WMF developers choose to adopt image software with some nifty features in other areas. See the discussion at Commons:Bots/Requests/GifTagger... -- AnonMoos (talk) 12:32, 18 June 2015 (UTC)

I will again emphasise to administrators and bureaucrats that it is not the job of Commons to set the standard that the sister wikis use, nor to think that we impose componentry on these wikis, they are their own bosses. This is the central repository that is used by all the wikis to make the images available to all wikis. We determine that files are within copyright, and that the files are within scope. More than that should be taken up with the wikis where we asked for them to come and contribute not enforce our opinion from limited conversations, in limited corners of Wikimedia.  — billinghurst sDrewth 14:07, 27 June 2015 (UTC)

FWIW, its really easy to make MediaWiki render non-animated gifs as PNG files, if there was wide agreement that that was the right thing to do. Bawolff (talk) 08:11, 2 July 2015 (UTC)

Imo there is no reason not to create PNG thumbnails of all still GIFs instead of obviously broken GIF thumbnails. --McZusatz (talk) 10:47, 2 July 2015 (UTC)
@AnonMoos: Would you oppose serving non-animaged greyscale GIF images as PNG thumbnails as well? --McZusatz (talk) 10:47, 2 July 2015 (UTC)
@AnonMoos, McZusatz:It still actually does not answer the intended reality of why that person actually edited/changed the images. But, in reality the Wikimedia rules or norms state the better the image, the better it is accepted in the Wiki community. Technically PNG images are always better than Gif images. Also, most Wikipedia chemistry articles have molecular arrangements as either a PNG or SVG image. So this person does have a point on the reason for converting it. But, I still do agree with the actions taken by the admins, for blocking the User since he/she should have talked or had a discussion before changing the images all of a sudden. @GifTagger: All wikis are communities were changes should be discussed before hand. From the friendly, Doorknob 747 (talk) 21:44, 5 July 2015 (UTC) :D

Help needed to classify old-timer automobiles[edit]

In Category:Hoek van Holland there is a series of pictures taken on 2010-05-16 of vintage cars. There must have been some event at that time. (or they where using the ferry)Smiley.toerist (talk) 21:48, 18 June 2015 (UTC)

I changed some by looking up the plate, but I did a reverse search on one of the images and found this page de:Benutzer_Diskussion:Spurzem/Archiv/2010#einige_Autos... a lot of the same images, but they're lower quality than the ones you're going through. But you can get some of the car descriptions from that page. Offnfopt(talk) 22:18, 18 June 2015 (UTC)
I made a new Category:Hoek van Holland (car park) to reduce the number of pictures in the :Category:Hoek van Holland.Smiley.toerist (talk) 11:15, 23 June 2015 (UTC)
You could also, look up Peugeot, Lotus, Ferrari, Maserati, and other european car companiey car companies to find the actual name of the car. If it looks like a bug, its a Volkswagen 75%. Froom the friendly and very helpfull, Doorknob 747 (talk) 22:13, 30 June 2015 (UTC) :D

June 19[edit]

Question about category usage[edit]

I have been working with categories and I would like some guidance/clarification on the usage of a category. Some articles contain categories for Category:People by name, Category:Men by name, Category:Women by name or some combination of the three. Some articles have flat list version of these categories, often in addition too those previously mentioned. Such as Category:Men by name (flat list) or Category:Women by name (flat list). My question is, is there some standard in place as to when to place one over the other or both? My opinion is that we only need one, either X by name or X by name flat last (I would prefer the former over the latter personally) but not knowing the history of these cats I was curious what the rules are for their usage. Reguyla (talk) 00:36, 24 June 2015 (UTC)

All people categories should be in flat Category:People by name that is our main people category. I have never noticed Category:Men by name, Category:Women by name, Category:Men by name (flat list) or Category:Women by name (flat list) before, but they seem redundant. --Jarekt (talk) 12:26, 24 June 2015 (UTC)
The category guidelines we have can be misleading. Trimming a file's categories to just a couple of super-specific categories will damage our project in the long term. Today, doing category intersections is overly complicated and beyond what can be expected of a new contributor. However, if we keep a balance between handy super-specific categories and generic categories with thousands of files (Category:Yellow flowers, Category:People by name, Category:Ethiopia) then in a couple of years time as well as searching out a useful category of images, even a new user could find, say, photographs of red-headed, 30 year old men, wearing sunglasses and with a bicycle because the WMF development team had done such a good job improving the mobile interface... :-)
Many volunteers disagree with my viewpoint on "bucket categories" and are busy 'diffusing' all images out of these top level categories, so perhaps the answer will be eventually implementing a system for "tags" where nobody cares that many thousands of files have the same tag, in addition to categories. It's been discussed here several times, so it may need someone with time to push forward a funded volunteer driven project to make this a reality. -- (talk) 12:47, 24 June 2015 (UTC)
The flat list categories for men and women were created (and, I believe, populated) by the same person just this past February. You could ask that editor what he/she had in mind. You could also propose deletion of the flat list categories. If they didn't exist until this past February, they probably have little function. I also see that these categories are under discussion at Commons:Categories for discussion/2015/02/Category:Men by name (flat list). --Auntof6 (talk) 05:37, 25 June 2015 (UTC)
Thank you. I do agree that in some cases it might be useful for an all encompassing "People by name" category but I think generally combining those into smaller more digestible subcategories such as Male/Female, Men/Women by name categories is better. For what its worth, as I learn the site and edit more, I don't really like the assumption of Category:Living people equals Category:People by name either. As for images, I also think that there is a lot fo confusion about what categories should go on Images and Which ones should go on Categories and thats probably worth some discussion at some point too.Reguyla (talk) 11:04, 25 June 2015 (UTC)
"Living people" obviously doesn't equal "People by name", because the latter would include dead people. As for which categories go on files/images and which go on categories, any of these "by name" categories that we're discussing, whether flat or not, would only be on other categories. This is the kind of discussion we get into when people who aren't familiar with the category structure create categories just because they can. --Auntof6 (talk) 04:07, 26 June 2015 (UTC)
You say Category:Living people doesn't equal Category:People by name but that's exactly where it and several other unrelated categories redirect too. I'm also not sure if the unfamiliar with category structure was directed me, but I have admitted that I am still learning and asking questions of how it works is part of that learning process. :-)Reguyla (talk) 13:42, 26 June 2015 (UTC)
Category:Living people redirects to Category:People, not Category:People by name. Apparently Commons doesn't need the distinction between living and dead. It actually makes sense that we wouldn't need it here: on Wikipedia they would need it because of policies concerning what can be said about living people. We don't have content that says things, so we wouldn't need it.
My comment wasn't specifically directed at you or any individual. It's just a fact that we get new editors who don't understand there's established structure, so they do things that have to be undone. I've done it myself, and I may inadvertently do it again in areas where I don't realize there's an established process. Since Wikimedia doesn't require training before people edit here, and since established structure isn't well documented anywhere, it's just a fact of life. --Auntof6 (talk) 19:47, 26 June 2015 (UTC)
@Auntof6:@Reguyla:The reason why is that Wikimedia just like Wikipedia is incomplete. For example, if you talk about a picture uploaded of gay marriage in Kentucky (or Kansas, don't even care about the state that was forced to accept rule) before the Supreme Court ruling, a tag illegal activity would be wise to put when uploaded, now after the ruling, you could put a tag named legal activity (not needed), but, it would be wise to replace the category 'illegal activity' with, 'previously considered illegal activity in Kentucky', since the image was taken in the time when it was considered illegal. As time, goes on, new categories may become necessary. That is why there is no justification, on what categories should exist or strict rules to enforce it. The only rule in categorization or creating category pages is that, it must be humanly possible, an example a category like 'real life flying truck/car' is not allowed and would be deleted (probably also user would be blocked).

Forgot this part, in the same sense, time is the major reason behind this! If a person changed there name via legal papers, thier birth name really is not official anymore, and for this reason the man/women by name thing does not count anymore, so its now catagory people. Same for people who went thru gender change surgery, they wont count for people who are man/women by name, so its would be people by names. There is a reason for everything. The reason for these is because things change.

I only used the gay marriage and Kentucky stuff for an quick example, I AM NOT GAY NOR DO I LIVE IN KENTUCKY or what ever state that was, I LIVE IN NEW YORK CITY!!!!!! SO IF ANY ONE CALLS ME GAY I AM REPORTING THEM to admin bulltin board or what ever board! From the friendly, Doorknob 747 (talk) 21:52, 30 June 2015 (UTC) :D

Pulled the plug on CategorizationBot & ErfgoedBot[edit]

I've been running these two bots for a while. Both still run on old pywikibot compat code that will break soon. I could invest time in updating these bots and I have been considering doing that for quite some time, but I realized I just don't have the motivation for it. Commons has become a toxic place dominated by a small group of very vocal users. These users don't seem to accept that Russavia is banned and to try to turn everything into "Commons community" vs "WMF" battle. I don't identify with that and I don't feel like investing time in a project where this behavior is acceptable. Multichill (talk) 10:40, 28 June 2015 (UTC)

@Multichill: I don't think that commons is a toxic place, there are a lot of volunteers doing very hard work (which never participating in AN discussion). I also fail to see a "Commons community" vs "WMF" battle. I am highly disappointed by your statement here and i think under this circumstances you should also step back your sysop bit. Best --Steinsplitter (talk) 11:48, 28 June 2015 (UTC)
I am rather surprised that someone feels Commons is a toxic environment but I can empathize because I feel the exact same way about the English Wikipedia. I understand not wanting to make the investment of time to recode the bots but I thank you for the effort you went through to do it this long. I have found in the past that taking a break often helps me feel differently about things and have found in the past that when I stopped doing something or "retired" I missed working on the project and came back. I also don't really see the "Commons community" vs "WMF" battle but I do see that the WMF periodically oversteps their authority and does things counter to what the community wants simply because they can and because its more convenient for them. Some members of both communities feel like they don't need the other when these projects cannot succeed without mutual cooperation between the communities and the WMF. IMO the Russavia case illustrates that disassociation. Most of the time the WMF won't help the projects when asked to do so but then when it wants to do something it does it, violating any previous statements it made to the contrary. Reguyla (talk) 13:10, 28 June 2015 (UTC)
  • Sorry to read this, Multichill; I was brought to this link by your post on the mailing list. What a shame that others had to make comments that entirely prove you right. Risker (talk) 14:17, 28 June 2015 (UTC)
    Risker, you have contributed to Commons on 4 occasions this year, and uploaded a total of 4 images in the 8 years of your account history. A part of having open discussion on Commons is that we regularly see people parachute in who seem to enjoy stirring up drama, are highly vocal off-wiki or off-project (as I know you to be), and dictate to long term contributors that we are terrible people for devoting our unpaid volunteer time here for the benefit of open knowledge. Many of us hide out on IRC and discuss issues on closed channels rather than risking becoming target of these sorts of unpleasant types of griefing. If you want to do something positive for Commons, take part in one of our projects for a while, and use that experience to form a solid evidence-based opinion on how our community works and how it can improve. -- (talk) 14:36, 28 June 2015 (UTC)
Yes, I know, Fae. I do not normally participate here because of my earliest interactions in which I was treated rudely and (in at least one early case) in a manner worse than I've ever been treated on English Wikipedia, which ironically I consider to be a pretty unpleasant place for a lot of users - it's the one thing Reguyla and I will agree about. I'm not a photographer, and I'm not looking elsewhere on the internet for images or other media; this wouldn't ever be my home. I do, however, care a lot about developers, oddly enough. I do care when people who have built useful tools that help the entire Wikimedia community feel that they're no longer valued. I'm shocked that the first reaction to his post was "please hand in your sysop bit". I'd hope you be shocked too. Risker (talk) 14:50, 28 June 2015 (UTC)
@Risker: I am disappointed by your comment :-(. Why you put oil in the fire? There might be some toxic users here, but there are also toxic users on enwp, dewp and other wikis. Calling a project "toxic" (putting all users under the same umbrella) is very immature. It is hurting to see Multichil, a commons admin, talking so bad about commons. I applicate his botwork and it is sad to see that he does not port his bots to core. If he really thinks that commons is so bad, he should probably step back his sysop bit or doing something to make commons better. But such "commons is toxic" posts are imho not helpful at all. It is not fair to call commons toxic. Commons is a wonderful place with a lot of hardworking users, and it is always a pleasure to work together with other users here on commons. --Steinsplitter (talk) 15:54, 28 June 2015 (UTC)

Thanks for your work Multichill. @Risker: shame yourself for your attempt to start a flame war. --Denniss (talk) 14:22, 28 June 2015 (UTC)

Sorry about that; it was not my intention to start a flame war, which frankly I think started with the first post. I am actually quite shocked that the first two comments are defending the atmosphere that Multichill finds distressing instead of "sorry to see you're not happy here, would you be willing to share your code with another developer can look at making the necessary changes?" That's what I'd really hope to see. If the work these two bots have been doing is valued by the Commons community (and given they have made a combined 1.92 million edits, I suspect their work has been useful), I'd hope that the Commons community would focus (a) on trying to find a way to continue the work of the bots, for its sake and (b) simply express sorrow that a longstanding member of the community has chosen to leave, instead of asking for them to completely disassociate with the project. (Seriously, asking for his sysop bit?) Risker (talk) 14:41, 28 June 2015 (UTC)
Emotionally colouring/escalating this thread by introducing words like "distress" and "shocked" does not help show your intention is to avoid creating drama. Could you confirm which email discussion you referenced that directed you here? It does not appear to be one of the open lists I am subscribed to. -- (talk) 14:57, 28 June 2015 (UTC)
The mailing list is wikitech-L, which is a public list. Thread is "API BREAKING CHANGE: Default continuation mode for action=query will change at the end of this month". Risker (talk) 15:05, 28 June 2015 (UTC)
Just to clarify, my post was not intended to flame anything but to thank them for their participation, encourage them to take a break and continue and to empathize for their feelings because I know how it feels from my time on ENWP. I would like to see at least some of the work done by these bots done by someone if not Multichill but I hope they change their mind. I also find it somewhat ironic that they blame this community for not being nicer to the WMF and then mentions 2 issues with his bots that are the direct result of actions by the WMF (although I admit they are for the long term good of the projects). I am also glad that me and Risker have some common ground. Reguyla (talk) 15:08, 28 June 2015 (UTC)
Thanks, for the record I see the link in this post this morning. I keep an eye on Labs-l but the tech list (and the IRC channel) is a bit too busy with so many different things I rarely try to follow them. -- (talk) 15:15, 28 June 2015 (UTC)
@Multichill: thank you for your hard work. I completely agree with you and feeling much the same way, sadly it isn't a nice feeling to have when you've lost interest and motivation. Bidgee (talk) 16:17, 28 June 2015 (UTC)
I know how you feel Maarten... It's really, really, demotivating. And yes, en.wp isn't too much better. —TheDJ (talkcontribs) 19:07, 29 June 2015 (UTC)
Depending on who you ask, myself included, ENWP is far worse than Commons. I will say though that things are getting better now that some of the historically problematic admins are getting to the Judge, Jury and Executioner (AKA , the Arbcom). So things are looking up there. Reguyla (talk) 19:55, 29 June 2015 (UTC)

Thank God for this decision! Erfgoed bot will not be missed (by me, anyway).--Strainu (talk) 18:58, 28 June 2015 (UTC)

@Multichill: Do you have a source code repo for your bot's code and what is the main entry point. Is /data/project/catbot/readme.txt still up-to-date? Also, could you share what was set in cron? Thanks in advance. If it is on Tool labs, which free license did you choose for the source code? -- Rillke(q?) 20:05, 28 June 2015 (UTC)

The cat script seems to be on labs and freely licensed (short test: seems broken). See grep -R "Please add categories to this image" /data/project/catbot/. On the other scripts i haven't seen a license :-(. --Steinsplitter (talk) 14:23, 29 June 2015 (UTC)

There is a general problem of bots and scripts which are widely used but not maintained properly (because the author is overstrained, tired, leaved the project or have some conflicts etc.). As regards ErfgoedBot, it is a very helpful tool but its development and maintenance appears to get freezed before years still. An example: ErfgoedBot updated daily a set of pages Commons:Monuments database/Unknown fields. However, its operator absolutely doesn't respond to reports about changes of field names in the affected templates. We cannot improve the templates nor correct their inaccuracies because WLM mill is not able to reflect the changes. I think, such important processes should be not so dependent on individuals and their fickleness. With all respect and gratitude for them. But I can't see inside where the problem springs from. --ŠJů (talk) 15:40, 1 July 2015 (UTC)

If someone writes a useful bot, and you find yourself running projects that rely on it, then you should feel free to ask the operator if they have shared the source code on github or similar. There have been several cases where we have tools and bots where this was never discussed and if a volunteer becomes seriously unwell, or is on a long wikibreak/seems retired for whatever reason, they are unlikely to find the time to cooperate. We should also encourage our most critical bots to have multiple operators and maintainers. This reduces the chance they will be unexpectedly lost and has the great benefit that the burden of maintenance is shared.
Lastly, as a fickle and tired bot operator I occasionally get thanked by users, but not very often. In fact far more often I get repeated complaints about things not working well, or chasing me about things that are out of my power to fix (like WMF server drop-outs!). If you use a bot, automated report or tool, and think to yourself how useful and neat it is, spend a few seconds by rewarding the operator with a barnstar to keep them sweet and engaged with the project. Bot operators are unpaid volunteers focused on the public good like everyone else. :-) -- (talk) 16:43, 1 July 2015 (UTC)
Yes - regrettably, the best and most important work is usually accepted as matter of course. We appreciate it just when we lose it, or when it get to be insufficient. Maybe, we should express thanks explicitly more often. However, when someone's work is widely used and have amounts of its fruit, that can be also a powerful source of satisfaction – that was also my way to express my gratitude for Multichill's (and other's) work on WLM. Maybe, improvement requests can be also perceived as expression of appreciation and thanks? I hope, some programmers will be capable to help Multichill with his work, or to take his relay, or to mediate related conflicts? It would be a pity to leave a monument database and lists to perish. --ŠJů (talk) 14:46, 2 July 2015 (UTC)

Allow WebP upload[edit]

Moved to Commons:Village pump/Proposals#Allow WebP upload

June 29[edit]

"taken with..."-Categories in the Categories of the camera manufacturers leads to heavy miscategorization[edit]

Hi, I see a problem concerning many categories below the Category:Photographs by camera manufacturer.
An instance: A ship in Taiwan file is Taken with Leica V-Lux 4, and therefore in the Category:Wetzlar, a town in Germany.


2013 we had a discussion in the german Commons-forum and User:H-stt told me to remove the meta categories from the Camera categories, but over the years this was reverted and so did User:Ghouston on 27th of June with all the "taken with Leica"-Categories, although I inserted the {{Cat see also}}-Template in all categories.

I know, it's a problem in principle and so I wanted to hear your opinion about a solution. I won't act as an editwarrior, but I must admit that I am very unsatisfied with this situation. Kind regards, --Emha (talk) 11:51, 29 June 2015 (UTC)

I still believe that the appropriate solution is to delete each and every "taken with"-cat. They mess up the category system and destroy valuable search functions. A more gentle solution would be to keep those cats but remove their top-cats from the cat branch of the camera-cats. That is because these cats categorize images, not cameras, which is why they must not be in the camera-cats. --h-stt !? 11:57, 29 June 2015 (UTC)
I think this is a misunderstanding of the category system. The relationship between a parent category and its subcategories is not necessarily "is a type of" or "located at". Thus a picture of a Leica camera is not necessarily taken in Wetzlar. Maybe the camera was made there, or maybe not. The pictures taken by the Leica cameras are then one more step removed. This happens throughout the category system. For example, what does File:Gravity escapement.gif have to do with Category:St Albans Cathedral? It's actually in a subcategory via Category:Burials at Saint Albans Cathedral, Hertfordshire and Category:Edmund Beckett Denison. --ghouston (talk) 12:17, 29 June 2015 (UTC)
That's actually a huge problem: We currently have no means to define the kind of relation between a (sub-) category and it's mother-category. The result is a chaotic network that next to impossible to grasp for any search function. COM:Structured data, though aimed at meta data other than content description in it's initial phase, could provide us with the necessary tools to define different types of relationships. --El Grafo (talk) 12:32, 29 June 2015 (UTC)
(Edit conflict) That's a general problem of our category system. There are numerous other examples that do not include "taken with" categories – I remember pictures of tennis matches showing up in the Category:Aviation subtree because they took place in a stadium that was named after someone who was related to someone who was a pilot (or similar). Movin the "taken with" categories to a separate subtree wouldn't solve the issue. We'd have to dissect the whole category system into single branches of "by location", "by author/manufacturer/…", "by date/time/period", "by color", "by taxon/type/series…", "by ID", etc. and use a search function that is capable to do proper intersections. I'd say it's impossible to do a conversion like that on the fly, but conversion to COM:Structured data looks like the way to do it. Then we could finally get rid of stupid constructions like Category:Women by country by century by occupation as well. --El Grafo (talk) 12:24, 29 June 2015 (UTC)
Sorry about the conflict. Yes, there are really multiple categorisation schemes that are independent, such as subjects, dates and locations. Combining them into ad-hoc intersection categories gives weird things like the ones I mentioned at Commons:Village_pump/Archive/2013/10#Category_madness_in_Hong_Kong. --ghouston (talk) 12:33, 29 June 2015 (UTC)
(Edit conflict) Ghouston, have you ever read Commons:Categories? The branch Category:Image sources is different from the Category:Topics, the how was the image made is not the same as what does the image show.
To stay at the given instance: This page lists all new files uploaded in the Category:Mittelhessen (including Category:Wetzlar). It doesn't work anymore if a ship in Taiwan is listed there.
Please don't argue with other miscategorized files, I know that there are many, but I seek a solution for the given ones. Regards, --Emha (talk) 12:54, 29 June 2015 (UTC)
Well, image sources are where the images were taken from, and it seems a bit different to how images were made. Perhaps you could argue that "hidden" categories are only for maintenance and shouldn't be part of the Topics category tree. However there's no policy for that, and hidden categories like Category:People by name fit into the topical tree without any problem. It's also logical that the "taken with" categories are associated with the cameras they refer to. My example of the Gravity escapement image was not to demonstrate miscategorization but to suggest that the category system varies in how categories are related to their subcategories. --ghouston (talk) 13:05, 29 June 2015 (UTC)
I don't know anything about User:OgreBot/gallery. Perhaps there is some way (or should be some way) of excluding particular subcategories. I don't think you'd want anything in Category:Leica included. --ghouston (talk) 13:10, 29 June 2015 (UTC)
Note: Commons:Categories#Major categories mentions Category:Image sources as a root category, which is wrong: it's a sub-category of the root categories Category:Media types and Category:Media by source. I'll change Commons:Categories#Major categories to reflect reality (put Category:Media by source in the place of Category:Image sources). Doesn't have to do anything with this discussion, just noting it here to avoid confusion. --El Grafo (talk) 13:33, 29 June 2015 (UTC)
(Edit conflict) User:Ghouston, the technical source of the named images is the camera, what else?
It doesn't make sense of dealing with the symptoms if the cause isn't removed.
Sure I want the Leica-Category and her content in this category, it's a company based in Wetzlar with a rich history including all the cameras that are produced here, but not the images of the cameras, because they are from elsewhere... Regards, --Emha (talk) 13:45, 29 June 2015 (UTC)
But then what about similar things like Category:Photographs by lens or Category:Photographs by film? "source" doesn't really make sense there and it doesn't really fit the description at Commons:Categories. If anything, I'd rather have a new root category for stuff like that. --El Grafo (talk) 13:52, 29 June 2015 (UTC)
The cause is varying relationships between categories and their subcategories, so that the relationship of a category with its grandparent etc., categories is unclear. A big part of it in this case is caused by putting the whole Leica company under Wetzlar. The relationship is "based in", I suppose meaning that its head office is located there. However Leica now makes at least some of its cameras in China [1] and it has shops and offices in other countries [2]. --ghouston (talk) 22:49, 29 June 2015 (UTC)
Yes, there are several root categories or entry points. But I don't think it's possible to separate them entirely. Simple example: Category:Photographs is in the root categories
--El Grafo (talk) 13:49, 29 June 2015 (UTC)
Yes, it is a problem with understanding our category system, but it is not me who misunderstands the categories. We use categories in a hierarchical tree. We don't use tags! That's actually all you need to understand about our categories and why this is a problem. --h-stt !? 12:30, 1 July 2015 (UTC)
@H-stt: I don't know what exactly it is, but the category system at Commons is certainly not a tree, as sub-categories can have multiple parent categories and it contains loops that (indirectly) make categories their own sub-categories. Commons:Categories calls it a en:Directed acyclic graph, but in reality even that is more like wishful thinking because of all the loops … --El Grafo (talk) 13:19, 1 July 2015 (UTC)
What is deeply troubling here is these two things:
  1. that we have an admin who not only doesn’t understand the underlying approach of our categorization scheme, but also comes to the Village Pump making a spectacle of it and demanding a “solution”, and
  2. that we have another admin who comes to the Village Pump to promote what would amount to vandalism were it perpetrated by an IP user («the appropriate solution is to delete each and every "taken with"-cat»).
-- Tuválkin 19:26, 29 June 2015 (UTC)
Tuvalkin: This is not the Administrators' noticeboard and not an admin-relevant cause. This is the village pump and I am not talking as an admin, but as a normal user like you.
I would be pleased if it would be possible to you not arguing personally but along the facts. And yes, I have a problem with the situation and would be glad if we could find a solution for this. If all the other users would have told me, that this is only my problem, that's okay for me, too. But as you can read, that's not the case. Regards, --Emha (talk) 09:37, 30 June 2015 (UTC)
Well, if we are looking for a solution for your gallery problem, it seems that User:OgreBot/gallery does have a facility for excluding subcategories. If you don't want to exclude Category:Leica, at least exclude Category:Leica cameras. --ghouston (talk) 23:37, 30 June 2015 (UTC)
User:Ghouston, thanks, but I know how to cure the symptoms but that's not the only concerned case and that doesn't heal the problem in principle. --Emha (talk) 08:10, 2 July 2015 (UTC)

@User:Emha: Categorization system of wiki projects is (since its beginning) designed as multifactorial, comprising many types of categorization relations to one interconnected modular complex. The main your problem is not the categorization system nor specific categories about cameras but your baseless assumptions about the system. The categorization tree here can be never supposed as a simple hyponymic hierarchy, nor a simple meronymic hierarchy. If we would extract pure hyponymic or pure meronymic threads from the categorization structure, wee need to mark a type of relation to every one category relation and filter such threads using these additional marks. --ŠJů (talk) 13:31, 1 July 2015 (UTC)

User talk:ŠJů: baseless assumptions? I read on Commons:Categories: The main principles are: Hierarchic principle.. And a ship in Taiwan is not situated in the city of Wetzlar. It's that simple. If you don't agree to this you have to discuss about changing this official policy of Commons. --Emha (talk) 08:10, 2 July 2015 (UTC)
Multi-hierarchic principle. Not only simple location hierarchy principle. The hierarchy is compound and incorporates many other kinds of relations than only current location. Categorization at wiki projects was allways multifactorial and allways supposed to combine more and very different types of relations. If it uses also relations as "originate from", "registered in", "produced by/in" etc., then, logically, we cannot suppose that every categorization thread will correspond with immediate spatial meronymy. If you don't agree to this you have to discuss about changing the basic principles of all Wikimedia wiki projects, or rather to create your own project which would be not based on them. There exist two ways to comply with your assumptions: to choose a simple-criterion system instead of the current all-embracing complex categorization, or to make the current system more sophisticated (using a thread filter by kind of relation). However, if you want to propose an alternative system, you should well understand the current system at first. --ŠJů (talk) 14:14, 2 July 2015 (UTC)

Electronics Categories[edit]

This problem is happening a lot with consumer electronics images, something should be done about this problem. From the friendly, Doorknob 747 (talk) 21:57, 5 July 2015 (UTC) :D

Computer Categories[edit]

Same problem is happening, a simple computer model will lead u to the country where the computer manufacture headquarters are located in the categories. From the friendly, Doorknob 747 (talk) 21:57, 5 July 2015 (UTC) :D

Scanner Categories[edit]

Laying aside the obvious matter that not all category relations are is-a relations, I still have a problem with some of this "taken with" stuff. For example,,_1954.jpg&curid=30755448&diff=164488000&oldid=149609367, as discussed in part at User_talk:BotAdventures#Scanner.... It seems pretty silly to say that a 1954 photo was "taken with" an Epson scanner. - Jmabel ! talk 19:49, 29 June 2015 (UTC)

Well, that should obviously be something like Photographs digitized with … and separate from photographs taken using a scanner as a camera … --El Grafo (talk) 19:57, 29 June 2015 (UTC)
You can digitize books, digitize photos, and image/digitize 3D objects with either a camera or a scanner. I'm not sure that creating separate categories for all these possibilities for every type of device is a good idea. I think there should be device categories (which will later be converted to Wikidata-like properties if Structured data ever happens) which can be assigned automatically from Exif in many cases. A single file can potentially have more than one device, e.g., the original camera, the scanner, or software that processes it (e.g., for enhancements or processing raw camera images), although sometimes only one can found in Exif. Any other information, e.g., that it's a book digitization, or a photo of a 3D object made by a scanner, can be added separately instead of creating intersections with the device type. --ghouston (talk) 23:02, 29 June 2015 (UTC)

Template:PD-NL-Gov/en depreciated or not ?[edit]

At Template talk:PD-NL-Gov/en is a 'kept' tag, while at Commons:Copyright_tags#Non-U.S._works under Netherlands it says it is depreciated. On Template:PD-NL-Gov/en it says: Images using the PD-NL-Gov tag need to be relicensed or deleted, as appropriate. To me it appears that if after a deletion request discussion it is decided it is 'kept', this would mean the template isn't depreciated. Do I understand this correctly ? Regards, --OSeveno (talk) 13:39, 29 June 2015 (UTC)

@OSeveno: They are not the same.. a template can be 'deprecated' (meaning it's no longer intended to be used) but 'kept' (because it has an unwieldy number of existing uses that need to be fixed, and deletion of the template would remove the ability to track them). In the case of this particular template, it was actually deleted after the original discussion, but then recreated later that same day as a 'deprecated' template, presumably after it was pointed out that it was in use. The 'kept' is misleading. As it stands now, the template is not intended to be used (but is linked quite a few times from old discussion). No files use it, and none should. Revent (talk) 16:10, 29 June 2015 (UTC)
(as a further note) The 'kept' on the talk page was added by a bot, when it noticed that the (recreated) page still existed after the discussion was closed. Revent (talk) 16:12, 29 June 2015 (UTC)
Thanks for clarifying that. Regards, --OSeveno (talk) 14:33, 2 July 2015 (UTC)


The Commons logo in the corner is blurry. Crisper versions are available.

SVG PNG PNG used in the corner

The PNG version of the logo also has other issues: a bump right where the outer ring turns into the arrow, and variable spacing b/t the arrow tips and the red circle. But the fuzziness is most noticeable. Is the simplest solution just replacing it with a crisper PNG? --SJ+ 06:07, 30 June 2015 (UTC)

It is suggested to use the .svg logo, not the png. --Steinsplitter (talk) 10:11, 30 June 2015 (UTC)
@Sj, Steinsplitter:The best was to resolve this problem is to convert the crisper .png, to a .svg image, "Convert. That was easy." From the friendly, Doorknob 747 (talk) 22:19, 30 June 2015 (UTC) :D
✓ fixed. -- Tuválkin 21:50, 1 July 2015 (UTC)

eBay upload script wanted[edit]

1909 postcard

From time to time, I find out-of-copyright images, such as old postcards, on eBay. There's one above. It would be handy to have a tool or script that will do the upload, capturing the appropriate metadata, like we have for Flickr. Does anyone know of one? Andy Mabbett (talk) 12:51, 30 June 2015 (UTC)

Not as far as I know. eBay has their own API, however as users can upload many photographs of objects, auction entries may be for multiple items and most sections of the auction page can be used in varied custom fashions by the seller, I think useful automation is a bit of a nightmare. Were I designing a tool, I would provide a webform, harvest details and then let the user fill in the blanks or correct details (like selecting which image is the relevant one to upload from a drop-list). All in all, it's a weak scenario compared to Flickr, where the API is explicitly returning fields about an image you want to upload, not information about what it may represent, tangential events, or as part of a group object.
P.S. Flickr upload tools are a bit of a hazard for mass uploads, however I'm reading your request as a tool for item by item uploads, not mass uploads. -- (talk) 13:06, 30 June 2015 (UTC)
Yes; item by item. A pre-filled form would be fine - anything that removes a step or two from the drudgery. Andy Mabbett (talk) 13:54, 30 June 2015 (UTC)
@: Idea: would you consider automatically submitting the eBay page to one of the web citation websites when your form is used? Eventually, the eBay link goes away and sometimes the licensing is questionable without being able to see it. Example: if a postcard is uploaded with the claim that it was published pre-1978 with no copyright notice and all we have is the front of the postcard, it's difficult to know that the license is valid because we can't see the back to verify that there is no copyright notice on there. But if the eBay link was preserved, then we could see the other side of the postcard and verify that there was no notice ... or we could see that the seller never provided a copy of the other side of the postcard, in which case it should be nominated for deletion. --B (talk) 15:24, 1 July 2015 (UTC)
Yes, auto-archiving the page would be possible, within whatever usage limits there are for the service. It may mean a bit more cookie-fiddling and a wrapped-in or popped-up human-check or similar jiggery pokery.
Sorry if my text above might be misleading, but I do not have the volunteer time to do this as a project, my real life commitments are likely to be a bit overwhelming for several more months (believe me, I would *much* rather be playing code monkey than handle the depressing stuff I'm going through). Anyway, though I can do this sort of API experiment, I'm actually rather hacky/bad and slow at it, and I have a big guilty secret backlog of projects I ought to return to and finish off. My description above was how I would initially think about it, with the hope that someone faster and smarter would have a crack at it. :-) -- (talk) 15:34, 1 July 2015 (UTC)

User:We hope has a personal method of uploading a preliminary image which shows evidence for lack of a copyright statement, then overwriting it with another image with only shows the area of real interest... AnonMoos (talk) 22:22, 5 July 2015 (UTC)

Tool for changing the descriptions of multiple images[edit]

Hello, is there by any chance a tool, similar to cat-a-lot, with which I can change the description of multiple images at once? That would be very helpful for Wiki Loves Earth, because sometimes there are 10 different images from the same place. Thanks in advance, --Лорд Бъмбъри (talk) 17:09, 30 June 2015 (UTC).

You could possibly use COM:VFC for the purpose, if you are simply replacing one block to text with another. Revent (talk) 19:12, 30 June 2015 (UTC)
@Revent:How can your utc time stamp for June 30, be 19 hour, when my UTC Wikmedia gadget displays 18 hour still. where did the hour go? Did I go back in time. Places check mark on discoveries to make "go back in time. Check". Doorknob 747 (talk) 22:23, 30 June 2015 (UTC) :D
@Doorknob747: It's a mystery, lol. My timestamp (just checked it) matches both the UTC time gadget, and a quick check against a network time server. Revent (talk) 03:10, 1 July 2015 (UTC)
@Revent: Sure is a mystery, now its 4 hours behind, that is after 5 days, its like as if the clock is slowing down, in the gadget on the top right corner of the page. Should I upload a screen shot image of the problem and add it on this page for people to see the problem? From the friendly, Doorknob 747 (talk) 21:51, 5 July 2015 (UTC) :D
@Doorknob747: I don'y know exactly how that gadget works, but I would assume it's based on local javascript... try clearing cache and see if that fixes it. Revent (talk) 22:30, 5 July 2015 (UTC)

Labadie Collection, University of Michigan[edit]

Labadie Collection, University of Michigan. A good deal of this should be in the public domain. - Jmabel ! talk 23:36, 30 June 2015 (UTC)

Notification of DMCA takedown demand - Baffin Island fjord[edit]

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

Affected file(s):

To discuss this DMCA takedown, please go to COM:DMCA#Baffin Island fjord Thank you! Jalexander--WMF 23:47, 30 June 2015 (UTC)

July 01[edit]

NoScript warning when trying to view full size photo[edit]

When I clicked on this photo I got a NoScript warning in Firefox "The resource could not be found. File not found: /v1/AUTH_mw/wikipedia-commons-local-public.2f/2/2f/Ford_A9745_NLGRF_photo_contact_sheet_%201976-05-11%20%20Gerald_Ford_Library%20.jpg ". With other images no problem. What is the cause? Wouter (talk) 05:30, 1 July 2015 (UTC)

Confirmed, Noscript in Firefox gives "NoScript filtered a potential cross-site scripting (XSS) attempt from" and a 404 page. Selecting "unsafe reload" to bypass XSS protection from NoScript options shows the image normally. Firefox 38.0.5, NoScript, Win7/x64. Note that is on my NoScript whitelist. MKFI (talk) 06:23, 1 July 2015 (UTC)
It sounds like noscript is replacing brackets in the image name with spaces, which unsurprisingly causes the image not to be found. I have no idea why noscript would do that. Bawolff (talk) 08:00, 1 July 2015 (UTC)
We had this with a custom gadget on de:Wikipedia:Technik/Archiv/2013#NoScript_meldet_Cross-Site-Script-Versuch in 2013. -- Rillke(q?) 08:11, 1 July 2015 (UTC)

Study reports that Wikimedia Commons media has large financial value[edit]

Blue Rasberry (talk) 18:02, 1 July 2015 (UTC)

Bluerasberry As with pretty much all "scientific" studies of Wikipedia, this one is complete rot. I strongly advice nobody on Wikimedia give this report any oxygen. The authors build a case by assuming many things that are patently untrue or seriously unlikely. Given that the paper is about "Public Domain", it is startling that their definition of this is completely wrong. They include CC-licensed works, saying "For the purposes of this paper, therefore, we include works which may be freely used under a Creative Commons license, even though in many cases the author technically retains title". There's no "technically" or "in many cases" about who retains "title". Claiming that works with a "free" licence are public domain, and that credit of authorship is merely a technicality is stupendously ignorant and disrespectful. They make further assumptions about how much Google increases traffic flow if a web page contains in image and how much a Wikipedia page is worth in terms of advertising revenue (I kid you not). At one point, when discussing novels becoming public domain and therefore sold at a cheaper price, they completely confuse "publisher" with "copyright owner" (i.e., author). I can only assume that Harvard Journal of Law & Technology is where C-grade term papers go to be archived, and shouldn't be confused with a serious academic publication. Sigh. -- Colin (talk) 18:44, 1 July 2015 (UTC)
Colin The paper at least demonstrates public interest in the value of Wikimedia content. If it is wrong then it gives the next researcher something to challenge. Blue Rasberry (talk) 19:18, 1 July 2015 (UTC)
When I read the list of assumptions in the paper half of them did not seem reasonable, so just like Colin I was taking the findings with a grain of salt; however I think it is a good start and the numbers are likely to be within of the right order of magnitude. Hopefully next iteration will get it closer. I always had a problem with estimations calculated by assuming dozen numbers with some huge or unknowable error bars and then multiplying them together to get your estimate. I am mostly interested in those error bars because that is what will tell me if the real value is likely to be off by 1%, 100% or 1000%. --Jarekt (talk) 17:10, 2 July 2015 (UTC)
Jarekt, I don't believe the figures are anywhere near the right order of magnitude because they ask us to imagine a world without PD and a world where Wikipedia is not a free content project that attracts freely licensed images and a world where Wikipedia's budget could pay for $250 million of photos per year (which is about five times WMF's entire annual budget). Their $250 million is naively worked out by taking the 2 million pages with free photos and multiplying that with the cost that Getty/Corbis charge to licence a single photo on a website for a year ($105-$117). Now I don't know where they get that licence figure from -- it often varies by the size of image you are licensing and quite possibly also by other factors like age or quality. The thing is that nobody who needs to licence 2 million images would pay the same unit cost as someone who licenses one. The incompetence of this economic mistake is staggering. (Getty own 80 million still images and annual revenue is $869 million -- there's no way 2 million thumbnail images licensed to a non-profit would bring in $250 million) And many images are sold for a great deal more than that for higher value purposes such as advertising. A world without public domain images or freely licensed images would much more heavily use photo agencies, so market forces means the price per image would also be less. And a Wikipedia that had to pay for images would undoubtedly licence 240px thumbnails rather than the multi-megapixel images that Commons actually hosts. Given the big stock photo corporations recent behaviour, I can well imagine them coming to a deal with Wikipedia that lets them use the small thumbnails for free, in return for a link on the file-description page that allowed viewers to see the high-resolution photo for a fee or to license it themselves. So in this world, without Commons, the cost of the images that Wikipedia uses is zero. $0. No error bars or orders of magnitude. Wikipedia pays nothing for the images. But there is no Commons for anyone else to use high resolution images. And you have to pay to look at any image in detail. How does someone put a price on how shit that world would be? -- Colin (talk) 18:57, 2 July 2015 (UTC)
If Commons didn't exist, most people would probably put their free images on some other site. There are already quite a lot on Flickr. --ghouston (talk) 23:20, 2 July 2015 (UTC)
Although some images from people who just want to put an image on a Wikipedia page, without caring much about license, probably wouldn't be CC licensed otherwise. --ghouston (talk) 23:25, 2 July 2015 (UTC)
--ghouston, they are imagining a world without free image content at all. So not just "no Commons" but "no CC anywhere". They aren't particularly interested in Wikipedia/Commons but in trying to put a value on the "public domain" (even though they completely misdefine it). They are only using Wikipedia as it is a publicly available source of information to mine. The two most disappointing things about this are that it is yet another study of Wikipedia / Free Content Project where the authors don't really understand the domain and that such a dire C-grade effort is being published at all. Have they never heard of Getty Embed -- 50 million images freely usable. What is most sad really, is the ideology that one needs to (or it even makes sense to) put a monetary value on things that are in the community (public domain images) or that are donated to the community (freely licensed images). -- Colin (talk) 07:31, 3 July 2015 (UTC)
Personally I think that study is overrated. We can see how much this site would be worth by looking at how much Wikia makes. Sure we get a bit more traffic, so add about 10 or 15% but its no where near the hundreds of millions they say in the study and we would lose a lot as well. If this site were for profit then that means the donations would stop and it would be taxed on what it makes. I heard that the WMF gets a little bit of a discount on IT services from the data warehouse they are using due to its ta free status so that would be lost too. A lot of the volunteers would leave and would need to be replaced by paid folks. So to me the whole study makes assumptions of things being equal when they would be far from equal in reality. Reguyla (talk) 00:30, 3 July 2015 (UTC)
Way back in the day there were several Spanish-speaking Wikipedians who forked and created en:Enciclopedia Libre to protest a proposal to make Wikipedia into a for-profit venture. We don't want that repeated. We can't do what the study proposes. WhisperToMe (talk) 08:38, 4 July 2015 (UTC)

Preview shows an image wrongly rotated[edit]

I found that image that is strangely converted by image preview (but is correct in thumbnails): File:Gyaru-ikebukuro-aug2007.jpg You can see the effect on Category:Gyaru if you click on the image (Gyaru-ikebukuro-aug2007.jpg), it shows up rotated but with the original dimensions, so it's weirdly squeezed. I checked the Exif and there's no rotation, orientation is top-left as usual. Might be worth checking out? SilverbackNettalk 19:22, 1 July 2015 (UTC)

The file has been rotated after upload but thumbnails seem to have not been updated properly. Purged. MKFI (talk) 06:22, 2 July 2015 (UTC)
I don't see anything wrong, except the dimension was changed from 650px → 640px with 0° rotation. ↔ User: Perhelion (Commons: = crap?) 17:57, 2 July 2015 (UTC)
@Steinsplitter: What do you mean? User: Perhelion (Commons: = crap?)  06:43, 3 July 2015 (UTC)
EXIF-Orientation has been fixed. Please note that MW is automatically rotating files based on EXIF-Oriantation. In raw-modus you see the the bot has worked correctly. --Steinsplitter (talk) 10:41, 3 July 2015 (UTC)

July 03[edit]

Batch upload advice[edit]

Hi there. I'm planning on tackling the batch upload request Commons:Batch uploading/Freshwater and Marine Image Bank. I'm looking for some advice on how to display various information from the files in the Information box. I moved my questions from here to the Batch Uploading page above. I appreciate any help or advice!

BMacZero (talk) 02:52, 3 July 2015 (UTC)

updated information suggest file name includes error[edit]

How do you fix a file name that communicates erroneous information? See [3]. --Smkolins (talk) 11:04, 3 July 2015 (UTC)

Have a look at Commons:File renaming, --Pristurus (talk) 11:14, 3 July 2015 (UTC)
Thanks --Smkolins (talk) 11:35, 3 July 2015 (UTC)

TIFF from LoC[edit]

Hi, On every file available from the Library of Congress, there are a 8-bit and a 16-bit TIFF files (e.g. [4]). I usually upload the TIFF (by URL) before uploading a JPEG version because downloading from WM servers is much faster (x10) than from LoC servers. ;o) The TIFF much be converted to JPEG (8-bit), so is there any interest to upload the 16-bit TIFF file? I do the conversion with Gimp, which only handles 8-bit. The 16-bit file is always available on LoC. So? Regards, Yann (talk) 12:53, 3 July 2015 (UTC)

How about just uploading tiffs? --Jarekt (talk) 14:12, 3 July 2015 (UTC)
@Jarekt: You mean not uploading JPEG? Or uploading both 8-bit and 16-bit files? Also noteworthy is that the TIFF files are Adobe RGB. So I convert them to sRGB before uploading JPEG here. Regards, Yann (talk) 14:23, 3 July 2015 (UTC)
I've actually recently come across another similar with LoC images.... from bulk uploads by Fae (not that I am faulting him, at all)... Photochrom images that were available from the Library of Congress Prints and Photographs Division directly in full-resolution as TIFF and JPEG, but also available (and uploaded separately) as lower resolution PNGs from the World Digital Library (which is a project run by the LoC). I'm tempted to nominated the lower res PNGs as duplicates, except that they are from a different 'exact' source, and not the same format. Examples:
As you can see, the World Digital Library PNG is far lower res. Revent (talk) 03:18, 5 July 2015 (UTC)
I'd say upload 16-bit if there is any useful information that would be lost by conversion to 8-bit, which would usually be the case. (Images with a substantial amount of noise wouldn't benefit from 16-bit. Most other scanned images would, notably anything with a compressed dynamic range, 16-bit HDR.) Anyone downloading the tiff is looking for maximum quality, while a JPEG should be "ready to print" or "ready to display" with full contrast and no HDR information. SilverbackNettalk 07:58, 5 July 2015 (UTC)
As I said above, the bigger TIFF is always available on LoC. The question is: "is there any use for it on Commons?" Yann (talk) 13:47, 6 July 2015 (UTC)
We don't know that it will always be available on the LoC. One scholar working on the Manhattan Project was complaining about scores of publicly available documents whose websites disappeared or stopped working, or in one case were corrupted by an ill-advised line-feed conversion. Not intentional, but there wasn't the budget and/or will to fix it. Maybe the LoC as library will be better, but one change and they could be made inaccessible indefinitely.--Prosfilaes (talk) 19:27, 6 July 2015 (UTC)

Source information incorrect on an image[edit]

I was working on some images today and noticed that the source information on File:Don_J._Jenkins.jpg is incorrect. We have the description that this person is Bryan Battaglia, Senior Enlisted Advisor to the Chairman of the Joint Chiefs of Staff but its actually Don Jenkins the medal of Honor recipient. I noticed this because I know what they both look like in person and I knew this was not a picture of Bryan Battaglia. Otherwise I probably wouldn't have caught this. When I looked at the source for the photo it is also incorrect. My question here is, what are our procedures for correcting something like this? I can contact the DoD source and get that fixed, but I'm not sure how we would change it here to reflect the correct information. Reguyla (talk) 16:28, 3 July 2015 (UTC)

  • We'd change it by editing, just like anything else that gets corrected. I'm not sure what you think is difficult, so I can't give you anything more specific. - Jmabel ! talk 16:44, 3 July 2015 (UTC)
    • Sorry for the misunderstanding, it wasn't a matter of not knowing how, I just wanted to make sure that it was appropriate since any change would not match the supporting source link. Reguyla (talk) 17:18, 3 July 2015 (UTC)
      • Presumably you will have some source for your change. Cite it. - Jmabel ! talk 19:57, 3 July 2015 (UTC)
        • Not for this one I don't other than anyone can look at the corresponding articles on ENWP and see that the image is not of Battaglia but of Jenkins. The group of images on Flickr all seem to refer to Battaglia even in the ones he isn't in and several have the exact same description, so I wanted to ask what should be done about that. Should we leave the incorrect description that was pulled in from the source, should we change it to reflect something more accurate or should we just leave it and maybe put a history statement or note saying that its wrong, but the description is actually X. Reguyla (talk) 20:55, 3 July 2015 (UTC)
          • I'd just replace the original description with a correct one. It will be easily verified as correct if there are other photos in Commons of the same person, or he's easily identified in a web image search. --ghouston (talk) 01:24, 4 July 2015 (UTC)
          • I think the photo shows Gary L. Littrell. Mrwojo (talk) 14:56, 4 July 2015 (UTC)
            • I'd include "probably" or "possibly" in the description if there was any room for doubt. You could also put "possibly Don J. Jenkins or Gary L. Littrell". Other sources may list who was at the event. --ghouston (talk) 02:24, 5 July 2015 (UTC)

NASA picture[edit]

How can I tell the copyright on this amazing photo of a Chrysler turbine engine? I would love to upload it, but I cannot seem to determine what its copyright might be? Thanks for assistance, mr.choppers (talk)-en- 01:48, 4 July 2015 (UTC)

It says the image was created by NASA so license tag should be {{PD-USGov-NASA}} Offnfopt(talk) 02:22, 4 July 2015 (UTC)

July 04[edit]

File transfer[edit]

I would like to use this image on the English Wikipedia for the page however since the file is hosted on the Hungarian Wikipedia (and I don't speak Hungarian) would it be possible for this to be transferred to the Commons? The C of E (talk) 12:53, 4 July 2015 (UTC)

Yes, you can load a copy of it here as public domain. The engraving was created/published in 1778 (based on an original in the UK National Portrait Gallery) and faithful reproductions of it are considered to have no additional copyright. It is possible to use bots to arrange transfer, but it's quicker for you to download and re-upload to Commons, just use the standard upload link and wizard. -- (talk) 13:31, 4 July 2015 (UTC)

User talk:Jameslwoodward protection[edit]

Are protections like that allowed? It looks like kinda disrespect to common communication practices from my point of view. --Base (talk) 17:03, 4 July 2015 (UTC)

I can understand, but semi-protection should be sufficient. Also a link to COM:AN would be useful. @Jameslwoodward: Jim? Regards, Yann (talk) 17:22, 4 July 2015 (UTC)
@Base, what is so hard to understand with "I will be on vacation, away from the Internet, through the first week in August"? If he hadn't protected his talkpage, then after 3 or 4 postings, the next poster wouldn't notice Jim's "absence message" and might become angry when he gets no answer. --Túrelio (talk) 19:05, 4 July 2015 (UTC)
Well in ukwiki we had a user who won't reply if you forgot to greet him when writing to him. He referred to need of respect but in that way actually disrespected community members who could just forget to say "Hi" because were thinking about other, more important atm stuff. It looks like a case of the same kind in some way. Are you really sure that a poster who doesn't get answer and fails to read the notice (though with some CSS you can do that a block of text sticks to the screen which probably solves the possible problem you mentioned in much better way than protection, IMHO) will be more angry than a would-be-poster who will not be able even post because of the protection? For example, in my case of TP use practice I'd rather write something on TP and get answer in several months because of the addressee vacation rather than I'm stopped from writing now and forget the thing I wanted to say in the months. Well not that I'm a typical case but I'm not too peculiar to neglect. (Just to clarify I didn't have anything to write the user now, just saw the action in WL and it drew my attention.) --Base (talk) 21:15, 4 July 2015 (UTC)
Jim should be aware that he won't get deletion notifications. -- Rillke(q?) 19:20, 4 July 2015 (UTC)
@Base: The answer is unambiguously no, Jim went against Commons:Protection policy which limits page protection of this type to "vandalism and edit wars". -- (talk) 19:45, 4 July 2015 (UTC)
Túrelio has my reason exactly -- in the past I have been away for much shorter periods and have had angry messages because I did not respond. Of course an experienced user would look at my contributions and see that I had not been on Commons since before he left the message, but novices just get mad. Therefore, I thought that it would be better to prevent novices from leaving messages.
However, I'll be perfectly happy with whatever Yann and Turelio decide. BTW, Rillke, I am well aware that I won't get deletion notifications, but I've never had one and I think I can safely say that my present 500 or 600 uploads are all safe from valid DRs. And now, I really am gone until August. .     Jim . . . . (Jameslwoodward) (talk to me) 23:21, 4 July 2015 (UTC)
Talk page protection is unnecessary as many people including me watching his page. If something unexpected happens, we can manage it (like replying to an angry novice). I think Yann or Túrelio can unprotect it now. Jee 03:33, 5 July 2015 (UTC)
I changed it to semi-protection. Regards, Yann (talk) 11:27, 6 July 2015 (UTC)

July 05[edit]

Central American / South American photos[edit]

I need help and have tried other avenues but am stymied. The Copyright law of Brazil article tells me nothing and I cannot access the source. There is no Copyright law of Guatemala article, Nicaragua, Uruguay, etc. The Media Copyright Questions page suggested I ask here. So, here is the situation. I wrote an article on Primer Congreso Interamericano de Mujeres and I would like to do a strip of photographs like were done on this article: Pan-American Conference of Women.

If anyone on here would like to help with this project, photos of any of the women in the article would be great. I asked a couple of editors on commons and no one knew the answers and suggested I post here. Some I am looking for specifically are: the very first Mexican movie star (Elena Sánchez Valenzuela); an MP in Canada (Beatrice Brigden); the first activist for disabled person's education (Eloísa García Etchegorken), Uruguayan; the first female lawyer in Nicaragua (Olga Núñez Abaunza); Sue Bailey Thurman. Any help would be greatly appreciated. I also am not sure how I get a response to this but if anyone is able to help, can you please notify me on my talk page on Wikipedia? Thank you SusunW (talk) 17:02, 5 July 2015 (UTC)
I found a photograph for the Canadian (Beatrice Brigden), which I believe was taken around the time of the conference or earlier, based on the clothing, hair style, etc. and the fact that I found a video of her from the 1970s. The poster gives his source as the Archives of Manitoba, but I was not able to find the photo there. [6] SusunW (talk) 23:50, 6 July 2015 (UTC)

Problem with UTC time gadget.[edit]

I feel the UTC-Time gadget is sending some users 4 hours back in time, according to the gadget. It can bee seen in this screen shot image I took, and compare it with the UTC time of the most recent edit with the time of the gadget in the top right corner of the page.

Problem with time gadget

But, aside from the send back in time joke, can someone fix this problem please. Thank you in advance. From the very friendly, Doorknob 747 (talk) 22:16, 5 July 2015 (UTC) :D

It's just a bit of JavaScript[7] that gets the current time from your own computer. It's working for me (currently 23:18:48). I guess you already checked that the time on your computer is right. --ghouston (talk) 23:18, 5 July 2015 (UTC)
What is your local time offset from UTC by the way? Is it related at all to the 4 hours? --ghouston (talk) 23:19, 5 July 2015 (UTC)
Then they should not even name it a UTC-Time gadget and call it a local time gadget. It can confuse someone. Also, another thing about this gadget, is that, there should be a option for 24-hour or a 12-hour mode. From, the very friendly Doorknob 747 (talk) 05:37, 7 July 2015 (UTC) :D

July 06[edit]

Replace picture with watermark[edit]

Hi, I have uploaded a new version of a file, without watermark. Can any administrator delete the two older versions (normal and not-interlaced). Here is the file, thanks a lot.

Is there any policy against pages with a large number of images displayed?[edit]

I'm just curious if there is any type of policy against pages that display a large number of images? I was looking at the "What links here" page for some of my uploads and came across pages similar to the "[[User:Josve05a/Possible DRs/2015 May]]" page. This page displays 2974 thumbnail images on the same page. This IMO is a waste of server-side and client-side resources. Offnfopt(talk) 06:31, 6 July 2015 (UTC)

There are a lot of such pages, and even some uploading software - as Commonist and Vicuna -routinely create user galleries, and as far as I know no problem has even arisen except for loading time. Dividing such pages into smaller pages wouldn't spare server resources. Probably making smaller galleries is a good practice, but not a mandatory rule to be enforced.--Pere prlpz (talk) 12:18, 6 July 2015 (UTC)
It's not a waste of client-side resources unless people are viewing them, at which the question might involved the client-side resources involved in splitting them up and having to deal with many pages instead of one. It's not a waste of server-side resources again unless they're being loaded, in which case, 2974 thumbnail loads is probably a drop in the bucket in the world of Wikimedia. It's all false optimization.--Prosfilaes (talk) 19:31, 6 July 2015 (UTC)
Parsing pages with that many thumbnails takes a while (So, there might be some latency on rendering the page). Thus I wouldn't recommend doing that for any page meant to be regularly viewed, as it will be slow to load for the user. Beyond that I wouldn't worry. Bawolff (talk) 01:24, 7 July 2015 (UTC)
I agree with User:Bawolff, it realm is not something that one should worry about. On, any Wiki there are pages with thousands of images. Also, if one does a search, they will see many images in one page, thus indicating that there is no need to set a limo to how many images can a specific page can have. Doing such a limit may have actually some adverse affects to the efficiency of this site even. From, the very friendly Doorknob 747 (talk) 05:26, 7 July 2015 (UTC) :D

Request for uploading a Flickr stream[edit]

Hi all, this is Scott (Russavia). After my request the African Union Mission to Somalia has kindly relicenced their Flickr photos from (C) ARR to CC-0! I have already uploaded one photo (and using externally for an educational purpose) and it can be found in Category:Photographs by the African Union Mission to Somalia. Would someone out there like to get a bot to do the rest of the stream? Heaps of valuable Somalian photos to be found. Once done, I'd recommend dropping a note over on Somalian Wikipedia and letting them now they are available. The Flickr stream is found at Cheers, 07:23, 6 July 2015 (UTC)

 Doing… --Zhuyifei1999 (talk) 08:17, 6 July 2015 (UTC)
Thanks Zhuyifei1999 for helping with this. 6,000+ photos from Somalia will be an awesome addition to Commons. I'll try to get some Somalian editors to help with categorising, etc. And hopefully they will find great use for them. Cheer, 11:07, 6 July 2015 (UTC)

World satellite map with political borders[edit]

Hi all. I am looking for a similar map to this one but also showing country borders. Is anyone aware that it exists in Commons? Thank you very much. Sentausa (talk) 08:54, 6 July 2015 (UTC)

File:Worldglobe.jpg, or file:Robinson world.png. Sincerely, Taketa (talk) 09:00, 6 July 2015 (UTC)
I saw File:Worldglobe.jpg before, but File:Whole world - land and oceans 12000.jpg looks much nicer to me. And unfortunately, file:Robinson world.png doesn't show the country names. So, apparently we don't have a version of File:Whole world - land and oceans 12000.jpg which shows the country borders, right? I wonder if I can ask for it be created in Commons? Thanks! Sentausa (talk) 13:01, 6 July 2015 (UTC)

SVGs as PNGs[edit]

Is it just me, or has the option for rendering SVG images as PNGs disappeared? On SVG file pages, there used to be a line of text saying something like "this image as a PNG", and then links to PNGs of varying sizes. But now I can't seem to find this. Zacwill16 (talk) 12:52, 6 July 2015 (UTC)

It's just you. On the description page for (e.g.) File:Bulletin Board with notes.svg the line says "Size of this preview: 800 × 568 pixels. Other resolutions: 320 × 227 pixels | 640 × 454 pixels | 1,024 × 727 pixels | 1,280 × 908 pixels." All links point to PNGs. --Magnus (talk) 12:59, 6 July 2015 (UTC)
That's not what I was talking about. Below that, there were options for 200px, 500px, 1000px and 2000px. Zacwill16 (talk) 13:25, 6 July 2015 (UTC)
Those were added by JS, not mediawiki. Maybe somebody disabled the gadget. Bawolff (talk) 01:21, 7 July 2015 (UTC)
Yes and no, this was (not a gadget) fully in the Common.js and was simply removed as deprecated because of redundant. User: Perhelion (Commons: = crap?)  01:33, 7 July 2015 (UTC)
I would like an entry box in the image description page so you can enter the desired width and then the new PNG is rendered in the new tab. This is easier than copying and pasting the PNG link to a new tab and then change the width value in the URL. Ideally there should be a PNG link with the same dimension as the SVG source. -- Sameboat - 同舟 (talk · contri.) 01:54, 7 July 2015 (UTC)

Policy clarification needed for images with {{OTRS pending}} but no license[edit]

Large fraction of images uploaded without any license use {{OTRS Pending}} instead. See for example here or here. For years I have been tagging such files with {{no license}} under assumption that COM:LIC requires all files to have a license template and uploaders should not be uploading new files unless they know under what license files are released, since the release might not be under license allowed on Commons. Unfortunately the COM:LIC policy does not mention OTRS, and COM:OTRS is more like guidelines instead of policy and it also does not deal with issue of WHEN is the license needed. Over the years I had a lot of discussions with confused uploaders, some new and some very experienced admins from other projects who were advising for years that license can be added after OTRS permission is processed. Lately User:JuTa was removing {{no license}} tags like here leading to this discussion. For me that is a sign that we do need clarifications to our current policies so we all follow the same rule book. As I see it our options are:

  1. Tag all images without a license with {{no license}}, even the ones that claim {{OTRS Pending}}. Such files will be often deleted after 7 days.
  2. Leave files alone and after 30 days {{OTRS Pending}} will begin displaying {{No permission since}} which will lead to deletion after additional 7 days. This option would cause files to remain without license for 37 days before it is deleted.
  3. Create some new processing pathway

--Jarekt (talk) 14:25, 6 July 2015 (UTC)

From a Commons policy perspective, and based on pragmatic experience as a OTRS volunteer and Commons contributor, I believe that Commons should adopt processes that disregard off-wiki correspondence about the possible copyright status of a file, until the point when a specific statement has been verified. Waiting for over a month for what may be damaging copyright violations to be removed, especially photographs of people or the recently deceased, is unacceptable. The OTRS pending template can be added by anyone, including those that may deliberately use this as a back-door for the Commons system to get images into high profile Wikipedia articles (noting that Wikipedians should be able to assume that images from Commons are at a more reliable free copyright status that on other projects). If anything we should focus on improving the undeletion work-flow and where there are associated claims of backlogged OTRS correspondence, these are flagged or escalated for attention if the file has been deleted or is about to be deleted.
We must encourage anyone to create reasonable deletion requests even when the OTRS pending template is added, and continue to be prepared to delete files after 7 days. Administrators are supported by policy in acting to speedily delete files where, in their view, copyright claims appear in any way suspect. Procedures should not leave an Administrator unable to act, just because they do not have OTRS access. -- (talk) 14:46, 6 July 2015 (UTC)
I do have a broad OTRS access, and I often search OTRS for records related to files with {{OTRS pending}} but no license before tagging them. In majority of the cases search for the name in the "author" field or search for the filename does not find any hits. As for the undeletion work-flow, many such requests, especially for files originally tagged by me, end up on my talk page and are promptly undeleted. --Jarekt (talk) 15:09, 6 July 2015 (UTC)
Sure it can be quick when all tools are in the same hands, the user is experienced enough to be relied on to apply their judgement and willing to be held to account afterwards. However we want to encourage more users to be active in chasing up possible copyvios, including non-admins who know enough about copyright to act, admins without OTRS access and OTRS volunteers who are not admins. Providing an easy workflow is key to gaining more helpers. (talk) 15:18, 6 July 2015 (UTC)
Hi, I think we need some common sense in these cases. I check if the claim of a permission is plausible. If yes, then waiting for about one month is fine for me. If it is just a file copied from the web without any real possibility to get a permission, then tagging them with "no license" or "no permission" is fine. Regards, Yann (talk) 19:27, 6 July 2015 (UTC)
But waiting for a month means that we are hosting a file with no license for a whole month before deleting it. I personally do not like it. Many people I interact with in last couple years while I was working with files with no license seem to have very little understanding of the concept of license. I have little confidence that many of them will provide proper OTRS permission form. --Jarekt (talk) 02:14, 7 July 2015 (UTC)

Cross post - can a sculptor CC license a photo of a traditionally copyrighted artwork?[edit]

I am seeking more opinions at Commons:Village_pump/Copyright#Sculptor_wishes_to_donated_images_of_copyrighted_sculpture. Blue Rasberry (talk) 14:44, 6 July 2015 (UTC)

cat redirect question (Category:Memorials)[edit]

Category:Memorials currently redirects to Category:Monuments and memorials in unidentified countries. I understand the choice -- it means that if someone adds the superseded Category:Memorials, things will be recategorized by a bot into a place appropriate to future work -- but it seems a bit bizarre. After all, it will put that "unidentified countries" category on something where the category may be perfectly well identified by another category or by the description. Shouldn't this just redirect to Category:Monuments and memorials? - Jmabel ! talk 18:51, 6 July 2015 (UTC)

I agree with your assessment. Sure, stuff shouldn't be put directly into Category:Memorials, but that is easy to fix. The redirect is just weird. --Sebari (talk) 19:36, 6 July 2015 (UTC)
Yes, if you type "Memorials" into the search box, you get this redirect. I suggest reverting the last edit by User:Butko. --ghouston (talk) 23:12, 6 July 2015 (UTC)
It actually is not right even, its like saying why even give a name to a memorial, if people are only going to call it a memorial. Even the 9/11 memorial is redirecting to memorial. From, the very friendly Doorknob 747 (talk) 05:29, 7 July 2015 (UTC) :D
OK, changed to redirect to Category:Monuments and memorials. - Jmabel ! talk 05:47, 7 July 2015 (UTC)

Lossless files on Commons[edit]

The FP I uploaded which caused the conundrum. 4,301 × 6,700, 48.8 MB
I'm breaking from my guide here and displaying a png on a readable page, but I feel it's justified here.

When I was working on some FP-candidates I got a few questions about why I had uploaded the files in png. They are original scans in TIFF format at a high resolution. I wasn't able to find any proper guide for when to use lossless versus lossy files on commons, so I've created one. For many this is second nature, but I think enough people will find it useful so I'm posting my work in progress here. Please judge all you want and try to give me feedback. I'll do my best to adapt the guide and see what can be improved or if it should be merged somewhere else.

Note: It is still lacking a number of sections that I wasn't able to just write of the top off my head. I'll be researching these and updating it.

Best, -- CFCF (talk) 21:31, 6 July 2015 (UTC)

In regards to "Lossless files should not be used in articles such as on Wikipedia. If a file only exists in a lossless version a new lossy version should be uploaded and used. This is in order to decrease load on the Wikimedia servers, both bandwidth and processing power used for rendering. Additionally it is better for the end user, and especially so when the end user has limited bandwidth capacity." - Tiffs thumbnail as jpegs so the bandwidth argument doesn't apply to them. You should not worry about processing power required generally speaking, we have servers so that they can be used (What takes more or less processing power isn't exactly straightforward either, I would recommend against doing anything because of processing power required, unless you are actually measuring the processing power or you are encountering some sort of limit like the large gif images don't thumbnail limit) [Obviously if you do notice something that's unambiguously exploding the servers, you shouldn't do it]. Bawolff (talk) 01:20, 7 July 2015 (UTC)
Okay, I didn't know TIFF thumbnailed as jpeg. Weren't there a number of issues concerning TIFF thumbnail generation as well? When it comes to pngs I know for a fact that they thumbnail as png, so they shouldn't be used for sake of the end user. I'll try to update it to cover that. As for the topic of computation needed to present a file I'm primary looking out for the end-user, I'll also try to make that more apparent. CFCF (talk) 07:15, 7 July 2015 (UTC)
P.S. The goal of the guide is to make more users upload and work on lossless files rather than the other way around, but to do so in a reasonable manner. I wonder if MediaWiki would support thumbnailing PNGs as JPEG? CFCF (talk) 07:21, 7 July 2015 (UTC)
What does computation needed to present a file have to do with the end-user? Is there any system that can handle HTML 5 that takes non-negligible time to process a thumbnail-sized image, either PNG or JPEG?--Prosfilaes (talk) 07:37, 7 July 2015 (UTC)
I'd really like to see numbers. There's a rule of thumb in the computer world that you should measure before optimizing and after, because many optimizations aren't, or aren't in real circumstances.--Prosfilaes (talk) 08:08, 7 July 2015 (UTC)
Hi, I think TIFF or PNG for large files should only be used as archive. For regular display, JPEG should be used. There is no shortage of disk space, so there is no reason to have 2 versions. Regards, Yann (talk) 07:57, 7 July 2015 (UTC)
Note that TIFF is not necessarily uncompressed nor is it necessarily lossless. Also, the masters of non-professional photographs are frequently JPEGs, but one should still not work on a lossy format; if you're going to be doing multigenerational editing, intermediate steps (and probably the final one) need to be lossless.--Prosfilaes (talk) 08:08, 7 July 2015 (UTC)

July 07[edit]