Commons:Village pump/Archive/2013/12

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

Fantasy railway maps.

There seems to be a whole category here Category:Tyne and Wear Metro development devoted to maps from some obscure book about fantasy extensions to the en:Tyne and Wear Metro. As far as I know none of these proposal extensions have any official backing. So I'm not sure what value this adds to the project. Any thoughts? G-13114 (talk) 00:46, 19 November 2013 (UTC)

Unless these images violate copyright laws or the original author/uploader requests, Commons usually does not delete them for reasons like original research or false information, to my knowledge. -- Sameboat - 同舟 (talk) 04:08, 19 November 2013 (UTC)
I second what Sameboat said. And I have found this other example very useful before (to settle an agrument with my kids :-) ). --Dschwen (talk) 04:50, 19 November 2013 (UTC)
I guess we need a warning template {{fictitious map}}, we already have {{Fictitious flag}}. The Yeti 05:48, 19 November 2013 (UTC)
However Commons does delete out of scope images and while world famous (in some demographics) Sodor railways are deemed worthy of wikipedia article, not every fictitious map or flag is. So unless there are some educational qualities to those images they might be out of scope. --Jarekt (talk) 14:02, 19 November 2013 (UTC)
From my reading of it, it does appear to fall foul of COM:EDUSE. In that it is of no educational value, and is self promotion. G-13114 (talk) 20:59, 19 November 2013 (UTC)
I'm totally open to deletion of files that don't qualify the scope of this project. The only problem I have is that many files that are clearly not used and not realistically educational still survive requests for deletion, such as our famous swinging bell GIF (not safe for work). -- Sameboat - 同舟 (talk) 23:04, 19 November 2013 (UTC)
Why does something need official backing to be of educational use? I think it's within scope for people to post material from whatever POV they might have here (or from others' POV), assuming that it can serve some realistic educational purpose. We are not (just) a repository of files supporting government press releases. These maps may not have official backing, but they seem to have a serious purpose, and one that can be seen as educational. (Quote from the book's description: "[the book] shows how best to transform the existing Local Tyne and Wear Metro network in[to a] developed Regional network".) The description of this material as "fantasy" and the analogies with Sodor seem way off base to me. --Avenue (talk) 02:16, 20 November 2013 (UTC)
The requirement of "educational" in its broad meaning of "providing knowledge; instructional or informative" and "be realistically useful for an educational purpose" is a low threshold to clear, so we only apply it to the obvious cases, but fantasy flags or maps are good candidates to consider. However one can probably argue that Category:Tyne and Wear Metro development could be useful for illustrating optimal metro design strategies, etc. --Jarekt (talk) 03:47, 20 November 2013 (UTC)
They are fantasy, in that they have no connection with reality. They have no official status nor any chance of ever being implemented. At least the Sodor maps have some purpose in that they illustrate a well known fictional series. I'm not sure what purpose these serve, other than being someone's vanity project. I think at the very least the name of the category should be changed to something which makes clear that these maps are not official plans, the present name suggests that they have some sort of official status. G-13114 (talk) 20:24, 20 November 2013 (UTC)
Maybe you're right about their chance of being implemented, but I don't think Commons should generally be in the business of deciding which transport proposals are "fantasy". They seem to be clearly intended as educational or at least for advocacy, which IMO is enough reason to keep them. I agree that the category name should be improved, maybe by appending the word "concepts" or similar. --Avenue (talk) 14:21, 21 November 2013 (UTC)
I would propose that the category is moved to Category:Samoilov G.K. THE TYNE AND WEAR METRO DEVELOPMENT PROPOSAL or something similar. To make it clear what they are. G-13114 (talk) 23:13, 21 November 2013 (UTC)
That doesn't follow our usual category naming practices. How about Category:G.K. Samoilov's proposals for Tyne and Wear Metro development instead?
Having said that, I don't think everything currently in the category belongs under such a title. Some of the maps (e.g. 1, 2) depict lines being renovated in August 2013,[1] i.e. not proposals. The subcategory Category:Tyne and Wear Metro test track contains photos of actual test track and vehicles from around 1980. These seem quite appropriate under the current category title, so we shouldn't move the whole category. I think we should leave the existing category in place, and create a new subcategory for maps about Samoilov's proposals. --Avenue (talk) 13:18, 22 November 2013 (UTC)
Ok, I'll get onto that when I get around to it. G-13114 (talk) 20:14, 22 November 2013 (UTC)

DONE! G-13114 (talk) 04:08, 23 November 2013 (UTC)

Sameboat, really? You can't have a simple discussion of scope without bringing up sexual pictures? The more people attack sexual pictures, the more people are going to knee-jerk defend them, and the harder it is to be to have a reasonable discussion of them. If you actually want to talk about what educational means in scope, then bringing up a sexual image that's going to get people's backs up is seriously unhelpful.--Prosfilaes (talk) 22:09, 23 November 2013 (UTC)
In fact I have just created a {{Fictitious map}}. Partially pr. The Yeti's suggestion, partially because I previously saw a need for it as well... --heb [T C E] 14:44, 28 November 2013 (UTC)
Should that be put on the maps discussed here? G-13114 (talk) 18:32, 1 December 2013 (UTC)

Upload from CDC

Is it possible to upload some of these pictures? We've got others from CDC. At least couple from page 2 or 3. Thanks. --Ganímedes (talk) 00:55, 1 December 2013 (UTC)

Yes, it is possible. Ruslik (talk) 04:20, 1 December 2013 (UTC)

Latest deletion spree on portraits


No doubt out of good motivation, there has been a purge on portraits that might be seen as selfies. Unfortunately quite a few high quality reasonably valuable photos and validly released photos of celebrities and notable people such as Professors have been swept up in the jealous hunting down of selfies. I would appreciate other volunteers taking a moment to look through my snapshot gallery created from DRs raised by The Photographer over the last week, and more opinions added to these DRs to ensure the babies don't get thrown away with the bathwater, in the rare and unlikely event that an admin fails to look at these more closely before bringing down the big delete hammer. Thanks -- (talk) 18:08, 27 November 2013 (UTC)

According to glamorous tool 20 of those images are being used on several wikipedias. I support The Photographer work to clean up some out of scope images, but agree with that it would be a pity to loose valuable images "in the rare and unlikely event" that nobody looks at them more closely.
--Jarekt (talk) 19:17, 27 November 2013 (UTC)
While I surely support the clean up out of scope images (having requested a few myself), I certainly see no merit whatsoever in The Photographer’s “work”: Not only the said spree was done in spite, after a tasteless racist caricature authored by him was marked for deletion, but also the only invested “work” was a blanket of lazy DRs with vague c&p rationales. That’s not work! Yes, some of those DRs were valid, but that’s the broken clock analogy: It could be said that among 19 million files there’s a lot of garbage that escaped a DR so far, but marking all those 19 million files for deletion, while being the surest way to catch 100% of said garbage, would be a certain waste of resources in telling the false positives apart. The Photographer has been wasting everybody’s time with this prank; I’ve seen week long blocks awarded for much less. -- Tuválkin 13:43, 28 November 2013 (UTC)
Fae, you said «rare and unlikely event», heh — irony scathes, it does. -- Tuválkin 13:43, 28 November 2013 (UTC)
No comment. -- (talk) 13:53, 28 November 2013 (UTC)

More eyes on this please—The Photographer is now creating DRs at a rate of several per minute with 56 95 boilerplate DRs being created so far today, see Photographer&withJS=MediaWiki%3AGadget-rightsfilter.js&lifilter=1&lifilterexpr=Starting+deletion+request list. Based on the most recent being raised in alphabetic order (over 30 starting in sequence from A to B), this is thoughtless automated editing that is going badly wrong.

In the space of just 15 minutes from 15:30 UK time, I note that 32 DRs were created. This would be impossible without automated tools using boiler-plate text. -- (talk) 16:03, 29 November 2013 (UTC)

Selected from just today's DRs, I have added a handful of cases below. -- (talk) 14:55, 29 November 2013 (UTC)

As the examples below show, this needs to stop. Andy Mabbett (talk) 15:34, 2 December 2013 (UTC)

Case 1: The dead Bishop

This deletion request is for a photograph of a bishop who died in 1954, the same photograph is used on his grave headstone. See pl:Maria Jakub Próchniewski. The photograph was uploaded last year. The Photographer warns the uploader against "self promotion" and that Commons is not a social network. -- (talk) 14:24, 29 November 2013 (UTC)

Case 2: The living Bishop

This is a thumbnail portrait of Bishop Claude Makengo, uploaded by himself in 2011. The Photographer warns the Bishop that Commons is not a social network. -- (talk) 14:28, 29 November 2013 (UTC)

Case 3: The Actor

This is a self portrait of a Brazilian actor, uploaded by himself last year. The Photographer warns him that Commons is not a social network. -- (talk) 14:43, 29 November 2013 (UTC)

Case 4: The DJ

This is a portrait of an international DJ at their mixing desk. A history and discography were on the image page. The Photographer uses their automated boiler-plate text to warn the uploader against "self promotion" and that Commons is not a social network. -- (talk) 16:47, 2 December 2013 (UTC)

Case 5: The Musician

Portrait of a well known and published Brazilian musician, the image page correctly categorized as a guitarist from Brazil. The Photographer uses their automated boiler-plate text to warn the uploader against "self promotion" and that Commons is not a social network. -- (talk) 16:47, 2 December 2013 (UTC)

Case 6: The Footballer

It turned out that this image was a duplicate of an existing photograph of the Argentinian Footballer which is used on 6 different language Wikipedias, so I marked it with {{Duplicate}}. The Photographer uses their automated boiler-plate text to warn the uploader against "self promotion" and that Commons is not a social network. -- (talk) 18:29, 2 December 2013 (UTC)

November 28

Credit embedded onto photo

Shouldn't we obstruct uploading images with credit captions on them? This is so much more ironic when the watermark is no other than a copyright symbol! :) Doesn't this turn the actual file by-definition unusable per Wiki's philosiphy? Please see a batch example in Category:Arraial Pride where almost all content appears to be from the same copyright-reserving source... Orrlingtalk 00:14, 28 November 2013 (UTC)

  • I don't see any reason to "obstruct" these. They come from Flickr; there's no apparent problem with the rights, and given the license there would be no problem cropping or doing other cleanup to make them useable. Marking with a copyright watermark does not reserve any rights that were explicitly released via a license. Most images on Commons are copyrighted (though not watermarked) and have their rights released similarly. So the only issue here is the watermark and, as I said, that can be worked around if someone finds the images otherwise useful. - Jmabel ! talk 01:13, 28 November 2013 (UTC)
    • What is it about the license that would permit the removal of copyright management info (vis a vis DMCA)? Saffron Blaze (talk) 04:14, 28 November 2013 (UTC).
Cf. {{Watermark}}. -- Tuválkin 13:28, 28 November 2013 (UTC)
  • No. Logically, if the copyright holder releases an image on Flickr, then the licence they give on Flickr either supersedes or runs in parallel with any mark/watermark on the image itself that may appear to be a licence. In effect, the re-user is free to choose whichever licence they prefer where the licenses are non-revocable. -- (talk) 14:12, 28 November 2013 (UTC)
  • Although the license obviously gives us the ability to remove the watermark, and the files appear to be properly licensed on Flickr, has anyone had a look at the EXIF data? Most (almost all) are taken with a "COOLPIX P6000" and downsized, this however: File:Lisboa (P), 2011, Arraial Pride 2011. (6236744605).jpg, gives me some concern. Is it possible that this is Flickr washing with EXIF being scrubbed and size altered? Is there a function within COM:OTRS where outbound request can be made? It may be a good idea to shoot off an email to the person listed in the EXIF date of the file listed above, just to check that everything is above board. Liamdavies (talk) 16:56, 28 November 2013 (UTC)
    • No need. Just go on to Flickr and ask the Flickrstream owner yourself.
    Hm, after looking into this, CSI style, I will be raising a DR on this image to lay out the problems with it.
    Now at Commons:Deletion requests/File:Lisboa (P), 2011, Arraial Pride 2011. (6236744605).jpg. -- (talk) 17:21, 28 November 2013 (UTC)
  • Thanks Fae, I guess my question now is: can we trust the rest of the uploads? Yes, I understand this is a difficult question to answer, but for the moment one to consider in relation to this file, the others being the same low resolution, and COM:PRP. Liamdavies (talk) 17:36, 28 November 2013 (UTC)
    As I suggested in the DR, if someone spots another suspect credit, then bam, I would push for blacklisting the stream as a bad source. This was a photo from a few years back, so maybe this was some kind of odd oversight when they were editing a batch. -- (talk) 17:49, 28 November 2013 (UTC)
Comment. Metadata can be modified. Dates appearing in the EXIF don't reveal anything . Uploaders don't have any onus on them to care that the date of creation or camera model be exposed, the only place where the true date needs to show up is the description box. Orrlingtalk 18:00, 28 November 2013 (UTC)
Odd EXIF data is a reasonable indicator that something is wrong and this example showed that concern was justified (in line with the precautionary principle). I agree that EXIF data is not entirely reliable, or that deletion decisions are made on that metadata alone. For example we have plenty of files on Commons where the EXIF data has an all rights reserved or non-commercial only licence within it, but so long as our release is credible, what is in the EXIF data does not overrule it. -- (talk) 18:27, 28 November 2013 (UTC)
"Although the license obviously gives us the ability to remove the watermark..." Strange, I read the CC licenses and nothing there makes it obvious, unless you are trying to abuse the term adaptation as a means to an end. The issue is not so much what the policy here allows, but whether it is illegal to remove copyright management inforamtion (i.e things like watermarks) given that the DMCA specifically forbids it and case law is there to support that position. 18:39, 28 November 2013 (UTC)
Removing a watermark is, very obviously, an adaptation. There isn't anything even remotely unreasonable in that interpretation. If that weren't allowed, we could never crop any watermarked images in any way. darkweasel94 18:52, 28 November 2013 (UTC)
I think the CC-BY-SA-2.0 license is somewhat unclear. Clause 4c says "You must keep intact all copyright notices for the Work" when distributing or publicly displaying the work, and the license doesn't define what it means by "keep intact". Perhaps it means that a watermark consisting of a copyright notice right shouldn't be removed, or perhaps the watermark can be removed if the copyright notice is reported elsewhere, e.g. with the author's credit. The first interpretation does seem inconsistent with the general idea of freedom to make derivatives, which makes that seem less likely to be the intended meaning. On the other hand, the second interpretation does seem to stretch the meaning of "keep intact", particularly if the image would then be displayed in Wikipedia articles without the copyright notice, which was instead relegated to the image description page.
Interestingly, the wording has changed significantly in version 4.0 of the CC-BY-SA license, to say you must "retain" a copyright notice, and that this can be done "in any reasonable manner based on the medium, means, and context". Removing watermarks containing a copyright notice seems a lot safer under this version of the license. And since version 2.0 (clause 4b) allows derivatives to be made available under a later version of the license, perhaps that offers a way out of the dilemma. --Avenue (talk) 21:44, 28 November 2013 (UTC)
DW, it is only obvious to those that have decided here on Commons that removal of watermarks is an adaptation. Certainly WMF legal has not supported this position we have taken. Moreover, I find no such clear consensus elsewhere and Creative Commons does not provide an opinion on the matter. Not certain any of the CC licenses are sufficient to waive this DMCA provision: To argue that is reasonable to remove CMI may serve our purposes here on Commons but it may not be legal even with CC 4.0. The safest approach is to delete all images that are uploaded with watermarks. Saffron Blaze (talk) 23:25, 28 November 2013 (UTC)
Saffron Blaze, nobody is arguing to remove or ignore the relevant CMI. In fact our processes carefully preserve it. In the cases under discussion the relevant CMI is "CC-BY-SA-2.0" and attribution is preserved by including an attribution statement. Per Avenue's quotation from CC above, the word "reasonable" applies, not the word "unreasonable". In the absence of any case law to back up strange and extreme interpretations of copyright, there seems no significant doubt for our conventional and widely understood interpretation of what these Creative Commons licences mean by the word "reasonable".
I agree the "safest" thing to do is for the WMF to forcibly close down and delete all Wikimedia projects, hand out remaining cash to the directors and pay off the staff, however that is not exactly a method of delivering on our charitable shared open knowledge mission. Thanks -- (talk) 23:43, 28 November 2013 (UTC)
Wonderful reduction to absurdity Saffron Blaze (talk) 00:20, 29 November 2013 (UTC)

Copyright information is moved when we crop a watermark, not removed. The linked law article does not apply to this situation. In CC licenses the author can say with what text (s)he wants to be credited, not where that text has to be located. Copyright information belongs to the image description page, not to the image itself. Jcb (talk) 23:49, 28 November 2013 (UTC)

(edit conflict) I'd agree moving copyright notices out of the photo and onto the file description page does not constitute "removal" as prohibited by 17 USC § 1202 (linked above), especially since that prohibits such removals only when done "with the intent to induce, enable, facilitate, or conceal infringement".
I don't agree that all CC licenses directly allow this, however. The requirement in CC-BY-SA 2.0 and 3.0 to leave copyright notices intact is quite separate from the requirement to give credit as specified. The latter allows this credit to be provided in ways "reasonable to the medium"; the requirement to keep copyright notices intact allows no such leeway. For that, I believe you need to relicense your adaptation under version 4.0. --Avenue (talk) 00:08, 29 November 2013 (UTC)
My guess: the CC licence allows you to make new works that include whatever aspects of the original work that you choose. You can make an oil painting from a portion of a photograph, and if you don't include the copyright notice, you aren't "removing" the copyright notice, you are simply choosing not to transfer that part of the original into your new derived work. Making a new photograph would be the same in principle. However the details of how the copyright holder has requested attribution under CC may be relevant. --ghouston (talk) 23:54, 28 November 2013 (UTC)
Although what Avenue wrote above about the change in wording is true: the v3 licence does support an interpretation that such a copyright notice must be included in any derived works. --ghouston (talk) 00:15, 29 November 2013 (UTC)
  • People seem content to spout Commons consensus without offering any support for that position. Regardless, if any of you have a single legal opinion, or even one from Creative Commons, to support your opinion that the DMCA prohibition regarding removing CMI from an image (read watermark) is not illegal given the way our processes preserve it then I will be satisfied. Unfortunately all I can find in searches is case after case of people being sued via the DMCA 1202 provision. Saffron Blaze (talk) 00:04, 29 November 2013 (UTC)
    • Hm, "spouting". Well we have experience of successfully hosting 20 million photographs, I am sure any copyright lawyer would take that experience and history of a lack of any significant legal problems into account. If anyone were to bring a legal case against one of our images, then we would be keen to find ways of complying with the law as stated. As for your interpretation of "CMI = Watermark", I'm afraid that I don't see many people being convinced by the legal justification (such as real case histories) that you have provided so far.
    As I don't see the point of "spouting" any more in reply to your odd assertions, for me at least this thread is plenty long enough, so done. Thanks -- (talk) 00:18, 29 November 2013 (UTC)
  • That's a good review of one important aspect of the courts' interpretation of 17 USC §1202, namely what sort of CMI is covered by it - CMI managed by dedicated information systems, digital CMI, or even analog CMI. But there are many other important aspects too.
For a wider ranging review, see pages 69-73 of this 2007 paper. In particular, page 74 notes the necessity for plaintiffs to prove intent to cause copyright infringement by removing the CMI. I think our usual watermark removal practices are unlikely to incur much risk of that. --Avenue (talk) 13:26, 29 November 2013 (UTC)
Concur, there is legal debate about what constitutes CMI. However, the latest case law is trending to a broader view and includes embedded watermarks. I also noted in some online digging that the provision in 1202 indicates "facilitating" as well as "committing" as a cause for action. So, while Commons may not intend to cause copyright violations, they may be facilitating it by having a policy of removing watermarks. 15:50, 29 November 2013 (UTC)
Perhaps, but are we intending to facilitate infringement? That is the critical point, and I'd think the answer is no. --Avenue (talk) 23:21, 29 November 2013 (UTC)
Of course we are not intending on facilitating infringement although I would bet some lawyers would argue that once we became aware of the legal issue surrounding the removal of watermarks and yet we continued the practice, that in and of itself would show complicity. Saffron Blaze (talk) 00:06, 30 November 2013 (UTC)
You just effectively admitted there is no legal problem, because you agree we're not intending to facilitate infringement. It's hard to see how being aware of a non-problem makes one complicit in anything. And lawyers would argue the sky was green if they thought it would help their case. So would the argument you suggest really help their case? I'll quote from the article I cited above: "In order to prove liability under the CMI provisions in § 1202(b), a plaintiff must essentially prove that a defendant was expressly contemplating copyright infringement [...] the burden of showing a culpable mental state on the part of the defendant is a heavy burden for any plaintiff. Rights-owning plaintiffs in several cases have failed to prevail on their CMI claims because they lacked admissible evidence of the requisite mental state of the infringing defendant." The article cites five such cases as examples. If that's the the only argument they've got, I think they're in trouble. --Avenue (talk) 01:04, 30 November 2013 (UTC)
You misconstrued my use of the term "we". I may not be intending and you may not be intending to facilitate infringement but others may be. If WM policy facilitates that then it is complicit. I do find it surprising given the recent case law on this very issue and way 1202 is being more liberally interpreted that you don't see this as an issue that could use legal opinion from the WMF or Creative Commons. In at least one case you cite the dismissal was overturned on appeal. Regardless, I am not here to argue which cases are actual precedent setting as it is clear the law is evolving on the matter and there is risk. Sure, the risk may be lower with 4.0 (not clear) but even you highlighted that earlier versions seem to support the protection of watermarks. Saffron Blaze (talk) 01:31, 30 November 2013 (UTC)
Yes, if someone is removing a watermark that contains a copyright notice from a file licensed under CC-BY-SA version 2 or 3 (and placing it instead on the image description page, of course), I think it is much safer to license the adapted material as CC-BY-SA-4.0 than to stick with the original license. Regarding "we", the intent of the person moving the copyright notice is the critical thing here. While I'd be happy to have qualified legal advice on this issue (e.g. from WMF or CC), I think the situation is clear enough for us to continue without such advice. --Avenue (talk) 02:12, 30 November 2013 (UTC)
Given the vast majority (read 90%+) of image reuses for images I have uploaded to Commons violate the terms of the license in one way or another, I am hardly going to get worked up one way or the other. I do think you are hanging your hat on one article's opinion while ignoring actual case law. Saffron Blaze (talk) 02:36, 30 November 2013 (UTC)
Would not the CC derivative permission, whereby unlimited alteration to the file can be made, so long as the result acknowledges the original author, and is licensed by the same or similar conditions, permit an alteration of the CMI (assuming we take at face value that the watermark is indeed a CMI)? If not (and we continue to take at face value that a watermark is CMI), that would mean that any crop made of an image with a watermark, which removes the watermark is a violation, therefore completely unable to be licensed as CC? I guess my point is very simply, either the derivative feature of the license permits us to crop the image, or it doesn't. If it does, we can crop the watermark out, if it doesn't, the license is invalid from the outset - it is impossible to license a watermarked image as CC unless it is also ND. Is that what you are saying As I said earlier, I am not a lawyer, but if this is seen to be a serious question, it should be put to legal, not the Village pump, we are completely ill equipped to deal with this issue here. Liamdavies (talk) 16:43, 29 November 2013 (UTC)
  • I would offer you made a very good summary of the issue. I have to imagine the new wording in CC 4.0 may have been an attempt to address the issue, albeit without expliciting stating as such. 18:00, 29 November 2013 (UTC)
What wording in 4.0 would waive the protection afforded by 1202? Saffron Blaze (talk) 00:09, 30 November 2013 (UTC)
See Section 3(a)(2): "You may satisfy the conditions in Section 3(a)(1) in any reasonable manner based on the medium, means, and context in which You Share the Licensed Material." (The relevant part of Section 3(a)(1) is 3(a)(1)(A)(ii): "[...] You must: retain the following [...] a copyright notice.") 17 USC § 1202(b) begins by saying "No person shall, without the authority of the copyright owner [...]" (my italics). If the copyright owner has licensed the work under CC-BY-SA-4.0, they've authorised anyone to retain copyright notices in any contextually reasonable manner; by removing the copyright notice from the image and providing it on our image description page, for instance.. --Avenue (talk) 01:52, 30 November 2013 (UTC)
  • This is my contention as well, the derivative adaption permissions inherently give reusers permission to alter CMI as long as the attribution condition continues to be met. Where on meta would be the appropriate place to raise this issue? I think at this point it would be best to get a more official response from someone working for the WMF in a legal capacity. Liamdavies (talk) 07:46, 30 November 2013 (UTC)
I'm not sure our positions are quite the same, as I gather you think moving copyright notices would be fine under earlier versions of CC licenses, and I think the text of those licenses doesn't support this. --Avenue (talk) 11:32, 30 November 2013 (UTC)
  • I have posted to the talk page of Geoffbrigham at meta regarding these questions. Liamdavies (talk) 15:32, 1 December 2013 (UTC)
I am coming to the conclusion 4.0 may just solve the issue with 1202 as illuminated by meta:User:LVilla (WMF) and Avenue (although it assumes copyright notice and CMI are the same thing) but I remain concerned for earlier versions. Saffron Blaze (talk) 14:58, 30 November 2013 (UTC)
I see Luis Villa has written about CC 4.0 licenses on the Wikimedia blog.[2] Is that what you're referring to, or has he weighed in on this somewhere else? --Avenue (talk) 20:54, 30 November 2013 (UTC)
I thought it was here within discussion, but it must be somewhere else. Too many browser tabs I suspect. I will have to look for it later and link to the comment properly. Sorry. Saffron Blaze (talk) 21:07, 30 November 2013 (UTC)
That's fine, there's no rush. I'm just curious about what he might have said. --Avenue (talk) 02:33, 1 December 2013 (UTC)
I am beginning to believe I am guilty of cross-pollinating his name with your comment above, so I struck out the reference. Saffron Blaze (talk) 03:41, 1 December 2013 (UTC)
I think files licensed under CC-BY-SA 2.0, 2.5 or 3.0 shouldn't be an insurmountable problem, since their derivatives can be released under CC-BY-SA 4.0. However CC-BY-SA 1.0 doesn't allow that, and nor do CC-BY 1.0, 2.0, 2.5 and 3.0. --Avenue (talk) 12:11, 1 December 2013 (UTC)
I am not so certain. The copyright like rights give away in 4.0 could be argued, quite reasonably, were not envisioned as part of the license elements. Amending BY or SA is one thing but adding in new rights is quite another. Saffron Blaze (talk) 00:44, 2 December 2013 (UTC)
"...nor do CC-BY 1.0, 2.0, 2.5 and 3.0." Avenue, I didn't get what you mean. As far as I know, a CC BY license (attribution restriction only) is not at all bothered about what license we choose for the adaptations. JKadavoor Jee 08:35, 2 December 2013 (UTC)
You're right, I was wrong, and I've struck that part out. --Avenue (talk) 11:26, 2 December 2013 (UTC)

I can accept moving the CMI to a convenient location, without removing it. But are we doing it properly? No; I think. I discussed it with the original uploader who collected those works from the photographer who is very famous in India. He said N. A. Nazeer is much demanding about proper attribution; so I did some changes there and advised to do in all his works. I don't think placing CMI elements under "Camera manufacturer" and "Camera model" in EXIF make any sense. JKadavoor Jee 07:45, 29 November 2013 (UTC)

I agree that where there is a known attribution statement this is best added into the licence template attribution field. Looking at the Nazeer example file, I'm unsure why the EXIF data is misusing fields (I would assume that the original was like that already). The statement in {{Watermark removed}} that attribution was "moved into the image metadata" should not mean that Commons contributors are hiding attribution by changing a file's EXIF data. I find that approach bizarre and wrong headed. I would find it acceptable to explain what has changed in the information box, so long as it is clear, as the default licence template gives a legal requirement to use the correct attribution, it does not actually matter whether this is embedded in the licence statement or further up the page in the description, author, or permissions fields.
If the {{Watermark removed}} is encouraging people to do odd things like tampering with the original EXIF data, then it needs more appropriate guidance text to be added to the template page. I don't recall ever using this particular template, even for images where I removed watermarks. At the end of the day, most of these templates are a little naff compared to a plainly written explanation of provenance or what has been done in the description. -- (talk) 08:10, 29 November 2013 (UTC)
Thanks. Hope we can improve our customs and best practices day by day. JKadavoor Jee 08:30, 29 November 2013 (UTC)
  • I have always understood the CC-BY-AT 3.0 license to permit derivative versions, so long as they: are licensed under the same or similar conditions and acknowledge the original author in a suitable manner. It has been upheld here - on legal advise - that the suitable manner is not whatever they author sees fit, but with a name/pseudonym they choose. If it has been found that the author cannot stipulate font size and proximity to the image (as one did not so long ago, they wanted it in the caption), then on the image would be the same situation. As a crop is a derivative adaptation I understand it to be a fully compliant action within the CC-BY-AT 3.0 license. This however, is my opinion/understanding, if there is serious concern legal should be requested to input on this. Liamdavies (talk) 08:41, 29 November 2013 (UTC)

Note: BTW, Dolf Patijn confirmed that [3] is a copyvio of [4] as concluded at Commons:Deletion requests/File:Lisboa (P), 2011, Arraial Pride 2011. (6236744605).jpg. JKadavoor Jee 07:18, 1 December 2013 (UTC)

November 29

Wikimapia question

According to this page, all user-contributed content on Wikimapia is licensed under CC-BY-SA 3.0. How do I find out whom to attribute?

Example: File:Hồ Đồng đò.jpg comes from but the Commons uploader appears to attribute himself instead of the Wikimapia uploader. Who can correct this? --Stefan4 (talk) 15:59, 29 November 2013 (UTC)

  • It seems likely given the similarity of the usernames that it is the same person. Have you asked them if they uploaded the images to both websites? Or are you assuming that they got downloaded from Wikimapia and uploaded here by another user? Liamdavies (talk) 16:53, 29 November 2013 (UTC)
    • How do you find the name of the Wikimapia user? I can't find it anywhere. --Stefan4 (talk) 17:12, 29 November 2013 (UTC)
  • If you click the thumbnail the name shows below the bottom right corner of the larger image. Liamdavies (talk) 17:37, 29 November 2013 (UTC)

Does wikimedia support several mapping projects? There is also ShareMap wich ask money from the foundation.Smiley.toerist (talk) 19:19, 29 November 2013 (UTC)

Wikimapia is not a WMF project. Dankarl (talk) 14:25, 2 December 2013 (UTC)

November 30


Hi, there is a little problem with that template. User:Ahonc changed the "grammar" of the month name to genitiv some days ago - see this. As he told me thats correct minimum for russian and ukrainian. But its wrong minimum for german. See also my (short) discussion with Ahonc here. Perhaps here is a template specialist who could "programm" with if/case/switch structures or similar that the "grammar" for each language can be set individualy. Thx in advance. --JuTa 16:15, 1 December 2013 (UTC)

In Swedish, the month has to be in nominative in this template. --Stefan4 (talk) 08:48, 2 December 2013 (UTC)

December 02

Pull geocoding out of EXIF?

I just uploaded a bunch of photos (like this one) that have geographic coordinates in their metadata, and I rather thought Commonist was going to put the relevant template in for me. It didn't! (Bad Commonist. Why not? It did before...) Anyway, I'm just wondering if there's any system in place that will put the right template in for me? Or should I go back over them all and do it manually? — Sam Wilson ( TalkContribs ) … 07:50, 2 December 2013 (UTC)

Hi, you can use {{GPS EXIF}} to mark the images and then a bot will process them. If you are expreriencing troubles with Commonist, you might want to have a look at Commons:VicuñaUploader. It has some nifty features (including geocoding, both from EXIF and manually) and my personal opinion is that this is by far the best uploader we have at the moment. Cheers, --El Grafo (talk) 08:20, 2 December 2013 (UTC)
Thank you! I'm checking it out now. Thanks for the info re {{GPS EXIF}}... tomorrow's job I think. :-)
Hm, nope... it left out the attribution template. I must've done something wrong... I'll keep investigating... — Sam Wilson ( TalkContribs ) … 08:56, 2 December 2013 (UTC)

copyright / license

File:Miss-venezuela-maria-gabriela-isler-miss-universo-2013-CP9.jpg seems to be from this website (or similar), source: Reuters. I guess that means, there is a copyright. (Sorry, if this is not the right place for this, I am not used to commons.) -- 11:02, 2 December 2013 (UTC)

very copyrighted. Thanks a lot--Pierpao.lo (listening) 11:19, 2 December 2013 (UTC)
I didn't know there were degrees of "copyrighted" (except that something may be copyrighted in all, some or no countries). Isn't "a bit copyrighted" and "very copyrighted" like "a bit pregnant"? ;) darkweasel94 11:28, 2 December 2013 (UTC)
I acknowledge You are a bit right :)--Pierpao.lo (listening) 11:38, 2 December 2013 (UTC)

Duplicate category?

I noticed there are two similar categories: Category:Lamps in Italy and Category:Lamps of Italy. The first category is for lamps found in Italy, the second one for lamps manufactured or designed in Italy. Even if there's a point in this distinction, it's the only case in Category:Lamps. What do you think about?--Carnby (talk) 11:47, 2 December 2013 (UTC)

The first is necessary, objective, and follows a usual pattern. The second is more specific, could be more useful for some users, but is less generally applicable since such information is not always available. I would say keep but it needs a more specific name or there will be ongoing confusion. The "Aircraft of" categories are a precedent for both the category organization and the confusion. Dankarl (talk) 14:21, 2 December 2013 (UTC)
If "Lamps of" is the more specific one, it will have to be renamed; otherwise, it's probably mergeable or deletable. Orrlingtalk 17:26, 2 December 2013 (UTC)

"Category:Photographs of"

I find it important to bring this discussion to the attention of as many of you, especially established categorists, since it appears that now, say better later than never, there has actually been made an attempt by an editor to cope with a question which seems more and more relevant with the apparent influx of categories-for-photographs. Please attend that page and share your opinions there. Thanks. Orrlingtalk 12:32, 2 December 2013 (UTC)

December 03

Swiss Archives problems

What's happening with the various uploads here: . Some of the images have been replaced with negative, rotated images. Can they be reverted to the proper images? I don't where where else to ask about this problem... Oaktree b (talk) 02:49, 27 November 2013 (UTC)

It's a work in progress, the originals are the negative. the description will be improved in the next few days. --Chandres (talk) 16:20, 27 November 2013 (UTC)
We upload the original (negatives) like they are exactly after scanning. Then we upload an edited version. - The samples you see here are only a few for testing reasons and the first version was already the edited version. So we make an version with the original picture and we will revert it to the edited version after it. Interested people can also get the originals here. That's the reason for uploading it. --Micha (talk) 19:42, 27 November 2013 (UTC)
I tried to fix an upload where it was reported at de:Benutzer Diskussion:Swiss Federal Archives that the JPEG was a negative and upside down. I created a positive from the originally uploaded TIFF, rotated it and uploaded it as a TIFF again ( File:Nördlicher Teil von Pfetterhausen - CH-BAR - 3241903.tif) - however, it seems that the TIFF resolution is too large to generate a preview image resp. thumbnails from it here, no preview image is shown and if you click on one of the JPEG options, you get "Error creating thumbnail: The resolution of the source file is too large. No thumbnail will be generated." So, if the Swiss Federal Archives TIFFs are too large for the Commons thumbnail generator to handle, it's very good we have the lower-resolution JPEGs (in TIFF containers) as well, after all. What do you think - upload this particular image again as a new JPEG-TIFF version? Gestumblindi (talk) 00:40, 2 December 2013 (UTC)
Upload process now works well and we have already uploaded more than 50% of the pictures. You can start to work on them (category-sing, linking, ...), descriptions pages won't be overwritten by the bot. Kelson (talk) 10:05, 3 December 2013 (UTC)

TARDIS - a tool to help compare image sizes in nearby categories

Now you can compare an image's resolution and filesize against the benchmark of up to 100 others in related categories.

In some deletion requests, such as those where there are issues with Photographs of identifiable people or Nudity, the discussion often revolves around the image quality. Some images are then kept on the basis of whether they are relatively rare or useful examples of the educational topic they represent. There being no easy or agreed measure of quality, I developed a simple report on image size (both pixel size and file size) to help with these types of deletion requests, pasting a table comparing with other images in the categories, parent categories and grand-parent categories. I have made this available as a WMF Labs tool for anyone to try. This is not the same thing as quality but does give an indication.

You can find the tool at TARDIS (should load immediately) and there are several illustrative examples below (when doing a category comparison, the report can take 30 seconds or more to run):

  • Example with no category picked.
  • Example with category being compared.
  • Bread a simple example with non-Latin characters in use.
  • Fish an automated redirect (from "Fish") and a comparison to other images in the Fish (singer) category.

P.S. This is not an "official" tool, just my hacking about as a volunteer. Thanks -- (talk) 18:18, 30 November 2013 (UTC)

Another great tool, Fae, thanks for it. There’s a minor issue, about the three lists of categories: I tried this and only a few of the listed categories were linked — namely 3 among 8 cats, one among 18 parents, and two among 27 grandparents. Bug or feature? -- Tuválkin 23:52, 1 December 2013 (UTC)
Thanks. I look forward to seeing it used in a few deletion discussions. I have just tried out your example and all the category links appear fine for me (I'm using Firefox 16.0.2). You might be getting fooled by the colour of the categories, the default colour is black with those you have recently visited shown in dark blue, but they are all linked and will jump to a new tab or window when you click on them. I can always change the colours around if this is confusing (still on a necessary extended wikibreak this week, but might be free to look at this again next weekend). -- (talk) 01:18, 2 December 2013 (UTC)
Meh, sorry. It was exactly that. I’m so dumb. -- Tuválkin 21:28, 2 December 2013 (UTC)
Pictogram voting info.svg Info A key design limitation/feature is the choice to truncate results to a maximum of 100 comparisons, this was a pragmatic decision after finding my browser would time-out on more than around 150 file size comparisons. My original Python script is not limited in this way, I might be able to work around it in some future version, though a sample maximum of 100 is pretty useful as a benchmark and pulling down metadata on large categories (such as 1,000 or 10,000+ images) would invariably mean waiting for minutes rather than seconds, hardly practical for this sort of web-page based tool. -- (talk) 01:18, 2 December 2013 (UTC)

December 01

Duplicate pics?

These pics are basically the same, I wonder if we need such duplicates:

What do you think about?--Carnby (talk) 11:53, 2 December 2013 (UTC)

They are not duplicates because they are in different formats. Ruslik (talk) 03:47, 3 December 2013 (UTC)

Zentralbibliothek Solothurn (Switzerland) releases ~1200 pictures


The Zentralbibliothek Solothurn (a public library in German speaking Switzerland), in collaboration with Wikimedia CH, has newly released around 1.200 pictures coming from one of their collections. This prints and drawings collection started in 1941 gathers notable historical pictures about the canton of Solothurn. The focus is on subjects from the canton of Solothurn and on Solothurn artists including Ludwig Midart (1733-1800), Martin Disteli (1802-1844) or Heinrich Jenny (1824-1891). There are also views of the canton of Solothurn by well-known artists from other regions such as the Zurich engraver David Herrliberger (1697-1777) or the Toggenburg landscape painter and engraver Johann Baptist Isenring (1796-1860). Pictures were uploaded in high quality (300ppi) and in a lossless TIFF file format, they can be found at Category:Media_contributed_by_Zentralbibliothek_Solothurn.

Kelson (talk) 10:08, 3 December 2013 (UTC)

A five-minute task new volunteers can do

I've just blogged about a five-minute task new volunteers can do, transcribing small pieces of text in images such as foundation stones or signposts.

Please feel free to circulate that through your networks, especially outside the existing Wikimedia community. Andy Mabbett (talk) 15:41, 3 December 2013 (UTC)

December 04

Category:Females and Women

I' am speechless :). Please someone with a very good English can answer to this objection User talk:Orrling#Category:Black and white photographs of women because he is reverting somewhere the natural logic of the tree. Thanks--Pierpao.lo (listening) 11:17, 2 December 2013 (UTC)

Synonymous women and females forcibly split?

I was asking one same colleague here and here to stop the recent flooding of women-in-art categories with "Females" parents that seem as baseless as they're only made up to sustain the formal hierarchy of "first-females-then-women" in areas where that split is sometimes patently absurd, such as here. Generally I get the impression that plenty of "Females" cats are being and have been created to solely contain "Women" (and the other way around) where they could be regarded as synonymous. Per me, many of these categories – themselves usually pretty esoteric or deep-search leves anyway – can exist each as a unified "Women"/"Female" notion (either one) that includes the subcats of that gender while saving the useless age pettiness. I can't shake off the feeling that it is about two different legitimate philosophies of categorization, but given that I am the one constantly repairing after him/her I need you to assist in judging whether my approach or theirs is the more constructive one in the light of this question: should every "women" cat in the Arts tree automatically have a "female" parent and viceversa? Constructing these duplicate entries seems to me needlessly obsessed & burdening the navigation without any sense. Orrlingtalk 17:26, 2 December 2013 (UTC)

Yes, you are right, this is one of the silliest ideas of the Commons categorization idiots that make this project almost unusable by contributors and other users. There is no shortage of silly categories, but these are esepcially dumb. --AndreasPraefcke (talk) 09:24, 3 December 2013 (UTC)
I complained about Category:Bathing females in art long ago... In any case, anything is preferable to another "Adolescent girls" type category fiasco (Commons still isn't 100% recovered from the "Adolescent girls" debacle). AnonMoos (talk) 02:27, 4 December 2013 (UTC)
Incredible... I can't figure how these ones two categories Bathing females in art and Bathing women in art have resisted the sense since 2009 on and still are existent as a pair of separate cats. No less than shocking. Orrlingtalk 03:30, 4 December 2013 (UTC)

Avoid becoming your own grand-parent

Faebot says kill all humans time-travellers becoming their own grandparents is a bad thing.
Category:Categories which are children or grand-children of themselves

Faebot has started churning through category trees to populate the self-parenting/grand-parenting maintenance category above. Populating this is happening slowly (it is page-read intensive) with only a few handfuls being added per day, and is likely to take a few weeks to complete. If this is useful, I'll think about turning it into a routine monthly housekeeping job.

If you would like to fix some of these badly linked categories, check the category tree by following the links in the category and break the loop by removing the category from itself or from its problematic parent category (which happens to also be a child).

Self-categorizations like these are bad both in terms of logical context and in terms of our category trees which then contain infinite loops, a potential hazard for any category tree-climbing bot.

Thanks -- (talk) 07:19, 4 December 2013 (UTC)

Yes, this is useful. In the case of parents, it would be fine for the bot to simply remove them. But for grandparents it probably needs human intervention. When you delink, remember to remove the category from this tracking category too. --99of9 (talk) 08:04, 4 December 2013 (UTC)
+1 --McZusatz (talk) 11:13, 4 December 2013 (UTC)
Going through those categories reveals a massive overcat backlog which can not be cleaned up by humans. Can your bot remove those overcategorization as well? --McZusatz (talk) 11:13, 4 December 2013 (UTC)
I'm on a bit of a wikibreak but could look at improving this in a couple of weeks. I don't understand what you are expecting in terms of fixing an overcategorization problem, could you spell that out a bit in terms of what you would expect a bot to do? Thanks
Update Since posting here the maintenance category has been emptied. Faebot has only processed a very small percentage of categories so far, and there have been about 50 that have been fixed by maintainers (thanks!). As a method of handling them, using a maintenance category appears more engaging and intuitive than a wikitable/report (which was available, but hardly anyone knew). :-) -- (talk) 17:40, 4 December 2013 (UTC)

Dating Passo Pordoi postcards

I have uploaded

Old hotel Pordoi.jpg

. The location is no problem and there is even a present day website. The colour prints in category Postcards of (Passo Pordoi) are presumed to be ca. 1914. The date is fairly critical as this region was part of Austria-Hungary and became Italian after WW I. Is the postcard editor: Joh.F.Amonn Bozen an Italian editor or Austria-Hungarian? By the use of Bozen for Bolzano I have the impression that it would be Austrian.Smiley.toerist (talk) 13:16, 3 December 2013 (UTC)

It would be better if you could upload a copy of the back, to make sure that the photographer is not mentioned. Example: File:Sandra Milovanoff.jpg. Regards, Yann (talk) 13:25, 3 December 2013 (UTC)
I always mention mention all relevant information of the backside. The German and Italian is taken verbatim from the backside. There is a small logo (an elipse with: Joh.F.Amonn Bozen) At the poststamp area there is a stylized logo of a big A with the legs used to form the J and F letters, surrounded with a crown where in the headband there ROZEN written in it. The high number M 9566 indicates that the editor produced a lot of postcards. The same backsite, except the text wich is at the front is shown
location and dating needed.
below. I suspect that the same traveller bough those postcards as I found them in the same old postcard box in the shop.Smiley.toerist (talk) 16:58, 4 December 2013 (UTC)
Perhaps I don´t understand the question (my English is bad, sorry): Johann Filibert Amonn started owning the company in 1849 and all the subsequent owners kept the name J.F.Amonn. As Bozen and whole Südtirol (the place he lived) and the Trentino (the region of Passo Pordoi) became a part of Italy not earlier than July 1920 it seems rather unlikely that he ever had the chance to gain Italian citizenship (and if, it would have been at a very old age). The road to Passo Pordoi was not completed before 1905 (by which time J.F. was presumably in his late 70s or 80s) so I doubt that he took the photograph himself. --Rudolph Buch (talk) 13:06, 5 December 2013 (UTC)

Renamings required within Category:Ville-Marie

If someone has the time, there are a lot of files within Category:Ville-Marie that have generic names that would do well to have {{Rename}}s done to them by someone knowledgeable about that area of Montreal. Thanks if anyone has the time.  — billinghurst sDrewth 22:44, 4 December 2013 (UTC)

December 05

HotCat save step for multiple categories

When you change just one category in HotCat, it immediately autosaves the edit. If you are amending more than one, you can press the (++) hyperlink, change the categories, and then press Save... but then it opens up an edit window displaying the diff, and you have to press save again. Is a technical change feasible so that we skip this final step? --99of9 (talk) 04:52, 5 December 2013 (UTC)

Symbol support vote.svg Support--Jarekt (talk) 05:07, 5 December 2013 (UTC)
Symbol support vote.svg Support I've wondered this for years! :-) — Sam Wilson ( TalkContribs ) … 06:16, 5 December 2013 (UTC)
Symbol support vote.svg Support Yes please --Jwh (talk) 07:56, 5 December 2013 (UTC)
Symbol support vote.svg Support--Asqueladd (talk) 11:16, 5 December 2013 (UTC)
There is a technical means to achieve that, but use it at your own risk. See Help:Gadget-HotCat#Intercepting the page edit HotCat will make. Just copy-paste the sample code there to your ./common.js, then reload your browser's cache, and you're done. Lupo 12:52, 5 December 2013 (UTC)
Fantastic, thanks for the tip - that worked for me. --99of9 (talk) 13:05, 5 December 2013 (UTC)

Category moves: We can't do it without you

More voices needed from editors at the important rename notice at Category talk:Bathing women in art as there appears to be some newly-come hindrance on that required merger already agreed on at Village Pump; Please be sure to stamp your clear sane "Support the merge" down that page or we're doomed to remain with a nonsense cat which will encourage the creation of others.

The same here at Category talk:Indigenous people (English speakers only pls!), where the formal procedure restricts me from performing the move once there's even one single invalid "opposition", please make sure you put at least one "support" so the sensible majority is documented and the reparative move can happen. Orrlingtalk 16:28, 5 December 2013 (UTC)

Reporting copyright infringement

Might a sysop look into the activity of Adys16? He seems to upload photos of Romanian royalty from a blog which clearly states © 2007-2013 ASR Principele Radu. Toate drepturile rezervate (en: All rights reserved). I suppose a bot should delete them from all Wikiprojects, since he inserted his photos in many of them. Cheers, --Dan Mihai Pitea (talk) 23:25, 5 December 2013 (UTC)

  • Looks like his contributions have now all been deleted. - Jmabel ! talk 01:46, 6 December 2013 (UTC)
Checkmark This section is resolved and can be archived. If you disagree, replace this template with your comment. Jmabel ! talk 01:46, 6 December 2013 (UTC)

December 06

Category move

Category:Southern Star Observation Wheel needs to move to Category:Melbourne Star. Please could someone more knowledgeable have a quick look and see if I've done the necessary correctly? Thanks. 04:20, 6 December 2013 (UTC)

Neat! Many thanks. 17:59, 6 December 2013 (UTC)

EU Copyright consultation

Just a heads-up that the EU is holding (another) formal consultation on copyright issues. (See eg IPKat here and here for overview).

Since Commons probably knows as much as anyone about cross-border copyright incompatibilities in the EU as anyone, especially in regard to different national views across the EU of what can or cannot be considered public domain, a response worked up here would be very useful, particularly if then channeled through national associations (as comments by organisations tend to carry more weight than comments by individuals).

There's a lot I think that could be usefully commented on here -- eg difficulties caused by different standards on copyrightability, difficulties caused by the limitations on fair use, difficulties encountered in trying to copyright-clear old material (especially in comparison with the U.S. clear-cut cut-off date of 1923), difficulties in access to library materials, whether copyright terms are now excessive and no longer optimal (eg blocking access to old material -- now cheap digital publication of out-of-copyright material, has this changed the balance of pros and cons); whether in a connected age it makes sense to lock up material in Europe if it can be freely published in the United States... etc.

I think there is a lot here, where there are some very important analyses and points, from a Commons/WP perspective, that really ought to be made to the consultation. Jheald (talk) 12:26, 6 December 2013 (UTC)

Scans from Google Books


Just a reminder that the first page with an warning from Google should be removed before uploading the scans here. If you can't do it, I can help. Is there already a list of files for which that should be done? If not, it should be created. Thanks, Yann (talk) 12:28, 5 December 2013 (UTC)

Ähem, you are aware that the recent "permission" for Google books by an US court is based on U.S. fair-use doctrine, which would exclude them automatically from upload to Commons. --Túrelio (talk) 16:56, 5 December 2013 (UTC)
Um, yes, I'm sure Yann does, but they have a lot of clearly PD works on there.--Prosfilaes (talk) 21:44, 5 December 2013 (UTC)
Yes, as Prosfilaes says, Google scans fall exactly within our policy of "exact copies of 2D works do not produce a new copyright." Regards, Yann (talk) 07:14, 7 December 2013 (UTC)

File:Armoiries république française.svg

I asked some months ago to add the Category:National Emblem of France to the file, but no one answered. Could you please add it?--Carnby (talk) 21:46, 4 December 2013 (UTC)

@Carnby:. I am wondering why you don't feel empowered to add it yourself? The HotCat gadget (Special:Preferences#mw-prefsection-gadgets) is available to do this really easily. I invite you to try.  — billinghurst sDrewth 22:49, 4 December 2013 (UTC)
The image file is protected from editing and I'm not an admin.--Carnby (talk) 23:15, 4 December 2013 (UTC)
✓ Done - Jmabel ! talk 05:13, 5 December 2013 (UTC)
For future reference: requesting edits on protected pages can be done best with {{Edit request}}. This template will add the page to a maintenance category where an admin is much more likely to notice the request. MKFI (talk) 08:23, 5 December 2013 (UTC)
Thanks, but could you add Category:SVG coats of arms of France?--Carnby (talk) 12:44, 5 December 2013 (UTC)
It's OK now.--Carnby (talk) 20:28, 7 December 2013 (UTC)
Checkmark This section is resolved and can be archived. If you disagree, replace this template with your comment. Carnby (talk) 20:28, 7 December 2013 (UTC)

Notification of DMCA takedown demand - Apis-mellifera-queen-worker-drone

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):

Thank you! Maggie Dennis (WMF) (talk) 13:37, 5 December 2013 (UTC)

This appears to have started off as a fairly simple copyright violation deletion, so it seems a pity that the copyright owner did not seek satisfaction using that simple and transparent process rather than the often closed one of DMCA (a process he appears to adopt by policy Is there any reason why the OTRS ticket referenced in the deletion log should not be visible to the OTRS Commons permissions queue volunteers? There may be benefit in blacklisting the source website. Thanks -- (talk) 14:03, 5 December 2013 (UTC)
Hi, User:Fæ. I believe that the ticket was in the "urgent" queue which is restricted access, but I'm not sure - I haven't moved it, myself, but it's possible that somebody else did. However, I have viewed it, and the substance is the same as the notice on WMF wiki, including the source website. :) (To be clear, it was also a DMCA takedown, just sent to OTRS as well as legal.) --Maggie Dennis (WMF) (talk) 14:06, 5 December 2013 (UTC)
Ah, thanks for the explanation, I appreciate the transparency for the DMCA notices where there are no Photographs of identifiable people or other legal issues beyond enforcing legitimate copyright. As you say, there seems no point in worrying if the same text is available on the WMF wiki.
Opening this up more generally, the site is quite aggressive with regard to a policy of pursuing re-users with invoices for copyright and handling fees, I believe blacklisting it would be a sensible precaution for uploaders that may become confused about the other policies on that site for allowing educational use for free. Thanks -- (talk) 14:25, 5 December 2013 (UTC)
Aggressive? I think he just stands up for his rights and tends to do so in an acceptable way: Saffron Blaze (talk) 02:12, 6 December 2013 (UTC)
I am sure Alex is a lovely person and I like the way he encourages educational use of his photographs. The policy on his website gives a clear warning that a failure to respect his copyright terms will be pursued, as he has a legal right to do. The question here is whether blacklisting the site is a good way to serve the interests of our uploaders, potential re-users and Alex. Thanks -- (talk) 05:14, 6 December 2013 (UTC)
That's OK; but we can't say a person trying to defend his rights is aggressive. We should respect their rights. Some attempts and negligence to not provide proper attribution in derivative works here (not talking about that watermark issue) also should be discouraged. We can enjoy the freedom that is granted; but we have no right to encroach and try to grab the rights not granted. Otherwise, it is better to limit our scope to PD/CC0 materials. JKadavoor Jee 05:37, 6 December 2013 (UTC)
I have no idea why you would want to derail this thread. If there is no appetite to take action to stop this happening again and we are just happy to have more DMCA notices or see our reusers prosecuted, fine. I have better things to do with my time than worry about it. Struck my suggestion. -- (talk) 05:52, 6 December 2013 (UTC)
@: I think that blacklisting the url will just lead to its removal from copyright notices. I would think that it would be more effective to write an abuse filter to note any addition for prompt review and action.  — billinghurst sDrewth 12:02, 6 December 2013 (UTC)
Special:AbuseFilter/124 created, and at this stage only the one url. Other urls for sites where we have DMCA concerns can be added, or people can leave a message on my talk page.  — billinghurst sDrewth 13:39, 8 December 2013 (UTC)
Uploading with this URI in place will most likely not trigger the abuse filter. Cf. bugzilla:19565. -- Rillke(q?) 11:17, 9 December 2013 (UTC)

File upload revert

Whether the file upload revert action can appear in edit history as a revert, not a new upload? I some user revert the new version of the file, in edit history revert should appear as well. --Rezonansowy (talk) 21:44, 6 December 2013 (UTC)

The default message when somebody reverts a file is, indeed, "reverted to version" + timestamp. Are you suggesting that it shouldn't be possible to change that? That wouldn't really prevent anything because it is always possible to download the old version and just upload it as a new version. darkweasel94 10:05, 7 December 2013 (UTC)
I mean the History tab section, like here and you see it appears in edit summary as new upload, not revert. --Rezonansowy (talk) 12:42, 7 December 2013 (UTC)
From a software point of view there isn't really a difference between a revert and a new upload. In any case nobody here can change that - you could report it at Bugzilla if you think this is important. darkweasel94 14:30, 7 December 2013 (UTC)
It's important as a MediaWiki-style revert, which counts as revert and go to the notifications applet. I wish to report it, but I don't know, how to do it. Can some report this bug? — Preceding unsigned comment added by Rezonansowy (talk • contribs)
See - thanks in advance! --AKlapper (WMF) (talk) 10:55, 9 December 2013 (UTC)
Reported. Bugzilla58209. --Rezonansowy (talk) 14:41, 9 December 2013 (UTC)

December 07

Copyright: Two images from MS

I post here because nobody answers for some threads on Copyright section. I think (and some others) that:

files are simple shapes, gradients and some text, just like in File:Windows 7 logo and wordmark.svg (DR). I post here because this make to me and I prefer to make sure that to you too. --Rezonansowy (talk) 12:39, 7 December 2013 (UTC)

Nobody can answer this question for sure. Ruslik (talk) 17:20, 7 December 2013 (UTC)
Well, that's a Commons, just rate whether they are under the threshold of originality in United States. I've provided an example of another Commons decision. --Rezonansowy (talk) 17:51, 7 December 2013 (UTC)
The wordmark is {{PD-textlogo}} but I do not think the icons are simple enough. --McZusatz (talk) 22:36, 7 December 2013 (UTC)
Concur. Dankarl (talk) 23:52, 7 December 2013 (UTC)
Both images have stylized elements that would seem to depart significantly from being just a textlogo. Saffron Blaze (talk) 16:10, 8 December 2013 (UTC)
See Commons:Deletion requests/Files in Category:Xbox logos. -mattbuck (Talk) 22:17, 8 December 2013 (UTC)
Other than the simple typeface/wordmark logos, any logo which includes a complicated graphic/symbol (and really, once you start putting in shaped gradients, you're getting complex) is non-free. —Locke Coletc 01:35, 9 December 2013 (UTC)
I don't agree with you, see COM:TOO. Besides non-free aspect has nothing to do with the Threshold of originality. --Rezonansowy (talk) 14:12, 9 December 2013 (UTC)

December 08

Rebuild palace?

I uploaded File:Dakar Palais du Gouvernement Général.jpg. This looks very similar to File:Resident of the president of Senegal.jpg. The central tower disapeared and a lot of classical style decorations. Is this the same building?Smiley.toerist (talk) 21:47, 8 December 2013 (UTC)

I am quite sure it is. Take a look at File:AOF-Dakar-Palais du Gouvernement.JPG. --heb [T C E] 07:08, 9 December 2013 (UTC)
I suppose the letters AOE on the front of the building stand for: Afrique OuEst?Smiley.toerist (talk) 09:32, 9 December 2013 (UTC)
It is AOF which means fr:Afrique-Occidentale française. Yann (talk) 09:44, 9 December 2013 (UTC)

Notification of DMCA takedown demand - Cranach Digital Archive

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#Cranach Digital Archive. Thank you! Jalexander--WMF 06:44, 5 December 2013 (UTC)

Please note: As I noted on the takedown notice itself (on top) the request demands 4 actions (3 file removals and 1 meta data removal), the WMF Legal team refused to do the image takedowns based on their determination that the files were within the public domain. They determined that there may be a copyright claim for the longer comment that was placed within the meta data described on [4] and so I removed that from the file itself and deleted the old version. Jalexander--WMF 06:44, 5 December 2013 (UTC)
Thanks for the notice, @James, and my thanks to the legal team for refusing to comply with the takedown demand. Is there any possibility that you guys could publish this nonsense of a request on I'd also like to note that the Cranach Archive images were mentioned in OTRS ticket #2012112910012701, and that, most probably, Cranach Digital Archive staff engaged in abusive deletion nominations directly here on Commons under the account Cranach Digital Archive (talk · contribs). odder (talk) 09:48, 5 December 2013 (UTC)
Thanks for the OTRS pointer, I'll make sure the legal team sees it. I uploaded the takedown to Chilling Effects shortly after I posted here so they have it and will hopefully post it. CE, sadly, does a lot of manual review before posting and I've never completely understood what they post and don't post (and why it takes so long sometimes). That said we always upload to them and I'm hopeful they will post this one given the abuse (and some more of our past ones once they get around to them). Jalexander--WMF 10:02, 5 December 2013 (UTC)
With thanks to the legal team I want to mention the OTRS Ticket #2012122210009986 with regards to file [2], a former complaint of Cranach Digital Archive Commons:Undeletion requests/Archive/2013-08#File:Jerome-cranach-vienna-infrared.jpg made by Stiftung Museum Kunstpalast, Düsseldorf / Cologne University of Applied Sciences. --Oursana (talk) 02:01, 11 December 2013 (UTC)
  • Maybe this is a problem which should be mentioned on COM:CB or something? If you download a PD picture from some other website, there is always a risk that there might be something copyrighted in the metadata, and in that case we can't use the metadata, which we became aware of with this takedown request. --Stefan4 (talk) 14:17, 6 December 2013 (UTC)

Original uploads

I'm posting this here, to gather as many opinions as possible. Some time ago, I proposed a new template that would contain every source information about a file that has been imported from a sister project. The original discussion gathered consensus, but didn't get to an outcome; it can be found here. --Ricordisamoa 03:11, 6 December 2013 (UTC)

Please, write your opinion so that we can reach an outcome. --Ricordisamoa 13:42, 11 December 2013 (UTC)

African women icon - an offensive amateur cartoon hosted on Commons

I would like to draw the wider community's attention to this apparently racist cartoon created by one of our contributors which is under deletion discussion at Commons:Deletion requests/File:African women icon.svg.

This modern cartoon was previously deleted (under the name Negrita), then undeleted for what appear a series of confusing rationales on "historic merit" (when it has none as it is a modern creation). Unfortunately there were very few viewpoints expressed in those discussions. I encourage people to add their views to the most recent deletion request above.

There can be no doubt that the cartoon will appear racist to the general public and is likely to only be used within the Wikimedia projects for disruptive reasons (such as use for a deliberately offensive user icon). In short:

  • It has no historic value
  • It has no educational value
  • It is not part of a notable artist's work
  • It has no cultural value
  • It is not in use on any other project

Drawings such as this, from otherwise non-notable Wikimedia contributors, are outside of the scope of this project unless there is no doubt that they have educational value or are likely to be used on Wikimedia projects. In this case, the only likely use of this "icon" would be to disrupt our projects in order to make some sort of political point - I am unclear what exactly that point would be, and hesitate to make any speculation about the creator's motivation, but if the intention here is to make some point about censorship then it seems highly misplaced. -- (talk) 11:06, 6 December 2013 (UTC)

Fæ, you know very well that we don't take side on the political correctness (or not) of our content. So this DR is certainly not valid. However, a warning in a description that some people find this image offensive would be OK. Regards, Yann (talk) 11:43, 6 December 2013 (UTC)
Yann, the rationale here is not "political correctness" it is that this image is out of scope. Just because Commons does not censor images out of "political correctness" this does not give a mandate to keep every random "politically incorrect" drawing that anyone wishes to upload to make a point. Again this is not historic, not educational, not notable and has no cultural value. Thanks -- (talk) 11:52, 6 December 2013 (UTC)
Well, you post here IS about political correctness. If the issue is only about scope, you don't need to, and you won't add a message here. The DR would be enough. And you probably worsen the issue by displaying it here. Yann (talk) 12:02, 6 December 2013 (UTC)
No, the fact is that this image will be seen as racist and offensive to the general public. Identifying that as a problem for an image that has no rationale that puts it within the scope of this project is not "political correctness".
If you take a moment to absorb my written words before marginalizing them as political correctness, my point above is that it benefits this project to have as many views as possible from the community in a controversial DR such as this. Community consensus is in no way something than administrators of this project should be finding ways to avoid. As for whether including the image here is "worsening" some problem, how can it, if, as you appear to believe, this image is in scope? Perhaps you believe this image is defamatory as it stereotypes black women? Perhaps you could make that viewpoint clear in the DR.
In terms of censorship, Commons is not in the practice of adding "NSFW" type notices to her image pages. I am surprised that you are putting something forward along these lines as a solution to this image being out of scope. -- (talk) 12:18, 6 December 2013 (UTC)
You totally overdo it here Fæ:
  • The image might be out of scope (but is still much better than many others for wich no DRs exist), but this should - as Yann said - be subject of a regular deletion request. Nothing to worry about here.
  • The image itself is only offensive if you consider it so. I can reupload the same image with the color of the womans face changed to white (or yellow or red or whatever color else I might like) - why should I be offended by any of these? The racist connotation only emerges in your (or everybody elses) head, it has nothing to do with the image. So not the image is the problem here but people minds, which occasionally tend to create such inappropriate connections to racism from an otherwise unproblematic image. But we can't delete peoples minds, can we Face-wink.svg. So this can't be solved by deleting the image either, so blaming it doesn't help. --Patrick87 (talk) 12:37, 6 December 2013 (UTC)
The image is out of scope and is raised up on a deletion request on that basis. Just because an image is offensive or defamatory is not actually a rationale to keep it. Considering the image was re-uploaded as "African woman" and was originally called "Negrita" (Spanish for Black girl), that the image is meant to stereotype black women is obvious, not really a question of being some imagined left-wing fantasy of political correctness. -- (talk) 12:47, 6 December 2013 (UTC)
(Edit conflict) I also don't see why this particular DR is in any way more special than any other. Are we now going to routinely use the village pump as a second DR nomination log? darkweasel94 12:39, 6 December 2013 (UTC)
No. -- (talk) 12:47, 6 December 2013 (UTC)
Then what makes it more special than the 83 other DRs that have been filed today until now? darkweasel94 12:58, 6 December 2013 (UTC)
None of those has been undeleted on the basis of having historic value when they are modern creations, or have rationales being put forward that they should be kept even though they are out of scope, for no other reason than resisting the "political correctness" of being concerned that Commons is hosting out of scope material that is defamatory for black people. -- (talk) 13:05, 6 December 2013 (UTC)
Curious, would this lovely image also be racist? 14:33, 6 December 2013 (UTC)
Hello again. I am surprised that the Canada Department of National Defence would want to be seen to be involved in these discussions. The image linked off-wiki is a commercially produced textile print that is under copyright and would not be hosted on Commons. A key difference is the outsize lips on this stereotype cartoon, while the textile print is simply a posterized drawing of a woman without any offensive stereotype exaggeration. -- (talk) 15:06, 6 December 2013 (UTC)
So let's get some things straight: the problems with the image The Photographer uploaded are such: (1) it's freely licenced, (2) it's uncommercial, (3) the size of the lips of the caricatural woman is too big. Did I get this right? odder (talk) 18:38, 6 December 2013 (UTC)
No, you don't have it right I'm afraid. It is out of scope. Commons is not a website host for defamatory stereotype cartoons that people make up one day and decide to upload. If you double check exactly where this image has been used on the internet in the past, you will get the idea about whether this is an "innocent" image or not. What's next, amateur drawings of stereotype grasping jews or child-snatching gypsies, just because they are freely licenced? Commons' educational remit puts this sort of disruptive crap out of scope. Let this nonsense get deleted and restore this project to being a non-"hostile environment" where contributors are welcomed without being confronted by uploaders showing off their personal collections of home grown defamatory cartoons and racist material is limited to being discussed dispassionately with the perspective of being historic artefacts. Thanks -- (talk) 18:50, 6 December 2013 (UTC)
Fae, are you in the habit of tracking down IPs so you can out them? 19:09, 6 December 2013 (UTC)
Hello again. Anyone aware of my history with Wikimedia will know the reverse is true. In your case, you appear to be openly promoting this highly recognizable IP address, as you insist on repeatedly using it (and not just in this thread) rather than a named account that would have the benefit of not associating a government network supported by taxpayer funds, with your edits here. -- (talk) 19:15, 6 December 2013 (UTC)
You always seem to ascribe nefarious motives to anyone that disagrees with you. I should be able to post as an IP without explaining myself nor deal with people identifying my employer and highlighting what they think is morally objectionable conduct. 19:42, 6 December 2013 (UTC)
Looks like Doxing or Outing to me. Identifying someone's employer as a means to intimidate fits the definition. Saffron Blaze (talk) 05:58, 7 December 2013 (UTC)
Nonsense, you do not appear to know what these words mean. Goto the signature link page, at the bottom you will see several handy links built into the wiki footer, this includes a couple that will identify the IP geolocation, for example Geolocate gives you precisely this information. As for the IP's employer, I made no suggestion about this, the IP self-identified the IP network owner as their employer. Anyone who publicly edits from an IP address has no reason to doubt that they are making their IP address public. -- (talk) 08:50, 7 December 2013 (UTC)
It is not nonsense. You didn't like what someone said so you took the time to research the IP to find out who or where they were located then when you saw it was a recognizable corporate IP you outed them in way of a personal attack. That it was easy is immaterial. It is disgusting behaviour. Saffron Blaze (talk) 17:07, 7 December 2013 (UTC)
No outing here, nor any doxing thanks. Links are built into the standard Mediawiki software in order to be used by anyone, so using them in well within the scope and accepted norms of this project; if you have a problem with the design of this system take it up with the WMF development team.
What I find disgusting is those who seem to get their jollies from trolling discussions to take them off topic. It makes it virtually impossible to hold sensible conversations in order to gain any community commitment to improve policies here as well as ensuring that any reader gets their pants bored off by having to wade through the off-topic chaff, wikilawyering and obsessive hounding that any controversial discussion seems to attract. If you have any ideas of how to solve that problem, I would welcome a proposal. -- (talk) 18:12, 7 December 2013 (UTC)
What you did by highlighting the IP was an employee of the DND is Doxing. You did it to intimidate via an Ad Hominem attack in trying to discredit the individual instead of the argument they were making. In fact the IP had a point germane to the conversation in that they provided another image of traditionally dressed African woman that wasn't too far from the image you provided. Then you complain about the copyright status of the image instead of acknowledging the clear point it made. Saffron Blaze (talk) 00:08, 8 December 2013 (UTC)

I personally don't like or hate these type of images. But I hate some images, even if they have some historical values. Coming to the point, I think this image is undeleted on the mediation of Russavia on the terms "The Photographer, if the image were to be undeleted, would you provided a detailed description in at least Spanish on the file page? This would surely give the image context, and perhaps make others less opposed to the image being on Commons." I think he asked the permission of Jameslwoodward too that he didn't fully agreed. I believe that is an acceptable compromise; that acceptable for me too to avoid a big conflict. So it is better to ask Russavia whether the conditions are satisfied or not. JKadavoor Jee 17:12, 6 December 2013 (UTC)

The undeletion request was closed by Fastily, not Russavia. As for context of the creator and uploader of this cartoon stereotype of a black woman, I don't see how that changes an out of scope image into an in-scope one. Again, Commons is not short of historic racist images if anyone needs one for an educational purpose, I have even uploaded some myself, but there is a huge difference between 19th century drawings featuring "niggers" and self-uploads of cartoons which will appear to the public to be nothing more than intentional defamation without any unique educational merit. -- (talk) 17:30, 6 December 2013 (UTC)

(uncollapsed): This is quickly becoming extremely un-mellow and unlikely to result in anything productive. I realize this collapsing may be an overly bold move from a non-admin (and I'm okay with being reverted by any uninvolved editor), but there is no point at all in discussing these things here rather than in the deletion request itself, or if somebody thinks some admin should be stripped of their right or ability to close undeletion requests, COM:ANU. darkweasel94 20:02, 6 December 2013 (UTC)

I have uncollapsed. This is not the admin noticeboard or Jimmy Wales' en.wp talk page. There is no need to hide discussion on the village pump from view, as these discussions are not "problems" that need to have a resolution. If you think this is unproductive, then you neither have to read this thread, nor take part in it. -- (talk) 20:10, 6 December 2013 (UTC)
Somehow this discussion reminded me of the Tintin in the Congo-controversy. While it may not have been made in 1930s Europe, which had a certain perception of Africans, it does in a way, provide a (-nother?) free version of that perception. A lot of the arguments from the Tintin-controversy may also apply here - or rather in a deletion request. Yes, I am editing this in my free time from my own private, connection, but please bear in mind that for some defence employees (i.e. deployed or on a ship), their only internet access is through the respective military-connections - even when editing in what-ever free time they may have. Don't hold that against either the employer nor the employee. --heb [T C E] 11:20, 8 December 2013 (UTC)
Thanks for that. I have spent almost a year away from home over the past three years and if the government hadn't provided me internet access it would have been difficult to maintain connection with family and friends. So while what I do here on my own time is part of the acceptable use policy it still is disconcerting that Fae went out of his way to try to identify me. 13:59, 9 December 2013 (UTC)
"Fae went out of his way" - utter nonsense. There is absolutely no excuse to be unaware that the network/IP address you are using is your signature. If you want more privacy, rather than making this information public, it is terribly, terribly easy, see Commons:First_steps/Account. -- (talk) 14:13, 9 December 2013 (UTC)
Fae, the part you keep conveniently skipping over is that you went through the process, easy at it is, of looking to find out what you could about the IP then when you noted the address belonged to a "highly recognizable" organization you brought that to everyone's attention. What purpose did you have for doing that? Saffron Blaze (talk) 22:09, 9 December 2013 (UTC)
If you want to keep on flogging this horse, please do open a new thread about whether it is okay to click on a signature link of an anon IP and actually look at the public results. -- (talk) 23:06, 9 December 2013 (UTC)
You can't even admit to yourself why you did it. That's sad. Looking was one thing; making an issue of it was quite another. Saffron Blaze (talk) 22:42, 10 December 2013 (UTC)
The argument for keeping the image as an illustration of blackface seems very clear. There's no reason we can't have modern media illustrating historical styles in a simplified or stylized manner. Just to give one example, I could imagine this image being used in an article on blackface which, rather than just discussing historical depictions, also wanted to clearly point out the features associated with typical blackface, or perhaps one which wished to illustrate how a modern digital rendering of blackface might differ from historical media renderings. Of course it is an offensive depiction (at least in US culture), but that is an issue of how the image is used, not an issue of scope. Dcoetzee (talk) 04:08, 11 December 2013 (UTC)
This is not simple "blackface", the lips are anatomically impossible. You may want to read up on how grossly exaggerated stereotype characteristics have, and still are, used offensively, in particular the phrase "nigger lips". Deromanticizing Black History may be a bit old, but it has some readable essays and is available at local libraries ISBN 9780870497223. Again, Commons has historical images of defamatory stereotypes of black people, we do not need contributors to upload their own drawings to illustrate Blackface, Racism or Defamation when we have the originals. Thanks -- (talk) 10:57, 11 December 2013 (UTC)

I have not been here for a while

real life being what it is, but was surprised to find File:Circus Max 1978 2.jpg gone. It was a picture I took maybe 35 years ago of some 12,000 year old place and uploaded here because wikipedia likes me to do that. I see that I have another picture on the cusp and I am now so uninclined to upload more. So, you might be wondering, what is the question? No question, just frustration about time and effort spent to no avail. Einar aka Carptrash (talk) 21:50, 8 December 2013 (UTC)

The problem was that while you were the author of the file, you were not the creator of the model it was a photo of. Unfortunately there is no Freedom of Panorama in Italy, so if you take photos of artistic works in public places, the copyright of those artworks is important. I'm sorry if you find this disheartening, but we must follow the law regarding copyrights. -mattbuck (Talk) 22:16, 8 December 2013 (UTC)
I'm not an admin here so I can't view the deleted picture, but if it's of something 12,000 years old then surely whatever it is is out of copyright (if indeed they had copyright laws back in the stone age)? Or am I missing something?  An optimist on the run! 22:32, 10 December 2013 (UTC)
The image is of a modern scale model of the Circus Maximus in Rome. INeverCry 05:59, 11 December 2013 (UTC)
If it's a reproduction of a non-copyright building, then it can be argued that the model is also non-copyright (see similar discussion here), and therefore FoP doesn't apply.  An optimist on the run! 06:49, 11 December 2013 (UTC)
The CM is the focal point of the image, and takes up the foreground, but it's part of a model of the whole city. The buildings and other structures are complete and look to be an artists rendition of how they would've looked in ancient times. INeverCry 07:27, 11 December 2013 (UTC)

December 09

A "Favorite" images gadget

Hi, all. During a recent discussion with some community members it emerged that some are missing the capability to quickly mark an image as a "favorite" in commons. This kind of social feature is present on many photo sharing sites such as Flickr or 500px. While Commons is certainly a different type of site with a different mission, I can see the merits of having such functionality. In the aforementioned sites this functionality is coupled with a notification of the author of the image that got added to the favorites. This provides a small token of appreciation and arguably some gratification to the contributor. We do have a similar system in place on commons: Thanks. However it is neither coupled to the concept of "favorite Images", nor is it easy to Thank for an image upload currently. Thanks is geared towards edits on pages and tied to specific page revisions.

I have written a gadget to bring this functionality to commons in a simple and robust way. The gadget can be found here: MediaWiki:Gadget-Favorites.js. In anticipation of a controversial discussion about the merits of social features on commons I have not yet added it to the preferences. But it can be tested already by adding


to your userscript page. (enable it on the Gadgets tab in the preferences --Dschwen (talk) 03:03, 11 December 2013 (UTC))

The gadget does the following:

  • On image pages a new tab (Fave/Unfave) is added on top. Clicking adds/removes the image to your Favorites list
  • Favorite images are stored in user space at Special:MyPage/Favorites
  • Adding an image to the Favorites will send a Thanks notification to the uploader (the gadget automatically determines the first revision of the file page, which is created at upload time and sends a thanks for it)

Some of the technical details (and code review points by User:MarkTraceur (thanks!)) are on the Gadget's talk page. There is quite a bit of magic going on to make sure that edits go through, that needed network traffic is minimal, that manual edits to your /Favorites page are preserved, and that the gadget feels responsive (local caching).

For now I'm interested if this kind of social functionality would be appreciated on commons or not. --Dschwen (talk) 18:55, 9 December 2013 (UTC)

  • Support, this seems like an easy way to build a short list of media for articles, and a simple way so spread wikilove. Liamdavies (talk) 06:01, 10 December 2013 (UTC)
+1. Cool. :) JKadavoor Jee 07:03, 10 December 2013 (UTC)
@Dschwen: Cool, but right now, it doesn't work at all. See File:Fave bug.png. ;o) I use Chrome on Win7 if it matters. Yann (talk) 08:35, 10 December 2013 (UTC)
This is a hilarious bug. Sorry, so far I only tested it on the vector skin. I'll have a look.--Dschwen (talk) 13:59, 10 December 2013 (UTC) Since I'm using the add portalLink utility function and act on its return value this might be a mediawiki script bug. --Dschwen (talk) 14:46, 10 December 2013 (UTC)
Ok, fixed. Works in monobook now as well. I tried adding the button to a section that does not exist in the monobook skin. This returned a null reference which oddly referred to every link on the page when using it to change the text of the newly inserted button. --Dschwen (talk) 18:07, 10 December 2013 (UTC)
Yes, it works now! Thanks! Yann (talk) 18:59, 10 December 2013 (UTC)
There is a little glitch, but it doesn't prevent your great tool from working: if I add an image, and then that this image is deleted, and then I add again an image, the new image is not inside the "gallery" tag in my list of favorites. Regards, Yann (talk) 19:18, 10 December 2013 (UTC)
Ok, thanks. I did not anticipate the opening and closing gallery tags to be on the same line. I'll work on a fix. --Dschwen (talk) 20:00, 10 December 2013 (UTC)
✓ Done. Gallery tag placement is now cleaned up and does not affect insertion anymore [5] --Dschwen (talk) 22:46, 10 December 2013 (UTC)
  • Dear Dschwen: Thanks so much for this very cool experiment! I really like your proof of concept, and think we can learn a lot from this trial. A number of users we've talked to in recent months have expressed an interest in this type of feedback tool, which offers three benefits in a single feature: 1) bookmark media files I find interesting; 2) share my appreciation with its uploader; and 3) surface useful content on our sites. Personally, I find this feature really useful, particularly if we can integrate it more closely with the Media Viewer we're now developing as a beta feature. Based on responses to your gadget from other community members, we'd be love to help you enhance it in coming weeks (for now, thanks so much for fixing the bugs I reported so quickly). From a design perspective, we will also want to think about is how this approach might integrate with our other user interface goals (e.g. how does a 'Favorite' list fit in next to a 'Watchlist' at a time when we are trying to simplify the overall user experience across our sites?). Overall, this seems like a good example of how community members like you and the foundation's development teams can collaborate to experiment with new ideas. I love that you responded so productively to user wishes for a fave-like feature during our recent roundtable discussions -- and that you are now working with our developers to get code reviews and suggestions for improvement. This process is very inspiring to me, and I hope that we can spark more collaborations like these on a regular basis. :) Thanks again for this wonderful step forward! :) Fabrice Florin (WMF) (talk) 19:14, 10 December 2013 (UTC)
Should be a gadget. Where's the harm? Saffron Blaze (talk) 02:28, 11 December 2013 (UTC)
Ok, added to the preferences. --Dschwen (talk) 03:03, 11 December 2013 (UTC)

December 10

Company logos on uploaded infographics

A recent post on wikipedia's talk page about South West England pointed to File:Regional profile of the South West.png which has been uploaded to commons, asking whether it would be useful for the article. Taking a quick look at it I see the ONS copyright statement and logo are included in the picture. I have a vague memory of some guideline or policy that the producers logo shouldn't be included - but I can't find the relevant info. Can anyone point me to relevant rules or comment about the acceptability/suitability of the graphic?Rodw (talk) 18:55, 10 December 2013 (UTC)

I think you are looking for Commons:Watermarks. However, this was probably created mostly with photographic works in mind, where they are much more disturbing. --El Grafo (talk) 19:17, 10 December 2013 (UTC)
Thanks - although it is not really a watermark. On a bit more checking similar files are in Category:Content created by the Office for National Statistics.Rodw (talk) 19:29, 10 December 2013 (UTC)

Wikimania 2015 jury

The jury that will be judging the bids for Wikimania 2015 was just announced and on it is one of our own: Ralf Roleček! He just posted on Commons:Forum#Wikimania_2015_Jury asking input from commons members on his decision making. He explicitly ran as a commons contributor when applying for a jury seat. So let him know what to consider when choosing the location of Wikimania 2015. --Dschwen (talk) 00:23, 11 December 2013 (UTC)

Need help notifying the author of an image

I am trying to notify the author of an image I would like to use, but I am new to WikiMedia Commons and can't seem to find a place to contact the user. Where should I look on the users page so I can send him some sort of message or email?

Thank you! -- 15:41, 11 December 2013‎ User:Baseklas

For starters you could tell us which image you are talking about. --Dschwen (talk) 15:52, 11 December 2013 (UTC)
Assuming the author uploaded his own image, he or she has a user page and a user talk page you can reach from the image page (look in the File History section partway down the pageunder User and click the link there). You can contact uploader on that talk page. Alternatively, look by the right margin of the user or user talk page for the section Tools; click this and look to see if one of the lines that results is Email user (not everyone has this enabled) and click that link. If it is a 3rd party image uploaded by someone other than the author the source field is often a good start for tracing it back but there is no general method. Dankarl (talk) 18:31, 11 December 2013 (UTC)

Identify video data

How can I see the upload date or uploader name of a video on Videos by Al Jazeera of the 2008-2009 Gaza War, e.g. this one

click this icon →

? --Wickey-nl (talk) 10:05, 9 December 2013 (UTC)

Hi, below the thumbnail of the video there's a small icon (right next to the figure legend, I marked it in your example). Clicking the icon should bring you to the file description page (in this case: File:Aljazeeraasset-AftarGazaWar1458.ogv) which contains the required information. Hope that helped? --El Grafo (talk) 10:36, 9 December 2013 (UTC)
Thank you. Not obviously. Usually such icons are used for getting full screen. For pictures, I normally click on the image itself to get the description. May be another icon is preferable for audio/video. --Wickey-nl (talk) 11:31, 9 December 2013 (UTC)
There is no such icon if we use the video in a gallery or use without the "thumb". In such cases, we have to play the entire video to see the link that embedded in the last frame. This example has no such embedded link too. A CC license violation? JKadavoor Jee 13:16, 9 December 2013 (UTC)
When I (running Linux, in case that matters) click on the thumbnail, the video starts running. When I hover my cursor over the video, a bar containing various buttons appear at the bottom of the video. I can click on "menu" and then on the link next to "Title" to get to the file description page. So it's possible to get there, but it should be much more prominent, imho. --El Grafo (talk) 13:27, 9 December 2013 (UTC)
I've only "play" option in Chrome, IE and FF in Windows. :( JKadavoor Jee 16:04, 9 December 2013 (UTC)
I have Linux too. If no link on Windows (after clicking on play!), it would violate the license indeed, in case no attribution in the video itself
  • As an icon for the link to the file description I thought of the notice icon:
Notice {{{1}}}

. --Wickey-nl (talk) 06:59, 10 December 2013 (UTC)

  • Hi Wickey-nl, J and El Grafo: Thanks for reporting this issue, which our Multimedia team is aware of. We agree that there should be an easier way to access file information for a video and have brought up this issue with our partner Kaltura, with a recommendation to add a direct link to the file info page more prominently in their video player UI. Currently, you can find that link if you click on the 'Menu' button, which puts it two clicks away. We would also like to provide author/source attribution more prominently in future versions. Our goal for next quarter is to integrate video and audio files into the new Media Viewer which we are currently beta testing with still images, so issues like these would be addressed in a consistent way across all media types. And yes, the (i) icon is being considered as a possible solution for letting people know where to find more info about a file. We'll post more about this once we have some proposed mockups or early beta versions for us review together. To be continued ... Fabrice Florin (WMF) (talk) 18:48, 12 December 2013 (UTC)
    • Thanks for your info. --Wickey-nl (talk) 17:00, 13 December 2013 (UTC)


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):

Thank you! Philippe Beaudette, Wikimedia Foundation (talk) 04:41, 11 December 2013 (UTC)

On December 7, 2013, the Wikimedia Foundation (“WMF”) received a Digital Millennium Copyright Act (“DMCA”) notice regarding this image titled “Barack Obama first meeting with Nelson Mandela.” The photograph helps to illustrate, among others, an English Wikipedia article regarding Nelson Mandela, and specifically the section which discusses Mr. Mandela’s retirement and failing health.
As the Commons file indicates, the photograph in question appears to have been taken by David Katz during his tenure as an aide to then-Senator Barack Obama. Ordinarily, as many Commons users are aware, works prepared or produced by U.S. federal government employees reside in the public domain, and this photograph has been labeled as such by the community. However, Mr. Katz has argued that he took this photograph outside of the scope of his “official duties”, which is a factual dispute that we are not in a position to immediately resolve based on available evidence.
WMF is strongly committed to investigating and challenging takedown notices which it believes lack proper legal basis with a reasonable degree of confidence. WMF has concluded that, for now, it shall remove the image from Wikimedia Commons (and, because it is hosted on Commons, from Wikipedia as well) pursuant to WMF’s published DMCA procedure.
Because, in our view, serious concerns remain about the alleged copyright status of this work, WMF will continue to research whether the image in fact resides in the public domain. We have retained outside counsel on this matter, and are currently engaging in active consultation with them to determine the likely copyright status of the image in question. We will return to the community on this issue once we have gathered sufficient additional information regarding the facts of this case.
Please note that, as always, under DMCA §512(g)(2)(B), any person may submit a counter-notice to contest removal of the content when appropriate. Upon receipt of a valid counter-notice, WMF will notify the takedown sender of the counter-notice, and the takedown sender will have fourteen (14) days to file a lawsuit to restrain the restoration of the content before WMF will repost the content.
Rkwon (WMF) (talk) 05:44, 11 December 2013 (UTC)
Legal Counsel
Thank you for this detailled explanation. I was just wondering the reason for this DMCA notice. Regards, Yann (talk) 05:56, 11 December 2013 (UTC)
Yes, and thanks for deciding to look further into the copyright status of this image as well. I'll be interested to hear what you conclude. --Avenue (talk) 09:13, 11 December 2013 (UTC)
Hopefully a Commons user will soon put in a counter-notice using the easy on-line forms, on the good-faith understanding (as supported by published interviews) that David Katz was a paid aide/driver for Obama at the time, and therefore only there to do his job on the tax-payer's dollar. As someone living outside the USA, I do not think it would be helpful for me to agree to "consent to the jurisdiction of a federal court in the district where your service provider is located" as I do not really think it is all that legally meaningful. If anyone wants to see what is under discussion, the photograph is published by the NY Times here and higher resolution here.
By the way, can someone please explain why the logs have been suppressed for this image? This appears excessive in order to comply with a DMCA notice where there appears to be no privacy issue or special concern for the uploader (I'm assuming the original uploader has been notified). -- (talk) 10:07, 11 December 2013 (UTC)
Hi @: There haven't been any suppressions performed for this image, it's just that Rubina added a wrong link to the file. (I fixed it, so you can see this yourself.) odder (talk) 10:15, 11 December 2013 (UTC)
Thanks for the correction, that's good to know. Unless I'm misunderstanding something about the original wikicode, there was a source link to a part of the Whitehouse Archives to an image that was a variation of this one. If the Whitehouse Archives were (or are still) holding this image as PD, this would be useful to know in order to give this DMCA notice some context. Thanks -- (talk) 10:35, 11 December 2013 (UTC)
Not sure at whom you're addressing this question, perhaps it might be useful to ping @Rubina directly. The now-deleted file description page says "Photographer David Lee Katz, creator of the original photo, was on the staff of Senator Obama, with the title of Special Assistant, and the meeting was scheduled by Obama's Senate scheduler." which suggests that it shouldn't be too hard to verify this. Perhaps a FOI request to the Senate, or a call, could help? odder (talk) 10:48, 11 December 2013 (UTC)
Actually these same assertions are all repeated in a number of newspaper articles, so already verified. Unless Katz is claiming that the press has all their facts wrong, there seems little more to do to demonstrate that use on Commons can proceed on good faith unless more solid evidence is forthcoming. Perhaps Katz was there on holiday and not employed to be (Senator) Obama's aide at that time but just happened to be giving Obama a lift when on his official, if last minute, schedule of meetings? If so, he ought to be able to provide some evidence of this, or maybe he can ask the Senate/Whitehouse to confirm why he was called Obama's aide.
Anyway, WMF legal are paid to sort this out, there appears to be plenty of material to work with. -- (talk) 11:09, 11 December 2013 (UTC)
I think what David Katz is arguing is that photography is outside of the scope of his “official duties” of a driver. --Jarekt (talk) 19:16, 12 December 2013 (UTC)

Just for Info: Is working for a US Senator to be considered the same as working for the US Federal government? Sounds like the author had no direct connection to Fedgov but was directly employeed by Obama as a driver. --Denniss (talk) 10:21, 13 December 2013 (UTC)

According to the newspaper articles he was employed as a Senator's official aide rather than a contract driver. He holds the same job today I believe. I imagine that Senators in the U.S. pay for official aides out of the tax payer's purse, though perhaps a student of U.S. politics might be able to supply a specific reference. -- (talk) 10:53, 13 December 2013 (UTC)

Vernacular names

On Phalaenopsis speciosa, for example, Фаленопсис специоза is given as a "Vernacular name". Isn't this rather a transliteration?
Season's greetings, Rich Farmbrough, 10:02 13 December 2013 (GMT).

Yes, and "vernacular name" is wikilinked to w:Common name, where they say that scientific names are Latinized. Maybe a Cyril transliteration can be the common name, or this is a clumsy/incomplete attempt of i18n. –Be..anyone (talk) 23:07, 13 December 2013 (UTC)

File:Office of Government Information Services logo.jpg

I think this file has a render bug. In lower resolution shows only blue text, I'm not sure why, because it's jpg not svg, so this problem shouldn't appear. --Rezonansowy (talk) 20:04, 13 December 2013 (UTC)

I applied a hard-purge to force new thumbnails to be generated. --McZusatz (talk) 20:47, 13 December 2013 (UTC)
Thanks! --Rezonansowy (talk) 21:02, 13 December 2013 (UTC)


Long time I've believed this arrangement was fine, but I'm not convinced any more, and I'm unable to decide about it.

Should the term "Riders" on Commons be reserved to *Horse* riders (= arguably being the most common, somewhat even defaultive association for "rider"), or, stick to the English language defining "riding" with more generosity and host any form of such occupation, including cyclists, bikers, camel riders etc.? Either resolution applies also to the gerund Riding.

It could be that the category itself, "Riders" (and "Riding"), is altogether redundant and can be redirected to Equestrians (and respectively Horse riding) - benefitting other "riding" media such as the cyclists being cleared from that ambiguous, unimportant parent. Basically I was the last one who handled this whole tree and designed its current shape (a year ago), so I'm feeling a bit responsible when the category as it is now feels somewhat unnatural, and I can't ignore the difficulty by users still categorizing their horse-rider images casually at "Riders" instead of one level deeper as they should. They can't be blamed. What do you think?

I'd much appreciate any opinion from as many of you. Orrlingtalk 03:47, 14 December 2013 (UTC)

It's complicated by the fact that, while you are right that "horse rider" is the most common, there are some proper noun phrases, like w:Freedom Riders that refer to other ones. I'd lean towards redirecting Riders to Equestrians, but I'm not particular convinced of that, either. JesseW (talk) 04:18, 14 December 2013 (UTC)
  • In an urban environment (and remember that the world just become over 50% urbanised) I reckon bicycle riding would be the most common form of riding. Because of that I would leave it alone, it does best as a 'catch all'. Liamdavies (talk) 06:46, 14 December 2013 (UTC)
yeah but we have Category:Cycling. People riding bicycles aren't referred to as "riders", but as cyclists. Then why double-parent them. Orrlingtalk 08:02, 14 December 2013 (UTC)
There is a difference between a cycle and a rider, one rides a bicycle, sure we may refer to them as cyclists, but that doesn't change the fact that they are also riding the bicycle. Liamdavies (talk) 08:27, 14 December 2013 (UTC)

Derogatory terms in file names for archive photographs

Though I would defend keeping original descriptions of photographs on image pages, there are a number of war time photographs that I have uploaded over the last year or so that contain derogatory terms for people featured. As an example, File:A further indication that not all Japanese fight to the death is this bag of 221 Nip prisoners of war HD-SN-99-02963.jpg appears offensive to modern eyes as it uses "bag of" to refer to the group of prisoners, "Nip" as a derogatory slang term for Japanese and appears to be calling these prisoners cowards for not fighting to the death (it is also reasonable to conclude that the prisoners were given no choice in being photographed while being held).

I have no problem in the original title being included in the image description for archive traceability purposes, but I am increasingly uncomfortable with the appearance of the filenames, which in the case of my uploads were automatically generated during the upload process. What are people's general thoughts on what should be good practice for Commons and the possibility of mass renaming to something more respectful? Thanks -- (talk) 11:35, 14 December 2013 (UTC)

Media viewer uses file name without extension as title. So I don't like these type of lengthy "file description" like titles. "NARA FILE #: 026-G-4669 WAR & CONFLICT BOOK #: 1304" is the preferred title for me in this case. But I don't see the need or recommend a mass file rename if it generate only "similar not so meaningful" names. You can rename case to case, if they contain "derogatory terms". Jee 12:10, 14 December 2013 (UTC)
I've moved it to File:Japanese POWs circa 1945.jpg, while keeping a redirect. - Jmabel ! talk 18:49, 14 December 2013 (UTC)
Checkmark This section is resolved and can be archived. If you disagree, replace this template with your comment. Jmabel ! talk 18:49, 14 December 2013 (UTC)

Resolution:Media about living people

SJ just announced the launch of Resolution:Media about living people. JKadavoor Jee 06:20, 6 December 2013 (UTC)

The changes to the resolution seems to be only tangentially related to Commons. We should be "ensuring that all projects [] that describe or show living people have policies in place calling for special attention to the principles of neutrality and verifiability in those articles". Luckily Commons do not usually have articles that have to be verifiable and neutral. And individual images are rarely neutral or verifiable. But we should be aware that galleries related to living people should meet those thresholds. --Jarekt (talk) 13:41, 6 December 2013 (UTC)
The resolution says we should have policies in place, not just awareness. I'm not aware of any policies here that address galleries' neutrality and verifiability. (The closest things I can think of are the bits of COM:PSP prohibiting "anything apparently created and/or uploaded for the purpose of vandalism or attack", and those don't seem expansive enough to satisfy the resolution's requirements.) Also, the change you quote might only apply to articles/galleries, but I think all the other changes to apply directly to Commons. --Avenue (talk) 14:09, 6 December 2013 (UTC)
Galleries are not articles.
In terms of verifiability Commons does rather a good job (compared to other media collections on-line) with regard to verifying releases and deleting those that are contested or confirmed as being copyright violations. In terms of "adding media about living people", the Photographs of identifiable people policy is well tested and provides considerable protection for living people (or those concerned for them) that raise a complaint that an image is potentially defamatory or an inappropriate invasion of privacy.
If someone wishes to use the resolution by the WMF board to drive improvements on Commons, I would be happy to see some Requests For Changes to existing policies, or some simple proposals on the relevant policy/guideline talk pages, that are well thought out and avoid conflating language directed at Wikipedia articles from the local issues on Commons of how best to ensure that media added to this project depicting living people are always well handled, and meet both legal requirements and our current Commons community processes with regard to the "moral obligation to behave ethically with regard to photographs of people" (quoting from current official COM:IDENT guideline).
To be honest, I don't see any significant changes needed in order to meet the resolution. Our policies already reflect everything needed, even if there are questions of where the community draws the line on educational use or what might constitute an invasion of privacy or what counts as ethical behaviour (such as on-going discussion on courtesy deletions). If the WMF board of trustees have something specific in mind, it would be best if that were spelt out more clearly if on-going improvements to this project are neither sufficient nor effective. -- (talk) 14:46, 6 December 2013 (UTC)
In some way, it could be argued that any page with textual information (such as this page) is an article. However, COM:IDENT already handles the file namespace, and harassment is already seen as unacceptable, so we probably only need to add a minor footnote to some page, if anything needs to be done at all. --Stefan4 (talk) 15:06, 6 December 2013 (UTC)
I agree that Commons:Photographs of identifiable people guideline handles most cases. En Wiki has several policies related to images of living people, like en:Wikipedia:Biographies_of_living_persons#Images or en:Wikipedia:No_original_research#Original_images, and they seem to be in synch with our guidelines/policies. At some point we actually tracked content related to living people but we decided to stop tracking it due to the impossibility of maintaining such a list. --Jarekt (talk) 16:28, 6 December 2013 (UTC)
SJ posted this notice on Commons:Courtesy deletions; so we can assume they give much importance for that proposed policy. As a person who has communication with many of them, I can assure that they expressed it many times. Unfortunately we faced much resistance from people on the development of that proposed policy. Hope we will restart to build it, strong.
Another important one is Commons:Photographs of identifiable people. Recently we made an amendment there; that also after overcoming a huge resistance from a single member. In that discussion, we even consulted Mdennis (WMF) and followed her suggestions. There we arrived in a consensus to make an amendment on Commons:Contact_us/Problems#Report_abuse too; but unfortunately that page is edit protected. I contacted Maggie, and she forwarded to the OTRS team; still waiting for a response.
Commons:PSP also need to be updated (agree with Avenue); currently there is a conflict with our sister projects. They remove media contents based on their BLP policy; but we refuse to remove those contents from the categories/galleries linked to that pages. :)
I fully disagree with on his argument "If the WMF board of trustees have something specific in mind, it would be best if that were spelt out more clearly if on-going improvements to this project are neither sufficient nor effective." IMHO, this approach is not helpful at all. We should open to listen them and ask them if any further clarification is needed. JKadavoor Jee 16:03, 6 December 2013 (UTC)
I doubt that what I wrote above was an argument, it was intended to be a viewpoint. As you are regularly exchanging personal communications with the WMF board, perhaps you could ask them if they "have something specific in mind", otherwise it looks like business as usual.
Of course any WMF board member is welcome to write their own views here, rather than using you as their mouthpiece, a position that I would find quite uncomfortable. -- (talk) 16:14, 6 December 2013 (UTC)
So you didn't see Sj's comment there almost two week above and on today? Further, there is a dedicated noticeboard for them. JKadavoor Jee 16:22, 6 December 2013 (UTC)
Hey all, the amendment of the resolution is to make sure that it applies to all things on our projects: both text and images and media. I think wikilawyering over whether it applies to certain types of pages misses the point: which is that the board feels Wikimedians should exercise equal care when dealing with all portrayals of living people on our various projects. -- Phoebe (talk) 17:37, 11 December 2013 (UTC)
p.s. in reply to User:Fæ, I haven't had any personal communications with anyone beyond what you can see on my own talk page and the board noticeboard, nor have we had any personal communications sent on to the board list :) I think you are right in that this resolution is not meant to drive to a very specific change, but rather is meant as a statement of principle that we can use to guide the development of process and policy. However, that doesn't mean that everything is great -- we should examine our policies, practices and individual decisions with care, much as with the past resolution about images of identifiable people (which is certainly an ongoing issue, AFAIK). (I speak for myself here, not the full board!) best, -- Phoebe (talk) 17:39, 11 December 2013 (UTC)
Seconding Phoebe's comment: the only communications about such topics have been public, on Meta. I posted an update to the courtesy deletions page because Jkadavoor pointed to it as the most recent place where relevant conversation (referring specifically to the BLP resolution and its specificity about text) had taken place on Commons. While the BLP resolution is one of the rare times that the Board has commented directly on project policies, this seemed like a straightforward update to clarify the original intent. --SJ+ 05:09, 12 December 2013 (UTC)
It is good to clarify that the 'living people' resolution applies to all content on all projects, however this WMF board resolution amendment is sloppy work I am sorry to say. In addition to it not amending 'articles' in "principles of neutrality and verifiability in those articles;" the resolution may also include media of unidentifiable people, which is common on Commons, whereas prose about unidentifiable people is not common on Wikipedia. I could easily read this resolution to mean that Commons must only accept photographs of identified people that are verifiable, which probably requires deleting a million odd photos of unidentified people and photos of living people where we don't have OTRS correspondence from the subject approving the photograph (model releases). That change of scope would solve many problems .. ;-) John Vandenberg (chat) 17:30, 14 December 2013 (UTC)
Not having an indication of model release may be one reason to delete; but as always one should use good judgement. Just as not having a citation can be an acceptable reason to remove content; though we don't use that to remove the 99% of Wikipedia sentences that aren't directly cited to a source. Here too, context matters: something controversial or misrepresentative or unkind or puffery (here: photoshopped?) requires better verification than a casual aside. And just as we accept the uploader's word for claims about copyright, we can accept their word for claims about releases - again, using common sense to avoid flickrwashing and the like. --SJ+ 18:42, 14 December 2013 (UTC)
Sj, inline citations are a red herring: that is the method of recording the verification, rather than the fact that everything must be verifiable to reliable sources, especially for BLPs. Wikipedia has speedy removal of contentious or libelous material, however their policy also includes process for questioning and removing any fact, no matter how minute, if a question on the talk page doesnt result in a reliable source being provided. So what does verification mean on Wikimedia Commons? What aspects of a photograph need to be verifiable? If it is that images must be supported by reliable sources, because the WMF board is now raising standards regarding media to the same level as text, then we have many photos that can only be verified by the photographer, and often that photographer isnt responding to email or has long gone with no forwarding address. John Vandenberg (chat) 19:12, 14 December 2013 (UTC)

Église de la Sainte Vierge

The Category:Église de la Sainte Vierge has to be renamed. This is not a French version of Category:Saint Mary churches but a highly specific chapel ship in Marchienne-au-Pont. I have no specific evidence that it is a Saint Mary church, expect that the other uploader classified it as such. It is a catholic church. See website. The the ship can of course be moved but has been there a long time, but I hesitade to apply geografic classification, except for the individual images.Smiley.toerist (talk) 10:21, 15 December 2013 (UTC)

This category Église de la Sainte Vierge appears to have been created in 2011 for images of the actual église Sainte Vierge. See [6]. In 2013, another user created a duplicate category for the same church, where he misspelled the name as Église de la Saint-Vierge de Marchienne-au-Pont. He removed the file that was in the first existing category [7] and he placed it in his duplicate category. What is the best name for the category? It is certainly not the duplicate category with the misspelled name. The official name of the church seems to be église Sainte Vierge [8]. But it is quite possible that it is commonly known as église de la Sainte Vierge. Users who know that area may be able to tell. If one believes necessary to specify the location in the filename, the usual way to do that on Commons for Belgian churches seems to be Église Sainte Vierge (Marchienne-au-Pont) or Église de la Sainte Vierge (Marchienne-au-Pont). Any of those category names or others could probably be acceptable. You could discuss it with users of Commons who are active on that topic and see what consensus there is for the best name for a category for that church. There just seems to be no category yet for the chapelship. It's unclear why the uploader of the photos of the chapelship categorized them into the category for the église Sainte Vierge. It could have been a simple mistake. This uploader contributed to Commons recently, so the best thing would be to ask him. -- Asclepias (talk) 15:30, 15 December 2013 (UTC)

Broken template

Can someone take a look at template:untitled. It seems to be broken and is displaying "script error" when invoked (example at File:Francisco de Goya, Saturno devorando a su hijo (1819-1823).jpg). I can't see anyting obvious in the last edit (July) made to the template. SpinningSpark 11:55, 15 December 2013 (UTC)

✓ Fixed. Lots of pages need purging, though. --Zhuyifei1999 (talk) 15:03, 15 December 2013 (UTC)

Lucien Waléry vs. Stanislaw Walery


As per the description in Category:Lucien Waléry and in Category:Stanislaw Walery, there is often a confusion between the two photographers. At least this image is attributed to both of them. I suspect that there are more misattributed images in Category:Male portraits by Lucien Waléry‎. Thanks for your help, Yann (talk) 14:51, 15 December 2013 (UTC)

Cut-off year for copyright expiry

In terms of 70-years-after-death (yad), we can safely assume that the works of an author made in the 18th century or before are out of copyright. If they were working in 1854, age ten, and lived to be a hundred (which is feasible), their work is still in copyright. Do we have a year, which we have agreed is reasonable to assume makes a work out-of-copyright without further evidence (e.g. when the author is unknown)? For 70yad a date of 1832, incremented annually, would seem reasonable, if not over-cautious. Andy Mabbett (talk) 12:59, 14 December 2013 (UTC)

Considering life expectancy, I think that works made before 1880 are safe, after some basic home work is done for searching the exact name and dates of the author. It means that starting creation at 20 (born in 1860), the author didn't live passed 83 years. Not impossible, of course, but only a very very small percentage of people got to this age at that time. Regards, Yann (talk) 13:22, 14 December 2013 (UTC)
There was a study done on this for the ARROW digitisation program; the longest case it was able to identify was 151 years (published at 17, death aged 98 in 1940, public domain 1 January 2011). However, examples like this are incredibly rare - the next longest they could identify was 146 years. (See p. 14 of this study.) I suspect in practice almost all cases would be covered by a 120-year threshold. Andrew Gray (talk) 14:13, 14 December 2013 (UTC)
However, be extra careful with cinematographic works. In "years after death" countries, there are great differences in determining who the author is. In some countries, the author is a single person. In other countries, anyone who was involved in the work is an author, and you have to consider the death year of everyone, including child actors. An example:
  • The w:Copyright Duration Directive tells that the copyright term expires 70 years after the death of the longest living of four people, regardless of whether these people are authors of the work or not.
  • The directive also tells that it doesn't have the effect that the term of pre-existing works is shortened. Thus, you have to consider the terms under the former laws too. For example, under the old Swedish copyright law, the term expires 50 years after the death of the longest surviving co-author, and an author is anyone who makes a contribution to the film which is above the threshold of originality. Actors only seem to hold w:related rights and so presumably aren't co-authors of the cinematographic work, but lots of people who make background work may count as co-authors.
Thus, in Sweden, you have to determine whichever is longer: 70 years after the death of the longest living of four people, or 50 years after the death of the longest living of lots of people (possibly dozens or hundreds of them). If the film involved children as co-authors, a much longer copyright term can be expected. --Stefan4 (talk) 00:49, 16 December 2013 (UTC)
1832 is over-cautios, IMHO. If the author is unknown (not a known author with unknown year of death), accepting works from around 1860 should be reasonable, for two reasons: First, there will be very few 19th century works here created by authors aged less than 15 to 20 years. So, assume an author aged 15 in 1860, thus born in 1845, would have needed to reach an age of more than 98 in order for his or her work to be still protected as of 2014. Though this is possible, even if the combination of a publication at such young an age and such longevity is exceedingly rare (see the study mentioned by Andrew Gray), another factor should be considered: If the author is really unknown, the work might be PD anyway in countries which have a protection term of 70 years after publication for anonymous works, see Template:Anonymous-EU. This is true for many countries with a 70 years after death protection for works by known authors. Gestumblindi (talk) 04:06, 16 December 2013 (UTC)
So shall we adopt 180 years ago for 70yad countries? That is similar to what I was using when at some point I was replacing all {{PD-old}} with {{PD-old-100}} if the work was dated before 1800. --Jarekt (talk) 14:29, 16 December 2013 (UTC)
So you are essentially suggesting that we assume that an unidentifiable death year means death 110 years after the work was created, if no other information is available? That would be a safe solution, but is it necessary to use as much as 110 years? --Stefan4 (talk) 15:02, 16 December 2013 (UTC)
180 years? Sorry, but this is not reasonable. I think we should not question if the work is older than 140 years. Personally, I use 135 years. Regards, Yann (talk) 15:07, 16 December 2013 (UTC)
LOL, yes 180 years is way beyond the spirit of the precautionary principle. When I was helping with the British Library relationship, I believe that a rule of thumb that applied to their on-line publications was that anything published before 1874 was not worth spending the time researching copyright for, and could be considered "safe" to be released as public domain unless there was an obvious claim of copyright that was visible without much research (such as the archives being under a particular donation/trust condition). Could we refer to a policy held by an august body such as this as "reasonable"? -- (talk) 15:19, 16 December 2013 (UTC)
As said above, I think accepting images from around 1860 (so approx. 155 years after publication) should be very safe, as the most extreme case found by the study mentioned was protection lasting for 151 years after publication. Gestumblindi (talk) 15:56, 16 December 2013 (UTC)
Yes, it looks like we are in general agreement that a cut off somewhere between 1860 to 1880 makes sense for assuming that any published work before that year can be considered PD without significant doubt. I still think that adopting a standard from another large institution (such as one of the well known national archives) makes this easier than having a long community debate. Any thoughts on what would be the most realistic process to get a credible consensus to turn this into a guideline for admins and future uploaders? Perhaps this discussion is sufficient considering there are some experienced uploaders and contributors taking part already? -- (talk) 16:21, 16 December 2013 (UTC)
It varies a bit by the type of media. For example Leni Riefenstahl lived until 101 and the copyright on her first film (The Blue Light) will last about 141 years assuming no law changes in that time. Pablo Picasso lived to 91 however since his first works date back to 1890 he's going to generate copyright terms of 153 years. Things like paintings, short stories and photographs where young people can get things published (or at least preserved) are likely to generate more extreme cases than novels and full length films.Geni (talk) 17:57, 16 December 2013 (UTC)

Copyright status of photo from Australia.

I have a question about the copyright status of File:Crests of Fort Street High.png. I've looked at the copyright rules of Australia, but would like to make sure.

The items in the upper left and upper right was created circa 1911, so should easily be in the public domain.

The lower image is a colorised version of en:File:Faberest.png created in 1849, which is clearly pd. I don't think mere coloring, done in 1975, qualifies as a new copyright, but would like to make sure. Finally, the collage of the three items, then photographed, presumably requires a permissions statement from the photographer, which person is providing the image so should be straight-forward.

Any concerns?--Sphilbrick (talk) 16:22, 15 December 2013 (UTC)

Why do you think that the image created in 1911 is in public domain in Australia? Ruslik (talk) 19:19, 15 December 2013 (UTC)
Per the link, the time limits for copyrights was 50 years until amending in 2005. However, in 2005, the 50 years had elapsed. The new rules, depending on the publication date and author death date, were not applicable to works whose copyright had already expired. I don't think I need to identify the publication date or author death date, I simply need to observe that 50 years had elapsed by the time of the 2005 amendments.--Sphilbrick (talk) 22:00, 15 December 2013 (UTC)

File:Bundesarchiv Bild 146-1998-043-06A, Maasbrücke bei Rotterdam.jpg

As stated in the description (and by own knowledge), the name is wrong. But the image came from an archive. Is it advisable to change the filename to a correct one? --Stunteltje (talk) 07:54, 16 December 2013 (UTC)

I think a renaming would be desirable as the filename is really misleading. You can induce that by adding {{rename|Moerdijk Bridge 1940.jpg|3}} in the picture´s description (see Commons:File renaming). Additionally, it would be helpful if you registered the mistake at Commons:Bundesarchiv/Error reports so that the filename and the description can be corrected at the Bundesarchiv as well. --Rudolph Buch (talk) 13:52, 16 December 2013 (UTC)
Thanks for the link. I'll report first and wait for a response. --Stunteltje (talk) 18:47, 16 December 2013 (UTC)

December 16

New search (CirrusSearch) coming to Commons next week

Hi everyone! As you may or may not know, we're currently working on replacing the aging MWSearch setup with a brand new and shiny setup we're calling CirrusSearch. The new search engine is backed by Elasticsearch, and fixes a ton of existing bugs we've left in the old system for too long. A list of exciting things as well as most of the supported syntax can be found here on Right now, I'm planning to roll this out to Commons on Monday, December 16th as a beta feature. What does this mean for people? Hopefully, not a whole lot initially. The feature will be available in your preferences to try out as a beta feature. I hope you'll give a try after things are deployed Monday. I'll be sure to check back in on this thread to let people know the status, as well as to get any feedback you may have. Thanks! ^demon (talk) 23:45, 10 December 2013 (UTC)

I would consider a better search to be really important for Commons - our old search had a bad habit of turning up irrelevant and sometimes offensive results, which just put more ammunition in the hands of people concerned about our curation. Will be great to see this roll out. Dcoetzee (talk) 03:54, 11 December 2013 (UTC)
Thanks Chad for the announcement. That’s cool to have a new search − I guess many people would feel the same way than Dcoetzee.
I’ve been following a lot the discussion on CirrusSearch on the mailing lists, and how people devised ways to take advantage of Cirrus to tailor Wikisource & Wiktionary ; but I must say I am a bit at loss to have ideas for Commons. What is really possible with CirrusSearch?
For example, can we push Quality images a tiny bit higher? Rank images per their global usage? Cluster results by categories? I’m just throwing ideas as they come to me − surely there are lots of better devised ones to take from the extensive Commons:Requests for comment/improving search. Jean-Fred (talk) 09:46, 13 December 2013 (UTC)
The main focus of our first iteration is replacing the existing infrastructure with as much feature parity as we can get right (the infrastructure part is key, the current search setup causes no end of operational/engineering headaches). A lot of things are possible in the long run once we've done this. I've been noodling some ways that communities can help in ranking things (like you said, feature images being higher ranked, for example). Also, we've been talking with the Wikidata folks as well to find ways that we can make things like image metadata searchable. (Imagine, being able to say "Show me PD images of the Eiffel Tower taken by user so-and-so").
I had not seen that RfC page before, will have to take a look at that, thanks for the link!
Also, we just deployed the software about 30 minutes ago. The index is slowly filling up over the next hours, and should be available for you to start using via the query parameters (we'll turn BetaFeatures integration on as soon as the index is populated). ^demon (talk) 18:05, 16 December 2013 (UTC)
I just want to reiterate what Jean-Fred said. Please rank Quality Images higher as well. I realize you guys are focussed on the technological side, but this will only work well if you collaborate with the community on-wiki. Your answer about FPs suggests that you may not even have heard about the QI project. A tremendous amount of work has been and is being put in by community members to curate and assess the content we have. This work is an important asset that must be used by the search team. Finding and deploying a good search infrastructure is an important step, but it will be a waste of resources of the second equally important step of harnessing the work done by the community for assembling metrics to rank results is not taken. --Dschwen (talk) 18:17, 16 December 2013 (UTC)
Right, I was just using that as an example. Any implementation we'd come up with would be flexible so communities (plural, not just enwiki or commonswiki) can control these sorts of weightings (also plural). Right now, I'm thinking of doing something like where the community can define (via a MediaWiki namespace message, or something) templates that define a given example of "good" content (featured, quality, etc) or "bad" content (stubs, no licensing info, etc) and how much to boost/lower rankings as a result. ^demon (talk) 18:27, 16 December 2013 (UTC)
Ok, thanks that sounds good. --Dschwen (talk) 18:40, 16 December 2013 (UTC)

I suggest that once this is 'bedded in' in a few weeks time, that we encourage some user videos to be made to explain it. Most regular users will find the ability to do simple category intersections incredibly useful but may find the syntax a bit confusing, especially the idea of using numbers to force proximity or fuzzy logic matches. A one or two minute video walking through examples is a lot easier than reading about the syntax. -- (talk) 18:47, 16 December 2013 (UTC)

I've been wanting to put up a blog post since this is starting to get rolled out on a much wider basis to more and more users. Screenshots and clear language would be useful here. A video would be nice too--I lack the skills to really do it but I'd be more than happy to help explain Cirrus to anyone who's wanting to do such a video.
I'll also add, the page we've written on, mw:Search/CirrusSearchFeatures was explicitly written PD so all communities using CirrusSearch can use this as a basis for their own local Help pages. I hope that the information we've put there can help WMF communities in updating their local search documentation as this continues to roll out. ^demon (talk) 19:12, 16 December 2013 (UTC)

December 11


I'm trying to be as polite as I can while typing this ... why don't we just ban all signs and remove sign categories from Wikimedia Commons, if everyday someone is going to nominate some sign File:St johns church sign buffalo.jpg, File:GeorgiaTechHistoricalMarker.jpg, File:Leora brown historical marker.jpg, File:Lafayette Ninth Street Hill marker.png, File:Jackson Barracks Historical Marker.jpg for deletion? It is really frustrating to see so many nominations. I'm just thankful for Flickr. None of my signs will be nominated for deletion there. --Mjrmtg (talk) 22:46, 15 December 2013 (UTC)

Some signs are okay. Others aren't. It's frustrating to have to make so many nominations because people upload pictures of other people's copyrighted work.--Prosfilaes (talk) 06:37, 16 December 2013 (UTC)
Wikimedia Commons needs CLEAR, CONCISE explanations of what is allowable and what is not - the FOP page is not clear. How can you copyright File:Thomasville Dog Park sign.JPG? Flickr is looking better and better everyday. --Mjrmtg (talk) 22:56, 16 December 2013 (UTC)
How is OK for buildings only not both clear and concise?! --Dschwen (talk) 23:14, 16 December 2013 (UTC)
There are no clear and concise explanations if you don't understand copyright. You get a copyright (as you've claimed on that page) for taking a snapshot of the sign; the sign painter gets a copyright for carefully deciding and executing a drawing of a dog on that sign. Any sort of drawing, any extended text gets a copyright, and you may not make a photographic copy of it without permission.--Prosfilaes (talk) 00:35, 17 December 2013 (UTC)

Upload wizard error

I got a weird error when I uploaded File:Facebook users by age population pyramid.png. When I try to tag it for deletion, I get "edit conflict". Can someone delete it? The correct file is at File:P opulation pyramid of Facebook users by age.png. --Dennis Bratland (talk) 03:29, 16 December 2013 (UTC)

The revision #0 of the page named "Facebook users by age population pyramid.png" does not exist. - This is bugzilla:15430. -- Rillke(q?) 20:01, 16 December 2013 (UTC)

Undesirable characters in title

Hi, I am having a hard time to upload pictures from Angkor Thom. I try to give a descriptive title to everthing I upload to Commons, and therefore pictures from Angkor Thom include these 2 words in the title, but it will not work because at the very last step of the upload process with the UploadWizard I get an error, the title contains undesirable characters. Only after I skip "Angkor Thom" from the title, the upload is allowed, but taking it out and renaming the file later on shouldn't be the way to do it, especially when I upload dozens of files from this place. In MediaWiki:Titleblacklist I couldn't find the answer to this problem, any ideas? Poco2 01:38, 16 December 2013 (UTC)

You mean file names like File:Elefante, Angkor Thom, Camboya, 2013-08-16, DD 01.jpg? --Zhuyifei1999 (talk) 10:50, 16 December 2013 (UTC)
Yes, I had to update that file under the name "Elefante, Camboya, 2013-08-16, DD 01.jpg" and later on renamed to the current name (adding "Angkor Thom," inbetween). Poco2 12:52, 16 December 2013 (UTC)
Sandbox pages with the full title of the file created, no error occurred. Try to match the regexes, failed. Also I noticed that admins have tboverride right, so MediaWiki:Titleblacklist shouldn't be an issue. Could you check whether there is some hidden characters or not? @Rillke: Please have a look at this ^^. --Zhuyifei1999 (talk) 14:50, 16 December 2013 (UTC)
No hidden characters, and actually it happened again for all 35 files I uploaded today, as a example: "Bayon, Angkor Thom‎, Camboya, 2013-08-16, DD 03.JPG". Poco2 15:23, 16 December 2013 (UTC)
For hidden characters, there is now Commons:User scripts/Invisible charaters. I have not clue what is causing this error; File:Elefante, Angkor Thom, Camboya, 2013-08-17, DD 01.jpg uploaded as expected using UpWiz. -- Rillke(q?) 19:50, 16 December 2013 (UTC)
I have rebooted my PC, and tried to upload a further dozen of pictures with the title "Terraza de los Elefantes, Angkor Thom‎, Camboya, 2013-08-16, DD XX" with XX from 01 to 13 and got the usual error, see screenshot. The invisible characters test is negative. I renamed the files taking out "Angkor Thom," and the upload was successful Poco2 23:35, 16 December 2013 (UTC)
It is positive for your example "Bayon, Angkor Thom‎, Camboya, 2013-08-16, DD 03.JPG" which contains a LEFT-TO-RIGHT MARK (U+200e) at position 19. The move-script is, however smart enough to remove these characts; only UploadWizard is struggeling with them. The API-warning is:
-- Rillke(q?) 02:23, 17 December 2013 (UTC)
Thanks Rillke, it looks like you got the problem. When using that tool I was expecting a kind of feedback, red color, or something like that, I didn't pay too much attention to each single character. I have no clue what that "left to right mark" is and where it comes from, but whenn going through the filename with the cursor I realized that there is an invisible character there. All the best, Poco2 13:26, 17 December 2013 (UTC)
So indeed a hidden character (Left-to-right mark). Also try not to copy the page names directly since mw add the marks automatically. --Zhuyifei1999 (talk) 13:45, 17 December 2013 (UTC)

503 Service Unavailable

Though this error is not constant, over the last couple of months it has become a default "feature" for those of us who do large amounts of work on Commons using automated tools. This was reported here on bugzilla, but considering how long this has been an issue, could we have a better summary of what exactly changed in the operational set up to cause this problem and what might solve it?

It would be a pity to start training all new bot-writers or automation users to design all tools to work-around this persistent error rather than actually resolving it. Thanks -- (talk) 13:53, 10 December 2013 (UTC)

If somebody knew what exactly changed in the operational set up to cause this problem and what might solve it, it would probably be solved already... --AKlapper (WMF) (talk) 15:22, 10 December 2013 (UTC)
Not really, if it were something incredibly difficult to fix. If there are lots of unknowns behind this, could we at least start some sort of automatic daily test and report so that we can see if this is getting better, worse or demonstrates some pattern that might help with diagnostics? If every time people in the office turn on their cappuccino machine we get more 503 failures, that would be worth knowing... -- (talk) 15:37, 10 December 2013 (UTC)
i think you mean have wmf log/graph 503 responses (do we already do that? Im not sure). A daily test wouldn't be useful since errors aren't distributed randomly so random tests probably aren't good. There was a report in bugzilla (bugzilla:57026) about double rendering of pages, that seems likely culprit. Do you have some examples of pages where this is happening currently? When the error happens are they repeatable? Are they always the complex pages? Bawolff (talk)`
As in the bug report 503 errors occur in trivial text changes as well as image uploads. I'll think about adding a log to some of the stuff Faebot does, probably not this week though. -- (talk) 18:30, 11 December 2013 (UTC)
Trivial text change on a complicated page, or a trivial text change on a normal page? Bawolff (talk) 23:09, 11 December 2013 (UTC)
Both. -- (talk) 09:51, 13 December 2013 (UTC)

I happened to stumble upon which gives a graph of 5xx responses. Bawolff (talk) 09:41, 13 December 2013 (UTC)

Good that it can be done. A more specific breakdown by type and some indication of what makes for good or bad performance (i.e. statistical process control) would be more useful. -- (talk) 09:51, 13 December 2013 (UTC)


I would like to point out again, that these 503 drop-outs only started in the last month or so. Before then my Faebot activities worked fine. A recent example from today is where drop-outs are causing uploads where the image page is not being written, see this DoD photo. I only realised this might be happening after someone complained about two blank uploads on my talk page.

If this problem is not addressed, new bot writers will find writing a successful and reliable bot extremely difficult as just following good practice for pywikibot code will need to be supplemented by adding I/O routines that are increasingly defensive about potential failures on the WMF server side. -- (talk) 13:56, 18 December 2013 (UTC)

Watermarks and historical photos

Washington SR-520 bridge under construction - 1962.jpg

File:Washington SR-520 bridge under construction - 1962.jpg was recently tagged with {{Watermark}}. I really wonder whether on a historical image like this where the non-photographic content indicates archival information we would really want to remove that content. (It's certainly not a watermark in the narrow sense.) And when I say "I wonder" I'm not being rhetorical: I don't think the decision on this is obvious. Maybe create a version under a new filename? Similarly (but not the same) on something like File:Seattle - Smith Tower 1914.jpg where we have photographer {{w:Asahel Curtis}}'s mark on the photo. Would we really want to remove this, any more than we'd want to remove the signature on an oil painting? - Jmabel ! talk 20:40, 15 December 2013 (UTC)

I could be wrong in tagging it, but IMHO having the archival marking on the bottom can distract from the image. --Admrboltz (talk) 20:47, 15 December 2013 (UTC)
It's not a watermark in the narrow sense, but it's ugly and distracting as hell. It's very noticeable and distracts from the picture even at thumbnail sizes. I'd much rather have a photo with a neat copyright notice on the bottom then a bunch of numbers that may never be useful to anyone using our copy handscratched into the bottom. (The date's useful, but just as well might be on the metadata then scratched into the photograph.)--Prosfilaes (talk) 07:28, 16 December 2013 (UTC)
I would tend to support Jmabel when it comes to the carefully considering the possibility of keeping information which was on a historic image as originally distributed, and is not extraneously-added or an advertisement. For example, on "French postcard" type images (File:JA52-French-postcard.jpg etc.), the "SERIE" markings were on the postcards as originally printed, and are highly relevant to deducing their origins and copyright status, grouping related images together, etc. Marking of such types should not be removed without some thought... AnonMoos (talk) 14:48, 16 December 2013 (UTC)
But I'm guessing that "Washington SR-520 bridge under construction" was never published before now. These are notations added by some archive, not part of the original picture. Moreover, File:JA52-French-postcard.jpg is a postcard, and I have a hard time imagining how it might be used otherwise. If a project were to use that as an illustration of high heels or something, I would highly encourage them to upload a borderless version without the markings. Since the starting photo is most likely to be used as a picture of the bridge, the image is much more useful without the markings then with them.--Prosfilaes (talk) 02:08, 18 December 2013 (UTC)

Attempt to delete and censor image that is Featured Picture quality on multiple language Wikipedias

Please see the deletion discussion at Commons:Deletion requests/Files in Category:Streisand effect.

This is an attempt to delete and censor an image established as Featured Picture quality on multiple different language Wikipedias, including (1) English, (2) Spanish, and (3) Persian Wikipedia.

I really don't think this is the best way to go about addressing these inherent issues.

Please let's not censor and delete images that are in-use and in-scope as Featured Picture quality images across multiple different language editions of Wikipedia.

Thank you for your time,

-- Cirt (talk) 22:31, 15 December 2013 (UTC)

I prefer to see this more as an acid test of what the WMF resolution means rather than an attempt at censorship. At least we now know that for the classic "it's too late it's already public domain" example of Streisand's home, Commons follows the legal minimum, and is likely to ignore courtesy requests from celebrities to remove media that is firmly in the public domain already, even if it is claimed to be unflattering or offensive. If we intend to improve Commons policies, I suggest a more cautious process of assembling example case studies to continue to test the boundaries of this, hopefully showing some helpful examples of where deletion based on a common understanding of respect for the rights and dignity of a subject is appropriate and realistic, in particular for non-notable common folk as well as the newsworthy, those that can pay for lawyers or the notable.
The level of trolling and lobbying on this topic is painfully over-heated right now, and with many of us busy with Christmas, I don't see this resulting in a meaningful consensus until the beginning of next year. That should be sufficient time to get a case book together to help ensure any changes are made on solid foundations of fact rather than speculation and pure rhetoric. -- (talk) 16:29, 16 December 2013 (UTC)
Acid test or being pointy, I can understand how people could see it either way. I concur with Fae however that the topic is still too emotional to allow a productive discussion. And in light of that russavia's DR does not seem to be the prudent approach to work on this issue. --Dschwen (talk) 17:38, 16 December 2013 (UTC)
I agree, individual DRs are not a very productive approach to clarifying these issues at present. I like Fae's idea of assembling a casebook for future discussion. --Avenue (talk) 03:02, 18 December 2013 (UTC)
I don't agree with the assessment this case actually settled things to the degree offered by Fae. If someone uploaded an unflattering image of a celebrity (or anyone for that matter) I would not be opposed to removing it, particularly if they provide a suitable replacement or something better already exists. The Steisand case was a simple legal consideration of a claim to right of privacy. The court found that the claim was not valid. Her dignity was never in question. 13:19, 18 December 2013 (UTC)
"Firmly in the public domain" I would hope conveys something about the media in question beyond being free. Any photograph subject to significant press interest that is also PD is probably reasonable to include on Commons due to potential for educational value. This is not the same thing as a recent poor quality fan photo of a celebrity looking tired after a long day at a conference, when we have many better quality, and similar, photos to use instead. -- (talk) 13:29, 18 December 2013 (UTC)
Ok, that bit of clarity helped more than the hope. I think we are in violent agreement. Thanks. 14:05, 18 December 2013 (UTC)

Backlinks missing

I think File:Still Life with Rick Santorum, Lube, Dildo, and Justin Bieber doll.jpg should automatically include a link to n:Santorum neologism gains prominence during US election cycle as a usage, but doesn't. I am 99% sure it was listed under global usage ~18 hours ago. Is this a known bug? John Vandenberg (chat) 09:27, 17 December 2013 (UTC)

Looks like enwikinews admins uploaded a local copy due to concerns about it being deleted (Nothing freaks out the sister projects like when a used image goes up for deletion on commons). They then moved it to a different name, because they didn't want the global usage messed with while people were discussing it for deletion [9]. GlobalUsage extension apparently doesn't re-add imagelinks after local image is moved, properly (I think the extension was written prior to file moving being a thing, which is probably why). I null edited the wikinews article, which restored the global image links. Bawolff (talk) 21:55, 17 December 2013 (UTC)
Filed as bugzilla:58604 Bawolff (talk) 22:52, 17 December 2013 (UTC)
Thank you, Bawolff (talk · contribs), for your help in this matter, most appreciated, -- Cirt (talk) 18:09, 18 December 2013 (UTC)

Movable type image should be rotated 180 degrees

File:Metal_movable_type.jpg should be rotated 180 degrees. It was rightly first mirrored left-right to show the type correctly, but should not have also been rotated upside-down. The rotation will return it to its native position as the original viewer would have seen it. This will also not diminish the purpose of the image, which is to show the sorts, sort tray, etc.

User:Mike_Simpson 17:02, 2013-12-17 (UTC-6)

Yeah, I agree. I have added a rotate template. --Dschwen (talk) 23:33, 17 December 2013 (UTC)
I see that the bot has come along and posted a new version, stating that it has been rotated, but the rotation does not seem to have been effected. Sqwsqw (talk) 04:34, 18 December 2013 (UTC)
Try forcing the web page to reload by pressing Ctrl-F5 (this is the command for Mozilla Firefox; if you are using another browser, check the instructions). — SMUconlaw (talk) 05:16, 18 December 2013 (UTC)

December 18

Logo of Rio de Janeiro

Hi. In this picture says source is Flickr. However, very clear says "undated photo released by the Rio 2016 Olympic Bid Committee". I'm not sure if this is electible per PD Text-logo, so I'm asking for a second opinion. Thanks. --Ganímedes (talk) 00:05, 18 December 2013 (UTC)

I'd say it is too complex for {{PD-textlogo}}. — SMUconlaw (talk) 05:18, 18 December 2013 (UTC)
I went ahead and deleted the file. Thanks for bringing this to our attention, @Ganímedes. odder (talk) 10:01, 18 December 2013 (UTC)


Can anyone explain the strange stripes on this image? I ask because I haven't seen anything like it before, and I'm suspicious about this user's uploads because they all (except this one) seem too good to be true, appear on other pages in TinEye and lack EXIF information. UNless I'm being over-cautious, that is. Thanks Rodhullandemu (talk) 01:08, 18 December 2013 (UTC)

According to the metadata:
Color Space Data                : CMYK
MediaWiki has a few bugs with w:CMYK images and therefore doesn't display them correctly. The strange stripes are not there if you use a program which handles CMYK properly. For example, if I view the full-size image directly in my web browser, the strange stripes go away. --Stefan4 (talk) 01:20, 18 December 2013 (UTC)
Thanks. Strangely, today all the images of various sizes look OK, except for the Photoshop in the EXIF data! Rodhullandemu (talk) 17:55, 18 December 2013 (UTC)

Can we extract usable files from this lot?

Bunch of MMO art files have been released under CC-0. Anyone got the software to turn them into files that we can host?

Geni (talk) 18:14, 11 December 2013 (UTC)

Files of the type .fla (Adobe Macromedia Flash source workfiles) are supposedly convertable into (animated) SVGs. I am not sure, howevere, if the W3C has already developed SVG enough to cope with all bells and whistles of Flash (or ever will), and I never heard of a reliable exporting/converting tool. -- Tuválkin 02:13, 19 December 2013 (UTC)

December 12

Flies need renaming

In each case the word "sketch" should be replaced by "diagram". These are not "sketches"; they are very detailed and precise technical drawings.

The word "sketch" is currently misused on Commons and Wikipedia to mean anything that is black and white art, ranging from a very minimal drawing (a bonafide sketch) to a high elaborate and detailed steel engraving. If the image being described took more than five minutes to produce, then it isn't a sketch.

Thanks to those people who have recently adjusted the names on a number of Leonardo da Vinci paintings, at my request.

Amandajm (talk) 01:23, 12 December 2013 (UTC)

  • If you feel this needs doing, there are a number of ways to get it done. The simplest is probably to make the request by putting the {{Rename}} template (with appropriate parameters) on the file page. This isn't something that needs to come for general discussion at the site-wide Village pump. (And, by the way, speaking as an artist: there are pieces that take hours to complete and are still sketches.) - Jmabel ! talk 02:00, 12 December 2013 (UTC)
All three ✓ done. Added camel case, too. -- Tuválkin 01:31, 19 December 2013 (UTC)

Characters allowed in Commons filenames?

I have a huge list of files that are hosted on Commons, and I want to check whether there is any error.

For instance, a filename "]]::::%%%{{.jpg" would probably be invalid on Commons. Filename is the part after "File:".

Where can I find what characters are allowed/forbidden in Commons filenames? Or is any UTF-8 character OK, including newlines, RTL markers, etc?

Thanks! Nicolas1981 (talk) 03:29, 12 December 2013 (UTC)

See Commons:File naming; a proposed Commons guideline. Jee 04:08, 12 December 2013 (UTC)
Use the force, read the source. Start here: --Dschwen (talk) 05:56, 12 December 2013 (UTC)
Actually Title::SecureAndSplit would probably be more enlightening. The tl;dr version (for files. Normal page titles have slightly less restrictive rules) is:
  • Max 240 bytes (As encoded in utf-8, so 1 character takes between 1-4 bytes depending on the script)
  • Valid UTF-8
  • Unicode Bidi characters are stripped
  • Anything that matches the character class [ _\xA0\x{1680}\x{180E}\x{2000}-\x{200A}\x{2028}\x{2029}\x{202F}\x{205F}\x{3000}] is turned into a space
  • Not allowed more than three ~'s in a row
  • Not allowed a % sign if it forms a valid percent escape. (e.g. 100% is fine. %20 is not. Depending on the context this is either disallowed, or converted into what it is escaping)
  • .'s and /'s are not allowed if it looks like they would form a relative path (e.g. ../foo is not allowed. foo/bar is fine)
  • Colons are either not allowed, or automatically turned into a '-'
  • ><][}{ are not allowed
  • Anything not matching this regex is not allowed [ %!\"$&'()*,\\-.\\/0-9:;=?@A-Z\\\\^_`a-z~\\x80-\\xFF+]
  • Things on MediaWiki:Titleblacklist and mediawiki:Filename-prefix-blacklist are not allowed (Some of those entries only apply to certain user groups, such as newbies)

Bawolff (talk) 16:20, 12 December 2013 (UTC)

Isn’t the regex in the last bullet point above blocking most of the Latin script, all other scripts, and a whole bunch of puntuation signs? Yet there are many, many filenames in Commons using those characters… -- Tuválkin 02:21, 19 December 2013 (UTC)


Is there a way to allow a one single quote ( ' ) ? Or rather, is there a reason why it shouldn't be allowed? The filenames in this batch upload seem to have been normalized to exclude it, but numerous French titles look malformed without it (sample: compare filename and "title" in the template). --  Docu  at 14:10, 15 December 2013 (UTC)

There is no problem with single quotes on Commons. It's only a personal decision from the uploader to spell those filenames wrongly. Perhaps the uploader mistakenly believed that the quotes were not allowed or perhaps he did not care about making mistakes when transcribing the titles. -- Asclepias (talk) 14:49, 15 December 2013 (UTC)
Isn't the above regex blocking it ("Anything not matching this regex..")?. I guess you are right. --  Docu  at 15:10, 15 December 2013 (UTC) (edited)
The proper character to use in words is the “letter apostrophe” U+02BC ʼ and in punctuation (incl. English apostrophization) is the “punctuation apostrophe” a.k.a. “right single quotation mark” a.k.a. “single comma quotation mark” U+2019 . In other cases, more seldom, other characters should be used: U+02BB ʻ, U+A723 , U+02BD ʽ, U+02BE ʾ, U+02BF ʿ, U+02C0 ˀ, U+02C1 ˁ, — or others, outside the scope of the Latin script. The “single quote” character U+0027 ' should be reserved for established technical purposes rooted on its own legacy, such as programming and mark-up languages; its use in human-readable typography should be restricted to cameos in works such as The nineteenth Century’s own Y2k bug: How shortsighted typewriter and telegraph Moguls stiffled typographic Æsthetics away from the common Man for a dozen Decades… -- Tuválkin 02:05, 19 December 2013 (UTC)

December 13

Book of Aneirin put online by National Library of Wales

An interesting discussion on the WikimediaUK watercooler [10], about claims of copyright by Cardiff City Council over images of the medieval Book of Aneirin. A sad contrast to yesterday's release of over a million such images, by the British Library. Andy Mabbett (talk) 12:33, 14 December 2013 (UTC)

Well, this is clearly in the public domain, as per our PD-Art policy. Not very surprising, most online archives in France put some illegal restriction. Yann (talk) 14:24, 14 December 2013 (UTC)
Just to be sure that it stays freely accessible to everyone, and really in the public domain, here it is: File:The Book of Aneirin.pdf. That's the same quality as the original files. I didn't include images of 3D objects which probably have a copyright. Finally I may upload all individual images later. Regards, Yann (talk) 20:34, 14 December 2013 (UTC)

The same is true of several works recently digitised in a joint project by the Bodleian Library and the Vatican. Andy Mabbett (talk) 11:58, 19 December 2013 (UTC)

Assembly Seating Chart

How to create this type of chart i.e Assembly Seating Chart? If by any software, then please send me a link to this software. I want to create an assembly seating chart similar to this chart. Would be thankful if someone tells me. --███▓▒░░ Asad Warraich ░░▒▓███ (talk) 14:40, 14 December 2013 (UTC)

Inkscape, or, if you have the guts, any text editor. With extra guts you can generate such SVGs directly from a database: That can be done with a simple tool such as MS Excel, or programmatically. -- Tuválkin 03:08, 19 December 2013 (UTC)

Electron microscope images of insects

Hi All

I'm the Wikipedian in Residence at the Natural History Museum in London. I've been offered a small amount of time for someone to take electron microscope images of entomology specimens in the collection. What would be the most wanted images? Given the size of our collection we will probably have a specimen of most species. If you reply on my talk page in the few days that would be really good. Feel free to request images that have already been suggested, it will help me get an idea of the most wanted ones.


--Mrjohncummings (talk) 14:51, 18 December 2013 (UTC)

  • Hi
Millimeter insects. The pest of cereals, eg Stegobium paniceum. --Archaeodontosaurus (talk) 06:20, 19 December 2013 (UTC)

Newest media in a category

Is there a way to find the newest media in any given category, by ordering after date uploaded, getting notifications, or something like that? Otherwise it is almost impossible to keep updated on when new and better images of a subject become available here.FunkMonk (talk) 15:04, 14 December 2013 (UTC)

You can use Special:Recentchangeslinked for the category you're interested in (there is a link to an Atom feed in the left sidebar); until very recently, it did not include upload logs, but now it does. Perhaps the CatFood tool might come in handy, too, but I've never used it. odder (talk) 15:59, 14 December 2013 (UTC)
There's also the API (change the part at the end of the url to the category you want): (It lists things in order of the date the file was added to a specific category. Most of the time that's equivalent to upload date). However, the output is not exactly the best for human consumption. Bawolff (talk) 20:01, 15 December 2013 (UTC)
Ouch. Wouldn't there be demand for a less convoluted tool? FunkMonk (talk) 18:24, 19 December 2013 (UTC)
VisualFileChange can sort by timestamp. -- Rillke(q?) 21:40, 19 December 2013 (UTC)
Bawolff, there are a couple of things that confuse me about your query string. The API help states that clprop cannot be used with a generator, and the result complains about cllimit as well. Also is there a way to include iiprop data (imageinfo) in the results? Ande does the first prop=info actually do anything? And does the inprop=url do anything?--Dschwen (talk) 19:02, 19 December 2013 (UTC)
To answer myself after some RTFM :-) --Dschwen (talk) 19:10, 19 December 2013 (UTC)
Also, shouldn't you specify indexpageids=true to get a reliable sort order? Or will the properties in the returned object be listed in a well defined order in a for in loop? --Dschwen (talk) 21:03, 19 December 2013 (UTC)
I was more looking at the result directly (which is why I included the url) if you are doing js stuff you would probably want to do indexpageids (I have no idea if for in loops return reliable order). You can have as many props as you want so you could do both categories and imageinfo if you wanted. The limit issue is because with categories each category counts against cllimit, and each image has multiple categories so cllimit gets hits before gcmlimit. Where did you read that clprop can't be used with a generator.
in terms of an actual interface in mediawiki proper - I'm hoping date sorting stuff will be a natural thing to add when the chinese collation stuff lands in core (which will hopefully eventually happen) Bawolff (talk) 21:25, 19 December 2013 (UTC)
Here, but I think I must have edited your query. When I ran it I got some warnings. Now I also see the effect of inprop=url. Sorry about the confusion. --Dschwen (talk) 21:36, 19 December 2013 (UTC)
Hmm, that's odd, you definitely can. Maybe they meant if you were using categories as the generator, then gclprop would have no affect, which is true of all generators. Anyways, I've removed the line from the docs. Bawolff (talk) 23:30, 19 December 2013 (UTC)

December 15


Do we have an ORTS permission for it, or just this is fine? --Rezonansowy (talk) 13:09, 19 December 2013 (UTC)

  • ORTS => OTRS. OTRS would be better - it would be good if Tim Schulz would forward the email accordingly - but the permission is clear enough. This is how we used to do it before the OTRS system was established, although it appears to date from after that time. - Jmabel ! talk 00:44, 20 December 2013 (UTC)

Creative Commons Licence Clarification

Creative Commons have recently clarified their licence wrt image resolution. Users who upload a small 640x480 JPG to Commons under CC have in fact released the underlying "copyright work" under CC. So the licence applies to any other form of that "copyright work", including the 24MP 16-bit TIFF you may have thought was safely "All rights reserved" on your Flickr or other hosting site. Even if your larger version must be purchased, there is nothing stopping someone uploading it here. This is a surprising development. It is unclear as yet how this affects other variants of a copyright work such as cropping an image, a short cut of a video or even a single frame from a movie. Please discuss at Commons:Village pump/Copyright#Update. -- Colin (talk) 19:36, 19 December 2013 (UTC)

December 20

Interesting Commons images / images with interesting stories

I am looking for Commons images that have found use beyond mere images, like discovering new information or pushing the edge of knowledge. I remember for instance some GLAM-related image that was restored by someone who discovered that the image showed dead bodies, something that the original image donor did not know about. Can anyone help me discover that image and suggest others. Thanks in advance. Shyamal (talk) 06:05, 20 December 2013 (UTC)

Good idea. I think you can suggest it as a topic for forthcoming photo challenges. Jee 06:12, 20 December 2013 (UTC)
I think you might be thinking of File:Wounded_Knee_aftermath3.jpg. See some of the info about it on Durova's Wikipedia user page. Bawolff (talk) 06:15, 20 December 2013 (UTC)
Thank you for locating that. Look forward to more such examples. Shyamal (talk) 08:14, 20 December 2013 (UTC)

"Script error" in all image pages

{{Information}} displays a red "script error" in all image pages. It's probably an embedded template, but I don't have the time or expertise to investigate. If someone could look into it, it would be great. --Eusebius (talk) 08:55, 20 December 2013 (UTC)

Yes, very annoying. --AndreasPraefcke (talk) 08:59, 20 December 2013 (UTC)

maybe Bawolff (talk) 09:02, 20 December 2013 (UTC)
My bot is now null editing all pages in Category:Pages_with_script_errors --Zhuyifei1999 (talk) 09:20, 20 December 2013 (UTC)
Looks good, thanks. --Eusebius (talk) 10:16, 20 December 2013 (UTC)

Moratorium on user-generated art about living people

In light of the recent wmf:Resolution:Media about living people, and also the 2009 wmf:Resolution:Images of identifiable people, I propose a moratorium on uploads of user-generated (unpublished) art about living people.

I have spent the majority of a day populating Category:21st-century caricatures and Category:Caricatures of living people, and there are only a few unpublished artistic images in those categories that could be considered valuable images. I have tried to find them all, but I am sure others will be found.

I think a six month moratorium would be sufficient. During that six months, any deleted image must be documented on a specific page here on Commons, and after six months those images are undeleted so that the community can review the images and create a policy which permits only user-generated art about living people that is appropriate. John Vandenberg (chat) 19:21, 16 December 2013 (UTC)

I doubt that the resolution requires this. Ruslik (talk) 19:29, 16 December 2013 (UTC)
  • Symbol oppose vote.svg Oppose, it's interesting that this is based on a user's research of caricatures, which are essentially forms of political cartoons, a form of expression upheld in a unanimous decision in Hustler Magazine v. Falwell. -- Cirt (talk) 19:31, 16 December 2013 (UTC)
  • I can't see how that resolution requires this for all artwork. And what is "user generated" - does some nobody's Flickr photostream count as publication? And when does something become "art"? Too many problems and I suspect a point is being made here rather than a suggestion that has any reasonable chance of succeeding. -- Colin (talk) 20:06, 16 December 2013 (UTC)
    • Colin, do you think it's appropriate for a user to trawl through Flickr etc., look for every hostile cartoon or piece of art designed to vilify someone they hate, and upload the lot to Commons, even if the "art" was done by some completely non-notable person no one has ever heard of? Because it seems to me that's how we got Category:Santorum_neologism-related_images and Category:Caricatures_of_Rick_Santorum. Andreas JN466 20:16, 16 December 2013 (UTC)
      • No, I don't. That's why I voted to delete the non-notable cartoons at that very deletion discussion. This suggested moratorium isn't just about non-notable "user generated" hostile cartoons or hate artwork. It is too blunt and hasn't been thought through. And frankly I'm surprised at all the fuss about a WMF resolution that states the bleedin' obvious. Does anyone other than a complete troll think we should actually be a project where human dignity is ignored? -- Colin (talk) 20:31, 16 December 2013 (UTC)
        • Does anyone other than a complete troll think we should actually be a project where human dignity is ignored? Have you noticed that your views are frequently a minority view in these sorts of discussions? Andreas JN466 22:07, 16 December 2013 (UTC)
          • I don't understand the point you are making. Do you think I am wrong and should give up, or right but unlikely to succeed? Anyway, I disagree. There is a vocal minority on Commons who hold extreme world-views. They are no more representative of actual Commons users than Jimbo's talk page on Wikipedia is representative of actual Wikipedians, say. What is your view on human dignity? History suggests we ignore it at our peril. Colin (talk) 12:13, 17 December 2013 (UTC)
            • The views you ascribe to "complete trolls" seem to be mainstream in Commons decision-making. So I don't think blaming it on complete trolls is accurate. Andreas JN466 12:24, 18 December 2013 (UTC)
Yes, I agree with Colin that it seems a bit like a w:WP:POINT is being made here, it seems like that is not uncommon in all this drama related to these discussions these days, unfortunately. -- Cirt (talk) 20:08, 16 December 2013 (UTC)
  • I don't see a need for such a moratorium either. I am hopeful that pretty much every contributor has enough common sense and good judgement to distinguish between a picture painted with a dick (and the corresponding making-of) and aerial photography for coastline documentation purposes, and that only a small vocal minority sees the need to actually manufacture a controversy around this subject. So over here in the real world ;-) I would consider this using a sledgehammer to crack a nut. --Dschwen (talk) 20:13, 16 December 2013 (UTC)
"using a sledgehammer to crack a nut", most interesting analogy, Dschwen, I'd have to agree with that part. Cheers, -- Cirt (talk) 20:15, 16 December 2013 (UTC)
  • Symbol oppose vote.svg Unnecessary as the minuscule number of real problematic images or handful of requests for courtesy deletions we get each month (has anyone got any statistics?) can be handled on a case by case basis. In fact, the real problem here is that more effort is being wasted on knee-jerk drama-mongering and grandstanding rather than constructive proposals as to changes to guidelines or policy that is based on case studies rather than fantasies. If anything this is a good reason to look for fresh cases rather than closing the doors or pulling up the ladder to our ivory tower in order to put off new contributors. As for crappy user-generated art, just go ahead as raise them up for deletion as out of scope, we don't need a change of policy to remove non-educational offensive chaff. -- (talk) 20:50, 16 December 2013 (UTC)
  • Symbol oppose vote.svg Oppose Knee-jerk and Overreaching. Saffron Blaze (talk) 22:11, 16 December 2013 (UTC)

Propose a new policy that when Fae and I agree on something the issue should be speedily closed in our favour. Saffron Blaze (talk) 22:31, 16 December 2013 (UTC)

  • Symbol oppose vote.svg Oppose per Dschwen and . If people have a difficulty to evaluate each case using their common sense, it is their problem. We can't close the doors entirely to satisfy them. Jee 02:58, 17 December 2013 (UTC)
  • I'm not sure I even understand the proposal. Are you saying that photographs are "not art" so they are OK, but anything non-photographic is objectionable? - Jmabel ! talk 07:56, 17 December 2013 (UTC)
I don't think he said photograph is not art. Here the concern is on works like these or these; I think. Jee 08:02, 17 December 2013 (UTC)
Jkadavoor, actually Jmabel's question is a good one, and it appears you haven't understood the concern at all. The links you provide are irrelevant, as they do not contain any non-notable user-generated art of living people. In fact, I can only find one that is of a living person, being File:Adi Holzer Werksverzeichnis 899 Satchmo (Louis Armstrong).jpg. John Vandenberg (chat) 11:14, 17 December 2013 (UTC)
Yes; yes. I just mentioned it to give him an idea what type of works (in looks) you mean. We rarely (almost never) feature such works from non notable artists. Jee 11:54, 17 December 2013 (UTC)
Jmabel, I am referring to user-generated non-notable 'art' of identifiable living people, where the art is not presenting an accurate depiction. Caricatures and digitally altered media depicting living people would be part of the proposed moratorium. Accurate photographs of living people would not be, as that is a different problem domain and much more complex. John Vandenberg (chat) 11:14, 17 December 2013 (UTC)
This is a self portrait and acceptable, I think. "where the art is not presenting an accurate depiction." I don't know how it can be defined. It is difficult to depict a person "accurately" in art. I think you mean "not defaming"; that is why I said we can deal it only case to case. I don't know whether you see the recent DR. Many claim it is accurate depiction; many on the other side. Jee 11:54, 17 December 2013 (UTC)
We do not need a special rule or moratorium to remove images that appear intended only to defame or harass, COM:IDENT handles that perfectly well right now. Anyone, including anonymous IP accounts and those complaining in this thread, can raise images they have concerns about up for deletion, speedy deletion or complain about them by email to volunteers or to the WMF legal department. I suggest they get their fingers out and click on the Nominate for deletion link in the toolbar. Of course, reasonably illustrative political parody, widely published images, established internet memes etc. can be kept as validly in-scope. -- (talk) 12:47, 17 December 2013 (UTC)
Technically, COM:IDENT only concerns "Photographs of identifiable people" and given how fussy people here are about limiting any discussion to pre-agreed policy terms (rather than, you know, using their own brains) that is a problem. We could perhaps do with some policy concerning art and user-generated art and art that concerns a living person. I also think the "identifiable people" aspect of COM:IDENT sometimes problematic -- a revenge-porn image (say) doesn't become ethical just because some admin pixellates the girl's head. So I don't think our policies and guidelines are perfectly ok and welcome the WMF once again encouraging us to try harder. The solution to these issues is probably easier to discuss away from flashpoint images, but inevitably discussion gets drawn back to them. They consume so much of our time for so little value to the project. -- Colin (talk) 13:12, 17 December 2013 (UTC)
Yes; COM:PEOPLE is the basic policy for us to evaluate any media related to people. I don't think it is applicable only to photographs. Many pages here are badly named (like COM:FP, COM:QI, COM:VI, ETC.); most of them accept any media contents. Some people think "identifiable" means "facially identifiable"; so cropping out or blurring the face may enough. No; people can be identifiable from many metadata and many other personal identification information. John, I will support you, if you take an initiative to renew COM:PEOPLE, COM:Courtesy deletions or any other relevant policies to meet the Resolution:Media about living people. Jee 16:36, 17 December 2013 (UTC)
So what exactly is the difference between a drawing someone finds uncomplimentary and a photo someone finds uncomplimentary? While I definitely try to defer to reasonable requests by people who aren't particularly famous (and don't threaten to sue me, at which point it becomes a matter of principle for me) and who dislike a particular photo of them, I wouldn't want to give everyone, including famous people, a veto over their photos. And I don't see why the medium really matters. Certainly deliberately scurrilous non-notable images should not be acceptable, regardless of medium. - Jmabel ! talk 17:18, 17 December 2013 (UTC)
I too don't think a medium make much difference. Nowadays photos can be easily manipulated according to their needs and many caricatures have (in the above mentioned categories too) photo elements in them. Jee 17:32, 17 December 2013 (UTC)
There's usually a difference in subject matter of the ones we're talking about. We'd delete a indecent photo that wasn't consented to in a heartbeat, but we have a Lautoff drawing of a certain politician masturbating to death. An uncomplimentary drawing is often truly insulting whereas an uncomplimentary photo is often a fan not catching the person's best side. And photos unless specially manipulated are real; a drawing can make someone do whatever the artist wants, whereas unless faked a photo is limited to what really happened. Also, there's a sense of distance with photos; all the artist renditions of Santorum showed off the artist's political position, but the photograph, as a document of fact, could have been taken by a neutral journalist or an angry supporter of Santorum.--Prosfilaes (talk) 20:18, 17 December 2013 (UTC)
  • Symbol support vote.svg Support Actually, I think that John's proposal is a good one. To me, there is a lot of junk in the categories he linked to, and cleaning and straightening our policy would be a good step toward improving the quality of our content. Six months seems a bit long, may be three months would be enough. Yann (talk) 12:16, 17 December 2013 (UTC)
  • Symbol oppose vote.svg Oppose We don't need a moratorium to approve a new policy. I agree that we should be concerned by the problem of caricatures of living people, but we can deal with it using present policies until they are improved. Furthermore, derogative unpublished caricatures of identifiable living people by non notable artists and lacking educative value can be deleted by COM:SCOPE, and usually they are.--Pere prlpz (talk) 19:05, 17 December 2013 (UTC)
  • Symbol support vote.svg Support per Yann. A methodical approach makes sense. --Andreas JN466 12:19, 18 December 2013 (UTC)
  • Symbol oppose vote.svg Oppose – We shouldn't disenfranchise amateur artists or hobbyists. We shouldn't rob our amateur artists (if I understand what Vandenberg meant by "user-generated (unpublished)" correctly) of the ability to share caricatures. Many artists here simply want to share their artwork with the world without payment or restrictions. We shouldn't punish artists who choose to share their work online under a free license rather than in a physical art gallery or in a book that must be purchased. --Michaeldsuarez (talk) 23:19, 18 December 2013 (UTC)

As it seems there is little interest in a moratorium, I have proposed a policy change at Commons talk:Photographs of identifiable people#Caricatures and cartoons of identified or identifiable people. --John Vandenberg (chat) 02:10, 21 December 2013 (UTC)

Pictogram voting comment.svg Comment - To those that commented, above, please see this proposal at Commons talk:Photographs of identifiable people#Caricatures and cartoons of identified or identifiable people, it strikes to me of w:FORUMSHOPPING. Cheers, -- Cirt (talk) 02:14, 21 December 2013 (UTC)
  • Symbol oppose vote.svg Oppose We are custodians of a library of free content. We should never change our policies to delete previous content (or forbid types of content) without exceeding care, debate, and reason. A knee-jerk response against things that, on the whole, are probably minimally problematic, and, at worse, hamper our ability to cover encyclopedic content, is not the way to go. Adam Cuerden (talk) 04:46, 21 December 2013 (UTC)

December 17

Problem with Wikimedia Commons main page

The main page has a different picture of the day in Google Chrome than Internet Explorer. Blackbombchu (talk) 18:33, 20 December 2013 (UTC)

I guess you are logged-in with one browser and with the other one you aren't. That's because anonymous users get pages from the caching servers while logged-in users get it from the parser cache. Someone who has a link to a help page at hand? -- Rillke(q?) 19:17, 20 December 2013 (UTC)
Rillke, Is there a way to purge the page at 0:00 UTC every day? —Mono 03:10, 21 December 2013 (UTC)
Could be done with a simple wget/curl command using cron on the toolserver/tool-labs. --Dschwen (talk) 04:36, 21 December 2013 (UTC)
On my bot's crontab for now. --Zhuyifei1999 (talk) 04:48, 21 December 2013 (UTC)

Targaryen had left a note on the image's talk page a while back saying:

"It may be permissible as a free image consisting of simple shapes and text."

What do people here think?

Could this image be uploaded locally here to Wikimedia Commons, as a free-use image, as consisting of simple shapes and text?


Thank you for your time,

-- Cirt (talk) 20:37, 20 December 2013 (UTC)

The text clearly isn't copyrightable, and that curve is simple enough, so, yeah, I think it can be uploaded here.--Prosfilaes (talk) 20:52, 20 December 2013 (UTC)
Thank you, I've gone ahead and uploaded it here. Cheers, -- Cirt (talk) 20:59, 20 December 2013 (UTC)
Checkmark This section is resolved and can be archived. If you disagree, replace this template with your comment. Jmabel ! talk 00:54, 21 December 2013 (UTC)

Saving URAA affected media from Meyers Blitz Lexikon

Just a heads up that due to the URAA review, after about seven days many images scanned from the 1932 Meyers Blitz Lexikon will be deleted until 2027.

I have not been following attempts at saving these files on sites not affected by the URAA (maybe Canada, maybe Europe) but I did leave a note here in case others are wondering. This would appear to affect wikisource and wikipedia of other languages but a quick search finds this wikisource discussion (in German). I will post a heads-up and linkback on the German language version of the Commons Village Pump. -84user (talk) 15:54, 21 December 2013 (UTC)

December 22

Template:Closed cap

Can someone convert this to Autotranslate template, like for example in {{Delete}}? Thanks in advance. --Rezonansowy (talk) 21:05, 13 December 2013 (UTC)

Do we still need that template, now that we have a working TMH? -- Rillke(q?) 22:21, 13 December 2013 (UTC)
Not sure, is there any alternative to inform users about subtitles? --Rezonansowy (talk) 22:41, 14 December 2013 (UTC)
select CC
They can be directly selected through the video. The template needs rephrasing to match the current state, at least. --McZusatz (talk) 23:13, 14 December 2013 (UTC)
Yes! Anyone can help? Thanks in advance. --Rezonansowy (talk) 14:42, 15 December 2013 (UTC)

@Rillke:, @McZusatz: - could you help? --Rezonansowy (talk) 22:49, 16 December 2013 (UTC)

I will have a look. Hit me back on friday if I forgot it. --McZusatz (talk) 23:29, 16 December 2013 (UTC)
✓ Done a bit. Is the /layout ok? (I tried to keep the old one). If so, we can go on to create the translations. --McZusatz (talk) 01:16, 20 December 2013 (UTC)
Looks fine to me. --Rezonansowy (talk) 22:43, 21 December 2013 (UTC)

December 14

JPL images

I see a somewhat larger problem with the images from Caltech's en:JPL: JPL images are copyrighted.

JPL releases their images for usage by anyone, for any purpose, provided a copyright notice is present and "no endorsement is claimed or implied" (see Template:JPL Image Copyright). Very reasonable conditions, IMHO.

However there are as far as I can see currently (at least) three problems:

  • When uploading an image, one can choose NASA, but not NASA/JPL-Caltech for the copyright.
  • Most JPL images uploaded to Commons are mis-attributed as en:NASA Public Domain.
  • JPL images are used on the various Wikipedia projects, yet a proper copyright notice is very seldom present

The first problem is probably the easiest to solve I would guess, for anybody knowledgeable in the ways of Commons. However I am not knowledgeable in that area.

The second problem may or may not be easily solvable by a bot, which would be tasked with searching for JPL images (e.g. by URLs present in the description). Obviously if the description does not contain any hint to JPL (or other such terms), such images will be difficult to locate. Again, I am not knowledgeable in the area.

The third problem may be solvable on the various Wikipedia projects with bots, once the images are correctly tagged on Commons – I imagine that will be the most difficult problem of all.

Personally, I can not do much more than write down what I see as the problems, and I will have to ask someone here to take responsibility to look into what should (or should not) be done with regards to JPL images, choose the necessary steps, post the necessary notices in the right places, initiate the appropriate actions, and so on. I fear I am already out of my depth…

I know this is going to be quite some work and some people will be unhappy with the results, but I fear it might be necessary. Tony Mach (talk) 00:06, 14 December 2013 (UTC)

One more thing: The Template:PD-USGov-NASA should probably contain a bullet point about Template:JPL Image Copyright as well. Tony Mach (talk) 01:00, 14 December 2013 (UTC)

As far as I know, JPL and the rest of NASA can declare copyright all they want (there are often warnings against non-commercial use even on non-JPL NASA sites) but this is contrary to the fact that NASA and its subsidiaries are either federal government entities or operating under government contract, which means their works are public domain unless specifically mandated otherwise (such as the NASA logo having controlled use by law). This includes work performed by Caltech under contract to NASA or a specific NASA program/mission. Huntster (t @ c) 01:34, 14 December 2013 (UTC)
I agree that for the most part public domain images of this type are public domain without any legally binding requirement for attribution, though where attribution is requested it would be good practice to include it so that Commons benefits from the best possible context. Note, the same cannot be said for the UK Open Government Licence ({{OGL}}) where attribution is a legal requirement even though this is equivalent to public domain.
However, having researched a few images that were problematic from NASA, the uploader has a duty to take care to check the specific licensing for images from JPL/NASA's or any U.S. government agent websites. I have come across images that were published by a consortium of organizations where copyright appeared to be owned by NASA in conjunction with other bodies, such as university research teams and the images were effectively under a non-commercial reuse only licence. In those cases there can be no automatic assumption that a NASA public domain licence takes precedence. -- (talk) 05:54, 14 December 2013 (UTC)
Certainly a valid observation Fae, the uploader always has the responsibility of verifying. If you remember any specific situations like you mentioned in your last paragraph, I'd be interested in reviewing them. Huntster (t @ c) 13:36, 14 December 2013 (UTC)
No, the creations of organisations "operating under government contract" are not automatically in the public domain. The government might be assigned the copyright, or it might be retained by the organisation whose employees created it. (See w:Copyright_status_of_work_by_the_U.S._government#Exceptions.) I don't know which would apply to JPL. --Avenue (talk) 08:41, 17 December 2013 (UTC)
Well. All is "as far as I know" (which might be wrong), and for what it's worth, but: Despite the name (which usually includes NASA), JPL is not part of NASA, it is part of Caltech, which is a private organization. JPL includes the usage of NASA for historic reasons as far as a I know, as both sides see this as a beneficial for their public image. Tony Mach (talk) 09:01, 17 December 2013 (UTC)
One more thing: "JPL is managed for NASA by Caltech under a contractual arrangement begun in 1958 and renewed every five years. Thus, JPLers are Caltech employees." Thanks User:Avenue for the pointer. Tony Mach (talk) 09:51, 17 December 2013 (UTC)
Yes, I should have made that clear here too. (See related DR.) What I'm unsure about is whether Caltech retains its JPL copyrights, or has contractually assigned them to NASA. In neither case would the images be PD, though. --Avenue (talk) 10:21, 17 December 2013 (UTC)
Nowhere do Caltech, JPL or NASA claim that JPL images are to be treated any differently than images from any other NASA center. Caltech specifically disavows copyright ownership of JPL images ("Caltech makes no representations or warranties with respect to ownership of copyrights in the images.") NASA states that NASA still images; audio files; video; and computer files used in the rendition of 3-dimensional models, such as texture maps and polygon data in any format, generally are not copyrighted. and Photographs are not protected by copyright unless noted. Plenty of JPL images appear in non-JPL NASA web sites without any indication that they are copyrighted, indicating plainly that they are not copyrighted. There is obviously no blanket JPL copyright, and no particular reason to worry about copyright issues for an image from their web sites unless there is a statement that it is copyrighted, it is clearly a non-NASA image, or involves a logo. WolfmanSF (talk) 23:45, 17 December 2013 (UTC)
Under the FAR general data rights clause (FAR 52.227-14), the government has unlimited rights in all data first produced in performance of or delivered under a contract, unless the contractor asserts a claim to copyright or the contract provides otherwise. So since Caltech explicitly denies claiming copyright and since probably the contract (that I can't find right now, but i'm rather sure) doesn't specify this either, it's just fine. TheDJ (talk) 00:03, 18 December 2013 (UTC)
I'm sure it's fine for us to host JPL images (giving appropriate credit, as specified in their image use policy), but I'm also pretty sure that these images are not generally in the US public domain. They are not produced by federal employees, and contractual assignment of copyright (even to the federal government) does not make the copyright disappear.
WolfmanSF, the clause in their policy that says they make "no representations or warranties" about image copyright is essentially saying "don't blame us if you get sued". That is very different from saying that they disavow copyright ownership in all images on the website. --Avenue (talk) 02:54, 18 December 2013 (UTC)
If you believe that "these images are not generally in the US public domain" then you must believe that NASA is misleading us by indicating otherwise. Please give me some insight into your thinking here. Do feel that NASA itself is ignorant of the fact that JPL images are not public domain, that they are just careless in setting up their web sites, or that they are misleading us willfully? If I obtained a statement from someone in NASA that they are indeed public domain images, would you believe them? WolfmanSF (talk) 17:09, 19 December 2013 (UTC)
I think you are reading much more into the NASA and JPL websites than is justified, and that while you are misled, they are not wilfully misleading you. Neither NASA nor JPL have indicated that JPL images are generally in the public domain, as far as I can see. While I agree that some JPL images appear on other NASA websites (e.g. GRIN) without a note that they are protected by copyright, I believe this merely indicates that some of NASA's websites do not maintain copyright information as carefully as we try to. But if you got a statement that the images were PD from someone at NASA or JPL who was suitably qualified or authorised to give an opinion on their copyright status, I'd accept that. --Avenue (talk) 14:19, 21 December 2013 (UTC)
I will seek clarification of this issue from NASA once my holiday travels are over. WolfmanSF (talk) 15:34, 22 December 2013 (UTC)

December 5 image loss?

There are quite a few missing images which were mostly uploaded on Dec 5. Are there any known server issues for this day? --McZusatz (talk) 20:24, 18 December 2013 (UTC)

Hmm, on wikitech:Server_Admin_Log#December_5

  • 20:09 ori-l: increase in load coincided with sync of wmf6 to apaches and subsided on roll-back, but wmf6 was not enabled anywhere at the time of syncing
  • 20:08 ori-l: image scalers started overloading at 18:56, cause appears to be a spike of convert jobs exceeding limits & getting killed; swift-backend.log on fluoine has lots of InvalidResponseException; syslog on swift has "swift-object-se: page allocation failure".

Which might be related. Bawolff (talk) 21:15, 18 December 2013 (UTC)

I ran your list through some eval.php code I hacked up to try to recover the files from our data store in Tampa (the primary is in Ashburn). Looks like it found copies for all the files to restore. The gallery works now. Aaron Schulz (talk) 05:48, 19 December 2013 (UTC)
Thanks Aaron! I was just about to put this into Bugzilla. :) --AKlapper (WMF) (talk) 14:36, 19 December 2013 (UTC)
Aaron, Can I send you a list of missing files regularly so that you can run them through eval.php? --McZusatz (talk) 13:06, 22 December 2013 (UTC)

Translated SVG files

Since gerrit:25838 MediaWiki supports translatable SVG images. Since gerrit:95746, this is clearly shown in the UI: see this feature in action.

Are we going to change our habits and finally use single SVGs instead of multiple language versions? Would this practice be recommended? --Ricordisamoa 22:55, 18 December 2013 (UTC)

There's a category of such files on commons at Category:Translation_possible_-_SVG_(switch). There's also some (technical) discussion at Template_talk:Translate#Update_to_mention_support_of_.3Cswitch.3E_statements. Just as a heads up, the current drop down thing has a bug where the (default language) option is broken (My fault. User:Rillke found the issue) In the long term there is mw:Extension:TranslateSVG (by User:Jarry1250) which integrates the Translation extension with SVGs, but that's probably a ways off. Bawolff (talk) 01:44, 19 December 2013 (UTC)
So, once the SVGs are migrated to the 'new' system, can/should we nominate the old versions for deletion? Do you think a bot would be reliable enough for 'merging' translated versions? --Ricordisamoa 15:52, 19 December 2013 (UTC)
I won't start "migrating" any images.
  • If you create/upload a new image, using <switch> elements is probably the way to go.
  • If you improve an old image of which many translations exist, creating a multilingual version is probably advisable, too.
    Deletion of duplicate translations should probably be OK in most cases. Care has to be taken not to delete originals of which the multilingual file could be a derivative of.
  • If you don't change anything, there's absolutely no point in merging/migrating images in my opinion. It produces considerable amounts of work for no real gain (right now, I even think the new system might be too complex for beginners (article authors as well as graphists) and might actually make files less usable for those people; especially with the rather bad software support we have today). Any merging/migration techniques imaginable will probably also be prone to errors and might potentially offend authors of translated image versions in case of deletions of duplicates.
Either way, we're just at the beginning right know. Many possibilities and issues of the new translation system are yet to be found and it will probably take some time for people to actually adapt to it (and even longer for graphics software to support it, if it will at all). --Patrick87 (talk) 19:26, 19 December 2013 (UTC)
I second that. The issue with it is that some - if not even most - SVGs are not properly coded, including text-alignment and text-overflow. I doubt we can code a bot that is smart enough to resolve all these issues - and spending hours of work time with that - why should we? -- Rillke(q?) 10:41, 20 December 2013 (UTC)
I made some switch-localized SVGs a while back, and I'm happy to see that technology now works. I changed a file on de.WP; I hope the "lang=de" isn't needed, but I included it to be sure.
Switch-localized files have some appeal because everything is contained in one file. Having one copy means editing the graphics shows up in all the language versions. Those benefits do not mean it is a good approach.
In the short run, the switch localization is easily lost if someone edits the file with a conventional editor.
In the long run, I don't think switch localization is viable. It copies a lot of redundant information into source files, and that means maintaining multiple copies of the same information. If somebody provides a new localization of "foot", then 50 SVGs may need updating.
I'm not happy with the lang-to-lang translations of XLIFF, but that appears to be better approach. WP could have a master SVG. A separate specification describes how to localize that master to different languages. (An equivalence class could be used.) WP needs a mechanism to enable the localization.
For the most part, many illustrations just need word-to-word translations from a common database (e.g. place names, technical terms). New SVG files that just use word descriptors might localize without any additional effort. Some translations would have to be fixed to get the right meaning (think of id="wikt gloss").
The approach does a reasonable job of separating the work of localization from the graphics editing. It also allows existing graphics editors to be used to make the original image.
Glrx (talk) 19:50, 21 December 2013 (UTC)

Merge with care

Can you not merge-n-delete existing SVG sets? Some of them (like mine) have some visual elements tweaked to accommodate translations (stuff that goes from like making room for long french phrases to switching out flag icons). —Love, Kelvinsong talk 01:19, 22 December 2013 (UTC)

December 19


I would like to move file Not-star-shaped.svg from en:Wp to Commons, preferable with all its previous versions. It is flagged since two years for being moved. Trying different tools had been unsuccessful. Is it only possible to upload the last version, without older ones? sarang사랑 09:49, 19 December 2013 (UTC)

Here's a stupid idea, you could ignore versions 2+3 arriving at 1+4, or ignore 1+2 arriving at 3+4, because the first three versions are from the same contributor. Then you could copy this old version to commons, and replace it by the newer "simplified" version from another contributor. –Be..anyone (talk) 08:41, 22 December 2013 (UTC)
Thank you. sarang사랑 15:02, 22 December 2013 (UTC)

Update to Obama-Mandela DMCA takedown

I have an update to the DMCA situation we discussed here (now archived)

On December 11, 2013, we removed an image of then-Senator Barack Obama and Nelson Mandela. We removed that image after the purported photographer – David Lee Katz – submitted a formal takedown notice pursuant to WMF’s DMCA policy.

When we removed the image, we noted that there remained concerns about the copyright status of this work, which was originally posted as a public domain work of an employee of the United States Congress. We reported that we were in consultation with outside counsel and promised to engage with the community as we obtained additional information that bears upon this issue. We have looked into this issue more and have the following to report below:

In his takedown request, Mr. Katz states that he took the photograph in question on May 17, 2005. Mr. Katz was apparently employed by the United States Senate as a Special Assistant to then-Senator Barack Obama from January 18, 2005 through July 31, 2006.[1]

The copyright status of Mr. Katz’s image turns upon whether he took the photograph as part of his official duties as an employee of the United States Congress. A work is a work of the United States Government (and therefore not subject to copyright protection) when it is “a work prepared by an officer or employee of the United States Government as part of that person’s official duties”, and there is authority for the proposition that a Senator’s employee qualifies as an employee of the United States Government. White v. U.S., 419 Fed. Appx. 439 (5th Cir. 2011) (unpublished) (noting that a Senator’s constituent services representative qualifies as a federal employee).[2] We have found some sources that suggest that Mr. Katz was acting as an aide and a driver rather than an official photographer at the time the photo was taken.[3] Additionally, Mr. Katz’s counsel has contacted us, asserting that photography was not part of Mr. Katz's official duties as a Senate employee, that Mr. Katz used his own camera to take the photograph, and that Mr. Katz processed the photo in his personal time. We were also forwarded a message from Mr. Peter Rouse, who served as chief of staff for then-Senator Obama, attesting that Mr. Katz’s official duties as Special Assistant to the Senator did not include photography.

Because we have not found sufficient evidence that Mr. Katz’s official duties included photography, we cannot definitively conclude that the work is a public domain image and therefore the image was removed pursuant to the DMCA.

Rkwon (WMF) (talk) 23:15, 20 December 2013 (UTC) & DRenaud (WMF) (talk) 01:13, 21 December 2013 (UTC)

Thanks for letting us know (with such detailed links!), and for all your work on this. --Avenue (talk) 22:22, 21 December 2013 (UTC)

December 21

Yellow tint in some browsers for generated previews from source TIFFs

I uploaded a bunch of TIFFs and found that the Commons-generated preview jpeg seems to have weird color profile information. In some browsers (FF, Chrome) there is a yellow tint in the image. In other browsers (IE10, Safari 7) the previews are correct. Screenshot for comparison. See images in this category for examples. Is this a known bug? Should I delete them and convert to PNG before uploading again? Kungbib (talk) 12:25, 21 December 2013 (UTC)

(You may set 'gfx.color_management.mode' to '0' in firefox's 'about:config'.) --McZusatz (talk) 11:25, 21 December 2013 (UTC)
You could remove the bad PowerPhase 6x6 with AR1, daylight colour profile with a command like exiftool Nicodemus_Tessin_dy_dagbok_1688-000.tif -ICC_Profile=, and then re-upload the tiff file. Bawolff (talk) 07:55, 22 December 2013 (UTC)
✓ Done Doing… --McZusatz (talk) 15:55, 22 December 2013 (UTC)

Looking for maintainers for User:CategorizationBot

Hi everyone, User:CategorizationBot has been around for quite some time marking uncategorized files and trying to categorize them. I haven't really paid a lot of attention to it, so some of the functionality broke and the backlog of some jobs increased. You can clearly see this in the categorization statistics. On top of the that the Toolserver is shutting down so if the bot doesn't get moved, it will just stop working. I'm looking for one or more maintainers for this bot. Moving it to toollabs will be a bit of work, but after that the bot is pretty low maintenance. Positive qualifications for a maintainer: Already have a toollabs account, have some Python/Pywikibot knowledge, I already know you. Who wants to help out here? Please reply here or leave a message at User talk:CategorizationBot. Multichill (talk) 11:48, 22 December 2013 (UTC)

@Multichill: Do you have a source where I can clone from? I am planning to have some sub-bots hosted on labs in the future (such as User:FlickreviewR (clone))--Zhuyifei1999 (talk) 12:22, 22 December 2013 (UTC)
Of course all code is in version control, but the whole point is to have *shared* accounts. Why are you forking this? What's the point of having "sub-bots"? In my opinion structural bot tasks shouldn't be running under personal accounts. Multichill (talk) 12:49, 22 December 2013 (UTC)
1. ✓ Cloned, I'll look into them. 2. For sub-bots I mean the bot account(s) which are not my main bot account (User:YiFeiBot). 3. What do you mean by shared accounts? --Zhuyifei1999 (talk) 13:22, 22 December 2013 (UTC)
I saw user:Wdwdbot by User:Wdwd does such edits with a low edit rate. --McZusatz (talk) 13:18, 22 December 2013 (UTC)

Problems with removing a category using Cat-a-lot

Hi, I've had this problem a couple of times recently:

  • Create a new category, use catalot tool to add same to a list of files.
  • Attempt to remove a (now-superfluous) parent of the new cat from the same list (ie, files in the new subcat)
  • Tool instead removes all files in the list from the new cat! Very frustrating.

See, forex revision history of such a file. This happened twice in one day, as I was categorizing a huge file-dump from Flickr that was poorly-categorized. I was careful the second time to RTFM carefully before attempting the removal. Help? TIA, Pete Tillman (talk) 18:48, 22 December 2013 (UTC)

Also posted at Cat-a-lot Talk page.

Mediawiki creating ogg from old version of flac file

Dear Commons users, I need help fixing an issue with Mediawiki's flac to ogg transcoder. I've uploaded a new version of File:Lysenko-The-Sailboat.flac. After that the page said that the file was re-converted into ogg. However, the resulting ogg file is from the previous version of the flac file. I've also tried resetting the transcode as well as clearing browser's cache. Please compare the flac and ogg variants of the most recent version of the file to confirm the mismatch. The difference between the two versions of the flac file is clearly audible in the beginning of the file. Any ideas how to fix it? Thanks, YurB (talk) 00:05, 20 December 2013 (UTC)

Ok, I did some investigating. It looks like TimedMediaHandler isn't properly telling the varnish cache layer to delete old versions of transcodes. I've filed this as bugzilla:58733. This is a serious issue, so will hopefully be fixed soon. In the mean time there is not much one could do short of uploading under a new name (Things in varnish cache get automatically thrown away after 30 days, so the problem should under no circumstances persist longer than that, but that's a long time to wait). Bawolff (talk) 07:27, 20 December 2013 (UTC)
Thank you for your help. Perhaps I'll temporarily upload an ogg separately as a workaround. --YurB (talk) 12:04, 20 December 2013 (UTC)
Moving the file to a new name should fix the issue. It could be moved back once the issue in bugzilla is fixed. Bawolff (talk) 06:05, 21 December 2013 (UTC)
Thank you for the fix. YurB (talk) 19:07, 23 December 2013 (UTC)

Category "tree" loops

It is hardly a surprise that our category tree is not a well defined tree at all, but contains circular references or loops. This is questionable from a semantic point of view as it destroys the meaning of a "subtopic" if - following a sufficient amount of subcategory links - you end up where you started. The other problem this poses id for a programmer trying to traverse a subtree of our category system. To avoid having the code loop back on itself a programmer has to either flag visited categories or sore a breadcrumb trail of its traversal. This is obviously inconvenient. Anyhow, since I'm working on a fast category intersection algorithm I encountered this strangeness. The first tool I wrote to work on my new processed category tree dataset is a loop detector. It finds one loop for every category that loops back on itself (once it finds A-B-A, it won't return B-A-B or A-C-A). This gives a lower bound on the number of loops. Currently I'm only working with page IDs so I cannot yet give you a human readable list of loops (tomorrow maybe), for a few examples look at:

I'll be running more tests on my dataset (currently about 1000 loops are detected). I have a few other interesting statistics in mind regarding average subtree depth etc. --Dschwen (talk) 05:11, 21 December 2013 (UTC)

It's one of the features of Commons categories to end in loops. --  Docu  at 09:00, 21 December 2013 (UTC)
Are you seriously calling this a feature :-)? I would agree as far as the existence of loops is possible in this mesh like graph, but I don't quite see how this could be a feature (which to my understanding implies usefulness). --Dschwen (talk) 16:42, 21 December 2013 (UTC)
Known problem, even in Category:Commons. User:Foroa should know more. --Zhuyifei1999 (talk) 09:58, 21 December 2013 (UTC)
Nice work, thanks for taking the time to research and experiment. -- (talk) 10:24, 21 December 2013 (UTC)

I have tweaked my algorithm so that the loop detection for the entire commons category set takes under 8 seconds on my crappy netbook (1900 loops detected). --Dschwen (talk) 17:56, 21 December 2013 (UTC)

Can I haz the code? --McZusatz (talk) 18:14, 21 December 2013 (UTC)
I'll put it up on github. --Dschwen (talk) 19:01, 21 December 2013 (UTC)
Very preliminary code is at --Dschwen (talk) 02:08, 22 December 2013 (UTC)
Also we need to fix those loops and make sure no new loops are popping up. --McZusatz (talk) 18:14, 21 December 2013 (UTC)

Intersection is still a tough one. I have a test case where I try to intersect Bridges and People. Bridges returns about 350.000 files (many of those are duplicates) and People returns over 82 million files. The quicksort of that return set alone takes over a minute on my laptop. I'm experimenting with mergesorting pre-sorted file lists now to see if I can get the time down substantially. But just to make this clear, I'm working without a depth limit here. --Dschwen (talk) 21:39, 21 December 2013 (UTC)

  • Intersection - I'll admit I just skimmed the code, but rather than sorting wouldn't it be quicker to take only the shorter list (in this case Bridges), do some sort of hash on that to generate a 'dictionary', and then make a linear pass through the longer list (in this case People) running the same hash and concerning ourselves only with elements that test positive for a hit on that 'dictionary'? (Feel more than free to tell me if you think I'm totally confused.) - Jmabel ! talk 06:51, 22 December 2013 (UTC)
    Jmabel, hashing may be a good solution if one of the categories is significantly smaller. A simpler variant would be to only sort the smaller cat and to binary searches on it. Which is essentially what a hash would do internally as well (so it be even faster). This is very easy to add as an additional strategy. The other strategy I'm exploring is merge sorting the (now presorted) file lists for each category (with a heap merge). This would take advantage of the presorting and should be very fast as well. The merge sort could be used for sorting one category in the binary search scheme as well. There is lots of room for optimization. And the bottom line is that arbitrary category intersections should be possible within seconds, which would be enough to make this a useful query tool. --Dschwen (talk) 16:52, 22 December 2013 (UTC)
    Also, if using the hash/bsearch method the result set would have to be postprocessed to eliminate duplicate entries. This could be done with a quicksort of the results (likely to be small and thus fast). This would also produce a sorted set for consistency. --Dschwen (talk) 17:17, 22 December 2013 (UTC)
    I committed a first version of the bsearch strategy. This takes the intersection with People down to under 15secs. Still too much for "primetime", but a start. --Dschwen (talk) 03:04, 23 December 2013 (UTC)
Yes, our category contains loops. Some are clear errors (self-categorized categories) and some are very hard to solve. In the past I made a simple child-parent table for our categories. On this table it's easy to find loops of different length like for example Category:Megoura, Category:Adam_Senn and Category:Micrelus_ericae. Also filtering of overcategorization for 8 levels of recursion was no problem (web interface). This has been broken for quite some time. If someone has a toollabs project I can put this in, I would be more than happy to set it up there.
Also check out this discussion on the Analytics list. It has a lot of overlap with what you're trying to achieve. Multichill (talk) 11:27, 22 December 2013 (UTC)
Nevermind, should probably just use catcycle. Multichill (talk) 11:55, 22 December 2013 (UTC)
CatCycle is ridiculously slow. I can detect the loops in the entire commons category tree in 8 seconds with unlimited depth. With CatCycle I have to select a single root cat, select a depth, and have to wait (to just get a gpProcessorException in my case, but that is probably just a fluke). --Dschwen (talk) 16:21, 22 December 2013 (UTC)

Making loop checks proactive

In addition to reporting loops and repairing them, would it be sensible to add a check at the point of when categories are added (using tools), so that contributors are warned if they are accidentally creating a loop/infinite regression problems? -- (talk) 11:34, 22 December 2013 (UTC)

Yes, Fae. I agree. This could be a useful tool. I'll start with detecting and possibly fixing or tagging the current loops. But I guess we'll also have to have a discussion about this. The first reactions above were like meh, sure we have loops. The first step is acknowledging that we have a problem ;-). --Dschwen (talk) 02:59, 23 December 2013 (UTC)

Commons Category Intersection

I have made some progress on category intersections with arbitrary depths (see my post above).

What it does
Return all images that are found in the subcategory tree of Category1 and Category2.

This could have the potential to be a powerful query method. I have implemented the intersection (∩) of 2 categories. The union (∪) of two categories is trivial to add. And the combination of multiple operations "(Cat1 ∪ Cat2) ∩ Cat3 ∩ Cat4" is also a simple extension of the existing algorithm.

In my current code the intersection time is pretty managable for smaller categories (100k-1M files per categorytree):

./fastcci_intersection 18193 20405 > /dev/null

0.16s user 0.69s system 99% cpu 0.855 total

That is less than one second (including loading the 600MB dataset, which in a persistent server would not be necessary. The pageids stand for Category:Bridges (343388 results) and Category:Uruguay (166919 results). The intersection contains 1856 images. --Dschwen (talk) 02:03, 22 December 2013 (UTC)

Looking at the results I found it contains this file (a bridge in Spain). I was already thinking bug, but there is a category chain (20405|1080476|2174197|110977|705494|9982954|26422515) from Category:Uruguay to Category:All_medias_supported_by_Wikimedia_España. Which makes this image part of the intersection. Maybe I should exclude all hidden categories? --Dschwen (talk) 02:30, 22 December 2013 (UTC)
Are you reinventing the wheel? Multichill (talk) 11:54, 22 December 2013 (UTC)
Sigh, looking at their example it looks like I'm inventing a wheel that is at least four times faster. But yeah, what the hell, somebody is apparently being paid for what I'm doing for free. I guess I can go home now. The question remains, why is this project not publicized more widely (or did I have my head in the sand?). And that project was brought to us by the same person who developed the Limeskarte, completely ignoring the existing community mapping projects. --Dschwen (talk) 15:31, 22 December 2013 (UTC)
Hi Dschwen, neat -- if your approach is 4x faster, you may want to let Johannes and particularly Daniel K know about it. --SJ+ 07:13, 23 December 2013 (UTC)

Wrong description

This flag is wrongly identified (see discussion). Many articles misuses this flag because of the description. How is to proceed now? --Arturius001 (talk) 10:28, 23 December 2013 (UTC)

GeoHack not working

When I click on the Camera location link on a file, I'm getting this error message:


You don't have permission to access /geohack/geohack.php on this server.

Additionally, a 403 Forbidden error was encountered while trying to use an ErrorDocument to handle the request.
Apache/2.2.22 Server at Port 80

Anyone know what's gone wrong? - MPF (talk) 12:38, 23 December 2013 (UTC)

down, see bugzilla:58888 --Steinsplitter (talk) 14:08, 23 December 2013 (UTC)
Thanks! Working again now ;-) MPF (talk) 16:48, 23 December 2013 (UTC)

TUSC/WMFlabs Geograph2Commons not working

The site is presently down with a "403 Forbidden" response. The site was temperamental, last night, but was working O.K. a few days ago. Does anyone know when it'll be working again, as I have some more images to transfer?

N.b. I have also been unable to transfer images from Geograph's grid square SJ3289, in particular, and at all times, if that is a separate problem. Whether that is a problem with Geograph or the TUSC software, I'm not certain. EP111 (talk) 14:05, 23 December 2013 (UTC)

down, see bugzilla:58888 --Steinsplitter (talk) 14:07, 23 December 2013 (UTC)
Thanks for the info. EP111 (talk) 14:13, 23 December 2013 (UTC)

{{Please use TeX}}

We have {{Please use SVG}}. I think will be good idea to have similar template for TeX and add it to MediaWiki:Gadget-UserMessages.js. Uploading images with formulas is typical beginners mistake. --EugeneZelenko (talk) 15:54, 15 December 2013 (UTC)

Symbol strong support vote.svg Strong support - use case - @EugeneZelenko: can you create it? Thanks, --Ricordisamoa 15:44, 19 December 2013 (UTC)
I didn't create such templates, but I'll try to find time to do. Or may be somebody more experienced may help. --EugeneZelenko (talk) 15:51, 20 December 2013 (UTC)
Just copy-paste the /SVG template and change SVG to TeX etc. --McZusatz (talk) 02:19, 21 December 2013 (UTC)
Symbol support vote.svg Yeah, I agree as well. There really is no reason to allow the upload of formula images anymore. --Dschwen (talk) 17:29, 19 December 2013 (UTC)
Template was created by User:McZusatz. I added it to MediaWiki:AxUserMsg.js. Translations are welcome :-) --EugeneZelenko (talk) 01:43, 24 December 2013 (UTC)

British Library "Mechanical Curator" collection -- update

The British Library's release of 1 million images to Flickr has had some great press coverage, and it's good to see Category:Images from the British Library Mechanical Curator collection starting to populate. (Note that there are some important details on the Commons project page as to how images should best be marked up, so we can keep their Commons pages in step with any information that gets added to the Flickr pages.)

However, one practical problem has been actually finding the potential collections of images of interest in all the material on Flickr. The BL are hoping that people will eventually be able to crowd-tag the collection, but as of yet very few tags are in place right now; so at the moment it's quite hard to find images about a particular place or topic. (And certainly hard to find all of them).

This is where we can come in.

Commons now has a full list up of all the titles that were scanned, and we are starting to work up a synoptic subject and place index to the book titles (though it's still at quite an early stage).

As a first step, the subject index has been been populated with a search for the word "history", for books with more than 15 images, and then the results organised geographically. This can produce quite a lot of hits, as shown by e.g the U.K. and Ireland page. But it's just the tip of the iceberg of what's still out there -- for most places and most topics there are still significantly better titles in the collection to be found than the ones so far included in the index.

So what would really make a difference -- what would truly unlock all this material -- would be every effort that can be given to go through the list of books, either from the top or for particular keywords, and systematically build up this index. We have the potential to really open the doors to make this collection usable. Jheald (talk) 20:00, 22 December 2013 (UTC)

Mechanical curator is certainly a misnomer (probably chosen with the mechanical turk in mind). There is nothing mechanical about using our volunteer "workforce". I hope the upload is done with care not to overwhelm commons with new uncategorized (and rather niche) images. It is tempting to jump the gun and flood commons with a million new files. --Dschwen (talk) 00:49, 23 December 2013 (UTC)
Whoa there!
So much in that short comment to respond to, and the night is already so late.
So please forgive me if I respond in stages.
  1. First up, there is a natural structure in this data. It's not just a completely random dump of images. There is a natural structure -- namely, the different books the images were scanned from.
    It's important to understand, I'm not asking for people to go out and classify a million images. What I'm asking is for people to classify 30,000 book titles.
    But actually it's not 30,000 book titles, it's much less. To start with, at least a third and maybe a half of those book titles are novels or poetry, which can probably be handled in different ways based primarily on their authorship, which we do have, rather than their subject matter, which we don't, but which (for non-fiction) it is really valuable to identify.
    And furthermore, we can get a huge amount of the value from much less than all of the titles. Just the top 200 books account for 20% of the images; the top 2000 account for 50%; the top 5000 (the first page of the list of book titles) account for 75% of the images; the top 10,000 (the first two pages) account for 90%. So a lot of the value can be gained, without having to do an unmanageable amount of work. Jheald (talk) 02:19, 23 December 2013 (UTC)
  2. Secondly, it can be done quickly. I've categorised 1000 of the titles in just the last four days, in my spare time. It's easy. Download the wikitext of the list of titles, cat them into grep, and presto, you have an instant list to populate a categorisation in the synoptic index, give or take a few false positives. (I actually use a little perl script that lets me gets lists for multiple different keywords at once, ignoring titles that have already been listed in the run, but that's just a wrinkle). Jheald (talk) 02:28, 23 December 2013 (UTC)
  3. Thirdly, it really makes a difference. If you go to, say the UK and Ireland page, and go down to 'Norfolk' in the 'East of England' section, if you are looking for material on eg Norwich Castle or the Norfolk Broads, you can immediately find it. That simply wasn't the case before. Jheald (talk) 02:39, 23 December 2013 (UTC)
  4. Fourthly, this is material worth having. It is already feeding into Wikipedia articles e.g. Barnard Castle (diff), or Napoleon III (diff), or Norfolk College for Young Ladies (diff) (different Norfolk), which is fundamental part of what we are here for. Jheald (talk) 02:53, 23 December 2013 (UTC)
  5. Fifthly, this is material worth having, even if it doesn't go into Wikipedia. Our Vision is for "a world in which every single human being can freely share in the sum of all knowledge". A key way we contribute is to find knowledge/content/material which wasn't previously readily accessible, to find it, to organise it, and to make it accessible. This fits directly into that, absolutely squarely. Now, I have a particular joy in the artistry of old prints and engravings, in seeing how places I know and love today (or places which have long gone, but were absolutely key in their day) were depicted topographically in the past. Yes, perhaps that is a "niche" interest. But I think it's a very proper one, for an outfit setting out to find information, organise it, and make it accessible. Previously this information information has been patchy, splintered, inaccessible: What topographical representations were there of a particular place in the past? Sometimes the research has been done, eg Cambridge, Oxford, London, variouslocal history projects. But for the most part places like the British Library simply have no idea of the full range of topographical images hidden in between the covers of their books -- it's simply never been catalogued, or at least not since what was available in the earliest years of the 19th century, before the battle of Waterloo. Nor are there very readily accessible systematic of the great entrepreneurial engravers, people like the Havell family before they got on to Audubon's birds. But now in an internet age, I can create a gallery like Illustrations and engravings of Cambridge, to 1923 and start to try to fill it (even if a lot of the images are still sitting on my hard drive, still waiting to be cleaned up and corrected and uploaded). And I can imagine that there could be a gallery like that for everywhere, and for each of the important artists and engravers, with all their known works, all systematically accessible; and to my mind that's *exactly* the kind of ambition that an image image bank like Commons sets itself up to be here for, with the vision expressed above, and can and should to be setting out to do. Coverage that is accessible, systematic, and complete. Jheald (talk) 03:57, 23 December 2013 (UTC)
Still more... Jheald (talk) 03:57, 23 December 2013 (UTC)
  1. What the BL has released is such a step forward. ... Comparison of eg what I had to do to try to gather together Loggan's Cantabrigia Illustrata (1690), scraped together from what I could find on print-shop websites all across the net, often lousy images that I've tried my untrained best to de-murk, patch together and sharpen up, and still even so with a fair number of gaps, compared to the quality and systematic completeness of what the BL is releasing; or Le Keux's Memorials of Cambridge (1847), cleaned up from University of Toronto scans with livid yellow-to-brown background gradients (though still better than anything from Google Books, dodged to a crisp), compared to the turn-up-and-go quality of the BL's Le Keux, Memorials of Cambridge, pretty much usable straight out of the box (even better if they could just get the files in page-number order) -- a set I only know about at all because I made the full list, did the grep, so now it's there for anyone to find in the merest moment, in its rightful place in the systematic index. Jheald (talk) 04:18, 23 December 2013 (UTC)
  2. When this material should get duplicated onto Commons. On the one hand, as soon as possible. Whilst there's something to be said for the care and love users give images when they hand-pick them and bring them here, there is also something to be said for the untiring systemicity and precision that a bot can bring to the task -- the advantage of a bot being able to install the images in an organised consistent pattern, before they can get uploaded hither and yon higgledy-piggledy by a disparate bunch of users all acting individually (yes, we *are* like herding cats, we know we are).
Contra Dschwen above, this would not be "jumping the gun and flooding commons with a million new [uncategorised] files" -- because, as mentioned above, there is a natural structure in this data, namely the different books the images were scanned from. That gives a natural categorisation of the files. But it also where the subject and place index that I'm so keen for people to come and help build becomes so important. Because this is the link structure needed to connect the new downloaded book categories with our existing ontology, to make the new categories discoverable. So the index is precisely what is needed to make sure that the new files do not become a morass of an uncategorised useless superfluous mess. Doing the transfer soon and by bot, before random individual uploads can have the chance to turn the thing into a complete midden, means that we can be sure that systematic organisation of each new image into a category by source-book will get done, and get done with as much consistency of naming as one can hope for, in the circumstances. An early and thorough transfer by bot would of course also have the benefit of avoid an olympic-sized shedload of completely unneccessary individual manual transfer work that would otherwise be inevitable; and, by lowering the barrier of work required, ought to make take-up and incorporation of the images into main wikipedias occure that much more quickly (though I do sometimes wonder if there is an attraction that killing an image in the wild and bringing it back to the camp-fire to eat may have, compared to simply picking same image off a supermarket shelf).
But there are a couple of considerations I can see that might mitigate against an early transfer. One is the rotation of the images. A huge number of the images are currently up on their sides, and need rotation. (In the past we gave the org a wikipedian in residence, and they set him to rotating images until his fingers bled... this time they just haven't bothered :g:). So maybe we should hold off until that has been addressed. On the other hand, maybe if we move the images here, we can experiment more directly and more immediately with how much can be programmed to be achieved automatically which we can then institute ourselves and show off the results of directly, rather than having to wait on the Library to do rotations when it sees fit. (also, Q: how does one try to automatically determine whether an image is appropraitely orientate. I can see one might try to consider any text around the image, or attempt to automatically identify ground and sky; but I would be interested in any other thoughts people might have tried and/or any links on the subk=hect).
A second, and I think stronger, consideration may be the sheer amount of duplication and repetition in the collection. Once a woodcut or a steel engraving had been created, it was not uncommon for it to be re-used an re-used again and again, turning up like the proverbial counterfeit penny. (See the comment on this image for one interesting "catch"). With decorative space-fillers and ornate set-off capital letters, which the collection is full of, the reiteration is even more extreme. I can see great value in creating a series of galleries each giving an alphabet of each such design with one example of each such letter, A-Z; but I am dubious that there is value for us in separately storing very many more impressions of essentially the same shape. The case is not quite so clear for the larger images, which would sometimes be adapted into subtly different "state" of the underlying design, which can be of relevance and interest in itself; I'm also not sure how one would determine which to be the "best" version of the image, when working on such a scale as to make manual intervention out of the question. But I certainly don't think we need 200 impressions of the same standard design of ornate boxed-capital letter T. So something I think would be good to develop is some kind of Tineye or Googe-like reverse image search capability, and then run it on the whole million images. I think it shouldn't be too hard -- I suspect there are essentially three things that are needed: firstly an algorithm to develop a multiple dimensional 'fingerprint' for each image, which is as invariant as possible to re-sizing or rotation; secondly, a way to identify close near neighbours in such a space, perhaps using a k-d tree, and finally a direct pair comparison algorithm, to make the final call as to whether any particular pair of images should be considered sufficiently similar or not. I suspect such a thing may well already, and (even be open source); but does anybody know, or know of any available research in the area? But it might be an idea to see the results of such a comparator, before making any transfer.
Finally, there's the question of the naming of the images. Do we go with the Flickr names, or try to organise out own, perhaps derived from any actual caption (which we ought to be be trying to read), or any entry in the 'List of Illustrations' (neither of which appears yet so far to have been used). But which or any of these should be allowed to hold uo the transfer, and for how long?
  1. one last section -- to cover how the project seems to have developed, the name "mechanical turk", and "are we being exploited?"; and to contend with Dschwein's rather negative view (held over) ... In contrast, it seems to me that for us this is an amazing opportunity - to show just how much going "open" can achieve for an organisation, to show just how much 'open source' can deliver, and deliver fast; and to develop a set of tools as free software that can be reused, eg that could be resued on images that could be extracted from the Internet Archive library; or the remainder of the BL's holdings (some enormous multiple of even this enormous release). If we can make a success and sufficiently exceed expectations on this releases, that might help them to find some more sponsors to scan and give away some more. Jheald (talk) 06:29, 23 December 2013 (UTC)
This is the best thing I've read all month, in so many dimensions at once. Thank you for tackling this project and opportunity with such elegance and organization. --SJ+ 07:22, 23 December 2013 (UTC)
This really is an excellent opportunity. Well done, and thanks for all the work you're putting in to help us make the most of it. --MichaelMaggs (talk) 08:27, 23 December 2013 (UTC)
Jheald, thank you for this excellent (and verbose ;-)) explanation. This does adress my (misplaced) concerns completely. I see that you have put a lot of thought into this project. I think I am excited now about how this is being handled. --Dschwen (talk) 16:30, 23 December 2013 (UTC)
Hi Jheald, although I can´t see exactly yet how the images will be imported, it is fun to contribute to the geographical index. One question: Is it necessary to keep the entries in the geographical index in the numerical order so that books with a lower number stay further up even within the country-specific subchapter? --Rudolph Buch (talk) 16:18, 24 December 2013 (UTC)

December 23

Commons usability degradation

Hi, bad situation:

  1. some time ago iwiki was moved to Wikidata: [11], [12];
  2. today its was reorganized: [13], [14].

As the result link to ru:Юэ Фэй was deleted from Category:Yue Fei. Now I need multiple clicks to find information about category subject. Before these edits I was need only single click. So Commons usability was decreased. How to handle the situation correctly? — Ivan A. Krestinin (talk) 22:47, 23 December 2013 (UTC)

It was due to bad wikidata link (wikidata:Category:Yue Fei instead of wikidata:Yue Fei), I have fixed it now. --A.Savin 23:40, 23 December 2013 (UTC)
You reverted bot`s edit. I think bot will make it again. This is not single category problem. Bot reorganizes all Commons categories: [15], [16], [17], [18], [19], [20], [21], [22], [23], ... Ivan A. Krestinin (talk) 04:29, 24 December 2013 (UTC)
Yes, you're right. I see now that my edits have been reverted by Multichill with reference to wikidata:Wikidata:Requests for comment/Commons links. I clicked on it, but TLDR, as you may guess. --A.Savin 10:26, 24 December 2013 (UTC)
d:Special:Contributions/BotMultichill seem making this situation worse. Also I doubt if the task is approved (or matching d:Wikidata:Requests_for_permissions/Bot/BotMultichill) --Zhuyifei1999 (talk) 11:04, 24 December 2013 (UTC)

On Wikidata we had an extensive debate on how to link. Pending that RFC I put my effort to connect Commons and Wikidata on hold. Now that the RFC is finished I'm implementing results so that's linking Wikipedia Categories to Commons categories, Wikipedia articles to Commons galleries and interconnecting these items using category's main topic (P301) and topic's main category (P910). Tracking and purging is not available yet, so to describe categories here we still need tools like sum-it-up. Later we can fetch this information from Wikidata. Multichill (talk) 17:18, 24 December 2013 (UTC)

"Wikidata:Requests for comment/Commons links" is shown that your bots edits are not consensual. This RfC has more questions then answers. See its talk page for more details. Approach "lets today degrade usability and hope that somebody in the future will restore it" is destructive. We need improve it continuously. So please if you want destroy something create better replacement before destruction, not after. — Ivan A. Krestinin (talk) 18:51, 24 December 2013 (UTC)

December 24

US miltiary musical recordings

I see that File:Silent Night (choral).ogg is PD as a work of the US military, could we get the rest of the album; preferably in lossless format? What about other recordings by US military bands, choirs etc? Do we have a category for such? Andy Mabbett (talk) 11:15, 24 December 2013 (UTC)

(audio) "The Stars and Stripes Forever"
The United States Marine Band performs "The Stars and Stripes Forever", The national march of the United States
Problems listening to the file? See Commons:Sound.

I checked under Category:John Philip Sousa for any links to a US Military bands cat and couldn't find any. He'd be the classic case as he wrote many of his compositions while a Marine, and presumably the whole process from day 1 until today would have been carried out "100% Marine". But I think downloading US military band music en masse is a great idea. I'm no music copyright expert, but the musicians are professionals, any arrangements are handled in-house, the only copyright question would be on the original composition. More complicated, but related questions
  • Can en:V-Discs be uploaded?
  • Can a QRcode be posted in the Congressional Cemetery or at Sousa's nearby house that could give somebody a quick Stars and Stripes fix?
Merry Christmas. Smallbones (talk) 18:24, 24 December 2013 (UTC)

British Library releases one million images to Flickr

Just thought some of you might be interested. linky. §FreeRangeFrogcroak 00:08, 14 December 2013 (UTC)

Wow, nice. --99of9 (talk) 10:16, 14 December 2013 (UTC)
Effectively, wow! Where are the bots to upload them to Wikimedia Commons? ;o) Yann (talk) 10:22, 14 December 2013 (UTC)

It would be nice if someone were to write this up as a proposed batch project at Batch uploading. Andrew Gray was a bit worried about the number of problems with identification, images needing crops and rotations and various other issues. None of this seems insurmountable to me, but as someone who has uploaded some large batches of clip art, I appreciate how many complaints and general stress trying to make an upload of this sort work will generate. With a million images to get through, it seems reasonable to spend quite a while discussing and documenting the issues and solutions.

By the way, doing it is not hard, just pump it all through Flickrripper. Doing it right will take a little longer. -- (talk) 10:35, 14 December 2013 (UTC)

Please see the project page at Commons:British Library/Mechanical Curator collection - in particular, note the request to tag the Flickr pages of images uploaded here, to tie the two pages together, so that later metadata additions can be aggregated, in either direction. Andy Mabbett (talk) 12:38, 14 December 2013 (UTC)
I think just upload them is ok. It saves time of uploading individually. Rotations and crop can be done when the photo is taken to use. The user can create an "other version", and it may have more than one version because some user prefer cropped photo, some prefer cropped with the yellow removed photo, some prefer transparent png photo, others like svg photo.--維基小霸王 (talk) 11:55, 16 December 2013 (UTC)
What's the copyright situation? I assume that they all can be loaded up without issue. JASpencer (talk) 22:45, 16 December 2013 (UTC)
“These images were taken from the pages of 17th, 18th and 19th century books digitised by Microsoft who then generously gifted the scanned images to us, allowing us to release them back into the Public Domain.”--維基小霸王 (talk) 12:32, 17 December 2013 (UTC)
«Svg photo», really? (Also, wow!) -- Tuválkin 03:37, 19 December 2013 (UTC)
I mean some people may convert the photo to svg format before using it.--維基小霸王 (talk) 05:26, 22 December 2013 (UTC)
And how do you feel about photos rendered as/within an SGV file?… -- Tuválkin 18:09, 23 December 2013 (UTC)
I'm sorry for using improper word. I intended to say line drawing, which is abundant in the image collection. --維基小霸王 (talk) 05:23, 25 December 2013 (UTC)


Is this category meant to be "Mathematical graphs"? - or rather a catch-all for any graph? What is a "graph" that is not a "diagram" or a "chart"? Please come to this talkpage and share your opinion about the prefered fate of the odd too-general Category:Graphs, a duplicate of Category:Diagrams that is hosting a mix of specific mathematics-related graphs side by side with an endless span of all other misplaced types of schemes, charts and diagrams. I need to get this cat (currently placed as a subcat of "Diagrams"!) somehow cleared of its over-ambiguous incoherent usage and can't decide whether make it a Category:Mathematical graphs or decline it as selfstanding cat by merging with Diagrams or elsewhere. Thanqz. Orrlingtalk 19:34, 24 December 2013 (UTC)

December 25

Adding a consent field to the upload form

What is the right forum to discuss adding a field in the upload form?

I'm wondering what a simple addition to the form would be to indicate consent? E.g., something like:

Does this media feature identifiable living people? (y/n)
[if y] Did the subjects give consent for their images to be published?
[ No consent requested / consent unknown / consent to photograph; consent to publish / No consent needed (public place); no consent needed (public figure) ]

As with (c) declarations, templates could be helpful for each of the categories of consent / non-consent. --SJ+ 09:08, 23 December 2013 (UTC)

Hi Sj,
I undertsand that you want to have a better publishing process, but the issue here is that nothing is back and white. First requirements differ from country to country, and then you would have to define who is a "public figure" and who is not? Our problem is that, because of this grey area, people try to game the system. Yann (talk) 09:26, 23 December 2013 (UTC)
The dedicated UploadWizard page would be but that feels really dead and unmaintained, hence not a good advice either... --AKlapper (WMF) (talk) 16:57, 23 December 2013 (UTC)
In cases of satire and parody for purposes of free expression related to commentary on public figures, this is not required, as explained in the unanimous decision of Hustler Magazine v. Falwell. -- Cirt (talk) 17:11, 23 December 2013 (UTC)
Copyright requirements differ by country too, and people try to game the system. So? This way we at least document the uploaders assertion of why the image qualifies.
It seems to me our alternatives are a)ignore permission status unless the subject complains or b)submit every image of a possibly living person to deletion review or c) employ some system to apply the appropriate template.
This is separate from the discussion of whether Commons can accept satire and parody of public figures. Dankarl (talk) 15:16, 24 December 2013 (UTC)
@Cirt, that's why I included "no consent needed (public figure)" in the proposed list of templates. I'm sure there are many others that I've forgotten, just as the (c) templates have expanded with time. --SJ+ 19:23, 25 December 2013 (UTC)

Thanks, @AKlapper. I posted a comment over there. That page gest a lot of traffic, in multiple languages... it seems worth maintaining :-). If the current extension maintainers can't do it, do we need to recruit more? How are new templates or options added to the system now, on different languages / for different jurisdictions?

Ideally the customizable parts of the extension like the individual form fields and # of options would be in wiki-editable pages... since every project implementing such a wizard would have its own set of requirements. update: Bawolff notes below that the right way to modify the U.Wizard is to file a bug. --SJ+ 19:23, 25 December 2013 (UTC)

Files transferred from Wikipedia

I just came across this deletion more or less by accident. This is not the first example I have seen of files transferred from Wikipedia subsequently being deleted. In cases like this, where the deletion reason is inapplicable or invalid in the US, can we please arrange to transfer the file back to Wikipedia so that that project is not disrupted. Failing that, at least inform administrators on Wikipedia so that the original Wikipedia file can be undeleted. If no one notices on WP, the material is effectively lost. SpinningSpark 03:18, 25 December 2013 (UTC)

I agree. As a workaround to avoid files ending up like this, we could create a simple template, to be placed either on the talkpage or filepage stating the original filename and origin of the file, with date transferred, reason for being transferred to Commons, and the transferring user. Then, at any time if the file has to be deleted at Commons, we could at least trace back and restore the file on the origin project.
If this is done, the template has to be embedded into the many bots and scripts that help transfer files to Commons. We don't need a "template" as such, but it would be clearer, compared to the messy text bunch found in filepages serving the same mentioned purpose. It could also maybe help add or remove functions easily in the future. Rehman 03:36, 25 December 2013 (UTC)
This is not a case of the deletion reason being inapplicable or invalid in the US. en:File:Scipio.jpg is no less a copyright violation on English Wikipedia than it was here, and the folks over at English Wikipedia really ought to know better; see en:Wikipedia talk:Non-free content/Archive 25#Photographs of ancient coins. LX (talk, contribs) 15:44, 25 December 2013 (UTC)
No, the folks at English Wikipedia just aren't lawyers. Tag all your uploads with {{Keep local}}. Commons is where things come to die, unfortunately. - Floydian (talk) 16:43, 25 December 2013 (UTC)
Sorry, I cannot hear your trolling over the sound of the 19.7 million free files we have here on commons. --Dschwen (talk) 16:58, 25 December 2013 (UTC)
Do you have a number for how many files have been transferred and subsequently deleted without a care? - Floydian (talk) 17:13, 25 December 2013 (UTC)
I'd think you would have done some homework if you make bold assertions like files come to commons to die. Just look at the deletion log and take a random sample. We have an estimated 1000-3000 deletions each month. Most of it obvious speedy stuff. It would surprise me if transfers from en would contribute more than a fraction of a percent. --Dschwen (talk) 17:54, 25 December 2013 (UTC)
"No" to which part of my statement? And why would I tag my uploads with a nonexistent template? And if you're referring to en:Template:Keep local, I haven't uploaded any files to English Wikipedia. Sorry, but you're not making any sense. The folks at English Wikipedia don't need to be lawyers, but maybe they should consider the advice of the Wikimedia Foundation's lawyers. LX (talk, contribs) 19:18, 25 December 2013 (UTC)
But in all seriousness, we do need a protocol for back-transfering and tagging files as unsuitable for common rather than just deleting them in such cases. --Dschwen (talk) 17:00, 25 December 2013 (UTC)
Just don't block uploading a local file when one with the same name exists on Commons. Copy suitable files here and don't delete the local versions. Problem solved. - Floydian (talk) 17:13, 25 December 2013 (UTC)
That would be very messy and confusing if uploading of different files under the same local and commons name were allowed. --Dschwen (talk) 17:41, 25 December 2013 (UTC)

Doubts about licensing from Sukhoi Superjet International Flickr feed

from the SSJ feed

In Category:Files from SuperJet International Flickr stream we have been gifted with many absolutely stunning or super-high-value/quality photos illustrating their jet. Unfortunately some of the comments in some of the files raise concerns for me. In File:Flying over Komsomolsk (5579237612).jpg we read "We found some interesting pictures on thanks to our friend @AFL271", and in File:Nose Landing Gear (6311220985).jpg we have "Marat Gizatulin provided us with these amazing pictures he took..." yet both files are CC-licensed with SuperJet International as the author/copyright owner.

Is it possible then, that this stream is an initiative from somebody at Sukhoi with limited understanding of CC licenses? Should we remove some/all of these files? Should we try to contact Sukhoi about this? Ariadacapo (talk) 17:15, 25 December 2013 (UTC)

I'd see if I could get a response; third-party images typically require OTRS. Dankarl (talk) 19:27, 25 December 2013 (UTC)
These files are in Category:Files uploaded by Russavia (Superjet), which is my cleanup category, and yes there are some photos like that which I have been deleting, and will delete these two. I'll make a concerted effort to go through the rest of the uploads in the coming days and clear that category entirely. Any photos left over should be all ok. I am in contact with SuperJet International, so please come to me and I can liaise if required, as I am talking to them about a couple of other initiatives which will benefit us here on Commons. russavia (talk) 02:10, 26 December 2013 (UTC)
Thank you Russavia, for the import and for the tireless clean-up. At some point, we should make the effort to thank Superjet in a more official way, for opening up such photos. Ariadacapo (talk) 07:45, 27 December 2013 (UTC)

Using DRs to confirm consent

TL;DRDR: A recent bulk DR of a category of photos of children suggested that photos of children that did not have an express statement of consent by their guardians should be deleted. The question came up: are DRs an appropriate? the most effective? way to ascertain consent in such cases, and what other processes might help with the backlog of identifiable photos without any consent indicators? We agreed that in addition to process changes, new tools are needed for indicating consent on upload, and policy should be clearer about what is required / what might be grounds for future deletion... but that was just the two of us having a chat, and a more public discussion s appropriate. So we're moving the discussion here. Please chime in:

All photos taken from the OLPC wiki now have a statement of consent. Yes, please update the consent policies so that all future photos of children require a statement of consent by the uploader. However, please do not DR and remove the [hundred] thousand existing Commons photos that include children but have no such express consent. --SJ+ 22:32, 21 December 2013 (UTC)

@SJ, can you clarify? I don't think anybody's about to take on nominating all such pictures for deletion, but it does seem to me that deletion review is the most efficient tool available for complying with the WMF resolution referenced above. Several examples of images I have nominated for deletion. In most cases, I think they have had good outcomes (i.e., clarity of consent, leading to deletion in some cases, but not in others):
Note: while none of these pictures contain nudity, the second might be a trigger depending on your perspective on bondage and sexuality. -Pete F (talk) 23:57, 22 December 2013 (UTC)
My main motivation in those and other deletion requests has been to encourage deliberation around how the WMF resolution on identifiable people should impact files currently on Commons. In many, maybe even most, cases involving an identifiable person, there is work to be done: contacting the photographer or subject, or otherwise making a determination about consent. Do you think DR is an appropriate way to move toward making a determination of consent, or not? Do you think those were legitimate nominations? -Pete F (talk) 23:23, 22 December 2013 (UTC)
@Peteforsyth, DRs are never great ways to clean up large backlogs, because of the timeframe imposed and the aggressive focus on 'deletion', when the intended outcome is actually some other form of cleanup. (often including cleanup beyond what is required to 'avoid deletion'.) The original uploader or photographer is regularly unavailable on the timeframe of a DR to contribute to the discussion.
DRs may be efficient for individual items, but not for thousands at a time. And there are thousands of images of identifiable people that need confirmation of subject consent.
When dealing with backlogs of this magnitude, a common approach is first to fix it for newly uploaded files, and then to work through past files in batches.
The upload form and process does not currently offer a way to indicate consent. This needs to change.
Commons:IDENT is relatively weak here: it does not currently require that the uploader positively indicate the subject's consent to publish, nor does it state that the lack of explicit consent is grounds for deletion. (It just hints many times that a positive indication would be useful; and hints that "published without consent" might be grounds for deletion.)
I could see policy shifting to say that, from now on, images of living people uploaded without indication of consent by the primary subjects may be deleted, without the active intervention and complaint of the subject. This would be a healthy change.
But dealing with the backlog will take longer. If this is handled similar to the way some other backlogs were: images with identifiable people can be catted as such (many already are), those with children with their own cat; then uploaders of those files contacted to update the image metadata with subject consent, linking them to the appropriate template. After a reasonable period of time, those that still haven't added consent could be catted as 'no-consent', and replaced with consentful images where possible, wherever they are in use.
After this I could imagine you nominating unused images for deletion if necessary; but before then it only seems appropriate in exceptional cases where you have specific concern about harm to the subject. --SJ+ 08:24, 23 December 2013 (UTC)
To your specific examples:
Karen Stollznow - the subject was asking for deletion. A clear case for a DR, appropriate of you to repost it and focus on the relevant problem.
Sauna (455500368).jpg - this didn't have to be rushed into a DR timeframe; you could have contacted the photographer and waited for feedback.
Black and White Striped Bondage.jpg - confirming consent was important. but by contacting the photographer and then ignoring his request, you may have simply insulted him.
Michael Paraire.tif - this was in use; the DR (instead of discussion on fr:wp) was not appropriate imo and offended the uploader, who only noticed it after (apparently) the deletion removed it from the subject's article
Lo'renzo Hill-White.jpg - the image was in use; a DR (instead of discussion with the uploader) was not appropriate imo.

@SJ, thank you for the detailed reply. There are some surprises in there for me - I think I have misunderstood some things about the way you, and perhaps other members of the Wikimedia Foundation board, have been regarding these issues. (I should say that, while I stand by my nominations and comments, I don’t want to make this discussion about me; I will follow up on that on your user talk page.) I am feeling hopeful that we can find some ways to move in a positive direction, so I hope you’ll bear with me through some details. I think there are two (related) common issues in the discussion around these issues, that I’d like to avoid: (1) extensive discussion of a "problem" without clearly stating what the problem is, and (2) finger-pointing and blame.

  1. The problem, as I see it: We too rarely defend individuals' personality rights, especially in contrast to the diligence with which we tend to defend the rights of those with intellectual property rights (like copyright). I would like Commons and Wikimedia to present a framework in which individuals feel they have reasonable options with regard to their personality rights.
  2. I will point out some ways in which WMF has inadvertently caused harm in this arena. I am very confident this is not through any ill intent, and bring it up for one reason only: I think that by developing a shared understanding of the dynamics, we might improve our ability to make things better. Although I think it's necessary to talk about causation a bit, it's not my goal to assign blame. (To the extent there is blame, I arguably own a piece of it, since I worked on the communications around the Upload Wizard as a WMF employee, and failed to anticipate these issues myself.)

In reply to your comment, there are two things I’d like to point out:

First, I think your statement above is inaccurate: (that COM:IDENT "…does not currently require that the uploader positively indicate the subject's consent to publish, nor does it state that the lack of explicit consent is grounds for deletion."). I say this based on the May 2011 Board Resolution ("…Strengthen and enforce the current Commons guideline … with the goal of requiring evidence of consent from the subject of media…), text from COM:IDENT at that time: ("there is a moral obligation on us not to upload photographs which infringe the subject's reasonable expectation of privacy,") and text from the current version: ("The photographer and uploader must satisfy themselves that, when it is required, the consent given is appropriate for uploading to Commons.") (All emphasis is mine.)

I'm simply pointing out that "must satisfy themselves" is vastly weaker and less enforceable than "must assert". The latter we can ask for, and any Commonsist can later observe whether or not the assertion was made. Since we don't currently ask for positive confirmation / assertion of this satisfaction, if I find an identifiable photo with no such assertion, I don't know what to do: should I assume the best and move on? Ping the uploader for more detail? Flag the image for slow cleanup and eventual (if not cleaned up) deletion? The latter is what we do now to ensure (c) cleanliness: even a good-looking image that you might assume the uploader had rights to, if uploaded with an inadequate license, will eventually be deleted. [–SJ]
Aha! OK, I was not getting that distinction, but it makes good sense. Yes, I think this is a good point for further exploration with the community. Thanks for the clarification. -Pete F (talk) 01:31, 26 December 2013 (UTC)

Second, you mention the Upload Wizard. I agree, this is a significant tool, which helps uploaders establish appropriate expectations. It's worth noting that the Wizard was moved out of beta in the same month as the Wikimedia resolution, but that even though personality rights were an area of timely concern, it it asked nothing whatsoever of the uploader about personality rights. And it has not been modified in the 2+ intervening years to do so. The Wizard was created by the WMF with substantial input from the Wikimedia community. But I am curious: does Wikimedia have a plan for how to shift the Wizard's behavior in order to support personality rights more effectively? Please note, I'm not saying it necessarily needs to be changed this moment; but if and when there is sufficient consensus around the significance of personality rights, do we know how to go about updating the Wizard accordingly?

I don't know how the UW is updated, and started a thread on the Village Pump about this before you left this comment. It is important for the community to be able to update and tweak the central tool that all projects use for posting media of any kind. --SJ+ 19:17, 25 December 2013 (UTC)
Absolutely. Thank you for pursuing that, I will follow up there and on Commons:Upload Wizard feedback. -Pete F (talk) 01:31, 26 December 2013 (UTC)

I think the WMF has some unique tools at its disposal to help calm the dynamics around these issues, and help us move toward a more collegial and productive dynamic that balances various interests. The WMF can help set the tone through the language it uses; and it has a significant ability to influence the software tools that guide the behavior of both new and experienced contributors. Can we find some ways to put those capabilities to good use here? A few suggestions to consider:

  • WMF could decisively disavow the phrase "Commons is broken," uttered in the past by WMF Board and staff members, which has gained some currency in many community members' commentary about Commons. I believe this statement, often accompanied by language about "rogue admins," is contrary to the principles of Assume Good Faith and Civility. I don’t think it's true – Commons serves many millions of files to many Wikimedia projects and beyond, vastly increasing the humanity's ability to share knowledge. There are problems, to be sure, but "broken" is a term that seems to imply that the negatives outweigh the positives.
    I hear you, however please note that these phrases and arguments have been made by non-Commonsists about Commons (and non-Wikipedians about Wikipedia, and Wikipedians about Wikipedia, &c.) for years. If you truly mean to follow your desire not to point fingers, please don't claim that this is something 'caused by Board and staff members' only later 'gaining currency in community commentary': that would not be correct.
    Commons is one of the gems of the free knowledge movement; and has the highest ratio of mindblowingly great work to active editors of any project (ok, wikidata and wikisource may recently be catching up here :) If anything, a core problem here is that we need better tools, and better way to recruit new contributors, to keep up with steadily geometric growth. [–SJ]
    You are right of course, and I am sorry I didn't speak more precisely. Of course I do not think the WMF caused these problems. I do, however, think it is particularly damaging when WMF personnel say things like this, and/or respond to them in public discussions without commenting on the incivility. I truly do mean my comment to be forward-looking; I think some dedicated "lead-by-example" commitment from WMF, more or less like "we commit to not disparaging the projects we exist to support," would be a very healing and welcome thing.
    As to your second point -- it truly warms my heart to hear this (even though I had no doubts as to your good will towards Commons). I put a fair amount of time into Commons, but not nearly as much as some others, and I'm sure others share my appreciation for hearing explicitly that the project is valued. Thank you! -Pete F (talk) 01:31, 26 December 2013 (UTC)
  • Update the Upload Wizard so that it asks questions relevant to personality rights, in addition to those relevant to copyright
  • Update policies and processes around bots and API use that upload large quantities of media: currently there is a great deal of attention to copyright in their design, but no attention to personality rights
  • Update the "No permission" sidebar item to clarify whether it means "no permission from the copyright holder" or "no permission from the subject"
    Absolutely on all of the above. I'm not clear on where each of these three updates should happen, but we should help speed them along. --SJ+ 19:40, 25 December 2013 (UTC)
Cool. I will watch the venues you linked above and help as I can, with regard to the Upload Wizard. And hopefully we can find ways to move forward on the other items as well. -Pete F (talk) 01:31, 26 December 2013 (UTC)
the upload wizard would need to be changed by devs. If the community thinks such a change is a good idea, please file a bug. In regards to "even though personality rights were an area of timely concern, it it asked nothing whatsoever of the uploader about personality rights" - devs aren't always up to date on the latest on wiki disputes (we have our own flamewars ;) and even when we are, its sometimes not obvious how to apply it to what we are working on. With that in mind, please tell us if there is something we should be taking into account when making things (via irc or bugzilla). Not all bugs get fixed, but zero of the bugs we don't know about get fixed.
the no permission thing in the sidebar is a gadget. Local gadget masters can change that if they agree with the change. Bawolff (talk) 20:30, 25 December 2013 (UTC)
Bawolff -- yes, I agree -- my comment above was not meant as a criticism of the devs, but more to raise the question of the processes that inform and interface with the devs. -Pete F (talk) 01:31, 26 December 2013 (UTC)
  • I have been bitching about COM:IDENT for a while now. I see an update in the UW going a long way to addressing the issues we have with lack of explicit consent. Saffron Blaze (talk) 05:32, 26 December 2013 (UTC)

Then there's always the approach seen in File:Kerkrade 2CV rechts.jpg, where even the dog's personality rights are fully protected! SFriendly.gif -- AnonMoos (talk) 00:36, 27 December 2013 (UTC)

Transferring to en-wiki

Clearly File:Olympic Stadium (Moscow).jpg is an FOP violation. Equally clearly, it would be appropriate for use with a non-free use justification on en-wiki. Is there any even partially automated way to transfer it there, comparable to the methods for transferring an image from en-wiki to Commons? - Jmabel ! talk 21:41, 25 December 2013 (UTC)

The file seems appropriate for use on en-wikipedia as a free file. As such, it should not have a non-free use justification. Template:FoP-US can be added to its description page. Commons fair use upload bot is unfortunately not set to transfer free-licensed files. It only transfers public domain files and non-free files. -- Asclepias (talk) 22:39, 25 December 2013 (UTC)
The uploaded version is poorly selected. The uploader toke the thumbnail from wikimapia. A version without watermark is available at the source The original, not the thumbnail should be uploaded to wp if required. --01:37, 26 December 2013 (UTC)
  • Asclepias, FoP-US is irrelevant: the building is in Russia. - Jmabel ! talk 07:49, 26 December 2013 (UTC)
  • I see, you meant FoP-USonly. - Jmabel ! talk 08:08, 26 December 2013 (UTC)
  • person w/ unsigned but dated remark: thanks. I guess that since the uploader is not the photographer, there is no harm at all in doing it that way & just ignoring the Commons upload. - Jmabel ! talk 07:49, 26 December 2013 (UTC)

December 26

Copyright of Finnish Airforce works

The images on are labelled "© Finnish air force". Does anyone know whether that implies an open licence for reuse, as for example, with crown copyright RAF images in the UK? Andy Mabbett (talk) 12:28, 26 December 2013 (UTC)

Minor point to avoid confusion. We upload recent RAF photos on an Open Government Licence rather than Crown Copyright. The latter is not free unless expired after 50 years. (talk) 00:47, 27 December 2013 (UTC)

Reducing filesize of PDF

I recently uploaded File:Shore Line Electric Railway Predecessor Companies.pdf (soon to be deleted as duplicate) but I was very unsatisfied with the differently sized pages. I'd been forced to scan it on two different machines, so the pages ended up very different sizes. I extracted the original images (22 PNG files, total size 687MB) and converted to .JPG (total size 121 MB). Each of these I then converted to .PDF - which resulted in a 787MB final product. Is there any way to convert the images into a similar PDF with a much smaller filesize? The existence of the first file (63MB) indicates it's possible, somehow. Pi.1415926535 (talk) 04:13, 26 December 2013 (UTC)

Cool upload, and interesting question. I don't have a definitive answer, but will throw out some observations in case it helps you move forward.
  • The file you have marked for deletion has an interesting line in its description, which makes me think it's corrupted in some way: "Original file ‎(10,600 × 14,600 pixels, file size: 61.26 MB, MIME type: application/pdf, 22 pages)" -- the other file has what seems like a more appropriate resolution (1,271 × 1,650). I don't know what to make of this.
  • Since JPG is a lossy compression algorithm, if there is any part of your process where you are re-applying JPG to something that has already been JPG'd, you are going to have a pretty terrible quality-to-filesize ratio. From your description it doesn't sound like that's happening, but I'm also not certain I fully understand each step you took (or what the software might have done without alerting you clearly). So, maybe worth a closer look.
  • As you may be aware, JPG is optimized for photos, but doesn't do so well for text. Nature of the beast, just pointing out on the off chance you don't know.
  • Have you tried converting the PNG files to PDF first, and then using JPG compression on the whole product (with the settings of whatever program you are using to create the PDF)?
  • Have you considered the DJVU format instead of PDF?
  • Have you considered uploading a low-resolution PDF or DJVU, and then higher resolution versions of the individual photos? (If it were me, that's probably the way I would go; and then reconstruct the work as a hypertext document, pulling in the photos, on Wikisource.)
  • Have you looked at the size of the color table? (Not easy for me to see in the metadata, so I haven't looked.)
  • It appears that your process is not generating a text layer. (See here.) It seems to me that PDF or DJVU offers two advantages over individual PNG or JPG files: (1) assembling them into a single file, and (2) creating a text layer. I don't know what software you have at your disposal, but maybe there is a way to address that?
Like I said, I'm a little mystified…I hope what I've said maybe contains some nugget that is useful. If not, sorry for wasting your time with my ramblings! -Pete F (talk) 05:01, 26 December 2013 (UTC)
You have very good thoughts here.
  • As I said, it scanned from two different machines. Both scanned at approximately 8.5x11" at 600dpi; one made the PDF 10600px wide, the other 765. Why? Who knows. That's why the original is so weird. It's not corrupted, though - still opens.
  • The scanners appear to have embedded the images as PNG files in the PDFs, but I can't confirm that. I don't have any Adobe editor, so I had to open them page-by-page in Inkscape. It appears the PDF conversion process takes them back to PNGs, though, hence the large filesize. I don't have any means to convert to text; they scanned as one solid image.
  • Again, not sure how to compress the PDF. Might take a while, too.
  • Have considered it; don't know how to work with it.
  • That I did consider; however, I'd rather have one high-quality source document than 22 scattered ones. I will be uploading most of the images separately as they'll be useful on Wikipedia.
  • Color tables I know nothing about. I usually work with individual image files - this is my first time attempting to assemble a full scanned document!
Thanks for your help; I'm going to continue thinking about this and try playing with DJVU. Cheers, Pi.1415926535 (talk) 05:34, 26 December 2013 (UTC)
Hi, The solution is to use DJVU instead of PDF. The same pages will come with a much smaller size in a DJVU file. However it is a bit tricky to create a DJVU without loosing quality. Regards, Yann (talk) 17:44, 26 December 2013 (UTC)
Thanks, Yann. I've now uploaded a much smaller (3MB!) DJVU file that can be used for most applications. Pi.1415926535 (talk) 19:56, 27 December 2013 (UTC)
I'm impressed! I found your recent upload: File:Shore Line Electric Railway Predecessor Companies 1961.djvu I'd be very interested to know more about the process that worked. Also, if you upload higher quality individual images, I'd be happy to help build a high quality document on Wikisource based on this... -Pete F (talk) 20:37, 27 December 2013 (UTC)

Eartha photo

Why was File:Eartha, the largest printed map of the world.jpg deleted? I don't understand the very brief reason given, "COM:DW". Was the photo of the globe deleted because its image quality was detailed enough to be considered a "derivative work" or "reproduction" of an unfree map? File:Eartha from outside.jpg also shows the same globe, but through glass windows. --Neitram (talk) 14:36, 27 December 2013 (UTC)

In the US, there is only freedom of panorama for buildings, not artworks, printed materials, etc. In the second example, it is a photograph of the building that houses Eartha and it would be practically impossible to do so without having the map show through a little. -- King of ♠ 22:06, 27 December 2013 (UTC)
But Eartha is a three-dimensional object, not a flat image. Isn't that the same as any photo of a globe? --Neitram (talk) 13:19, 28 December 2013 (UTC)
I think these images are copyright violations. I nominate them for deletion. Regards, Yann (talk) 13:32, 28 December 2013 (UTC)

Question about copyright status tag for File:YuJiaoLiEnglishversion.pdf

FOP for WWI and WWII memorials in France - mass delete?

December 29

Closure of Commons:Deletion requests/File:Jimmy Wales by Pricasso (the making of).ogv

December 27

GLAMwiki Toolset

Hello all. I want to inform you that the GLAMwiki Toolset, a tool to help GLAM groups (galleries, libraries, archives and museums) upload many pictures to Commons, was added to Wikimedia Commons on December 17, as a Special Page extension.

The main goal of the extension is to allow GLAMs the ability to mass upload content (pictures, videos, and sounds) to Wikimedia Commons based on respective metadata (XML); the intent is to allow for a wide variety of XML schemas. The extension goes about this task by presenting the user with several steps, represented by HTML forms, in order to set-up a batch upload process that will upload content and metadata to the wiki, which creates individual mediafile pages for each item uploaded. The project is co-funded by Europeana and a few Wikimedia chapters. Further information can be found at the extension page and at the project page. Your feedback and questions are welcome, feel free to contact us. Cheers,--Kippelboy (talk) 06:22, 27 December 2013 (UTC)

This is brilliant. Thank you Kippelboy, and everyone who is working on these tools. --SJ+ 10:58, 30 December 2013 (UTC)
Thanks I am sure we will put it to some good use. --Jarekt (talk) 13:31, 30 December 2013 (UTC)

Margaret Nowell Graham - Reynolda Barn (1922)


I just wanted to check in - There is an image of a 1922 work made by Margaret Nowell Graham on the cover/first page of the Spring 2013 journal: Would I be able to use this image, in that it was created before 1923 (I live in the US)?

Thanks!--CaroleHenson (talk) 16:13, 29 December 2013 (UTC)

  • The issue would be whether it was published before 1923. Nothing here really says either way. - Jmabel ! talk 19:47, 29 December 2013 (UTC)

December 30

Converting redirect

PLM is redirected to an Manilla university. There is also an French ex-railway company: PLM --> Chemins de fer de Paris à Lyon et à la Méditerranée of wich a category exist.Smiley.toerist (talk) 14:41, 30 December 2013 (UTC)

  • So turn it into a disambiguation page. Anyone can do that. - Jmabel ! talk 16:05, 30 December 2013 (UTC)

Server URL limit on Commons

Hello Face-smile.svg,

I have a little technical question for people who know about how Commons servers work, I also posted it on the French Bistro. When I need to do long and complex search requests, for example a keyword including or exluding several categories using "incategory", I would like to know what is the limit length of a URL we can use on Commons. I tried for my searches, it seems to be between 6500 and 9000 characters, but not sure because after 9000 I get an "Error 414", and with about 6500 sometimes the result page goes blank (then maybe the server seems not to like that lol). Sure it's already a lot of characters! Then if someone know exactly "how many" I can put on a search URL without getting errors or bugs, and without being considered as "harassing" servers... Because I didn't find how to do in other ways with tools like Catscan, because they don't include keywords. But if you have a tool or a tip to do otherwise it would be great too! And so sorry if I could do some mistakes in English Face-smile.svg. Thank you very much for your help! Jeriby (talk) 19:09, 27 December 2013 (UTC)

length limits on Urls can be worked around via post requests, but that's very difficult to do from a user perspective. (at least for the 414. The blank page indicates something wrong on php side which shouldn't happen). Bawolff (talk) 01:10, 28 December 2013 (UTC)
Thank you for your answer! Then the blank page should not happen? Where can I report this error? Maybe it depends on what I put in the search request, anyway if my user experience can be useful to resolve some bugs, I would be very glad to help php programmers with it^^ . Thank you again anyway. Jeriby (talk) 07:36, 29 December 2013 (UTC)
Because it looks like a search thing I'd report it here. If it turns out to be caused by something else whoever works on it will move it around. NEverett (WMF) (talk) 16:27, 31 December 2013 (UTC)

December 28

Geograph speedy category deletions

There are quite a few categories like Category:Images from the Geograph British Isles project needing categories in grid C0135 that are empty, and appear to have been nominated as speedy deletions by a template, but aren't actually listed in the contents of Category:Other speedy deletions. How does that happen? --ghouston (talk) 01:00, 30 December 2013 (UTC)

That's interesting. Has something changed recently? I'm sure I've seen them in the speedy deletion category before... Not that deletion of empty maintenance categories is time-critical. HJ Mitchell | Penny for your thoughts? 01:11, 30 December 2013 (UTC)
I have done a mass null-edit and Category:Other_speedy_deletions is now populated. The further speedy deletion tags by transcluding User:Avicennasis/emptycat will be hopefully observed by my bot. --Zhuyifei1999 (talk) 11:50, 31 December 2013 (UTC)


A while ago somebody has uploaded File:1995_Gaestel.JPG which was backed up by me on my computer. Now the file can be useful on one of my projects, but it seems to have been deleted. Unfortunately the information about the licencing wasn't recorded by me, and it'd be great if some admin can let me know the licence, the publication date, and the author, so that it'd be possible to correctly attribute the work. Thanks in advance. — Preceding unsigned comment added by (talk • contribs)

  • Author/uploader was User:Eurobas, uploaded 25 June 2009 with the remark "Peinture de Gaestel en 1995", licensing was GFDL cc-by-sa-3.0,2.5,2.0,1.0. User:Aavindraa cropped it 22 November 2010, removing the large border. It was deleted as being out of scope, no known problem with the license. It's a great image, too bad it was deemed out of scope, but I guess it's reasonable to say we're not a repository of user-created art, even good user-created art. - Jmabel ! talk 20:49, 30 December 2013 (UTC)
    • Thanks. This is exactly the information that was needed.
Checkmark This section is resolved and can be archived. If you disagree, replace this template with your comment. Jmabel ! talk 06:36, 31 December 2013 (UTC)

'Spirit of the Wild' opening

There are two issues with Category:Spirit of the Wild opening, Birmingham - 22 September 2005, which I have just created and to which I have uploaded a number of my photographs.

Firstly, all the files are mis-named ("Spetember", not "September"); my apologies. Could an admin batch rename them, please?

Secondly, as the photos were taken at the launch of a photography exhibition, I have been careful to only include parts of the exhibits, or distant views. Please can someone check that the images are OK, from a copyright PoV? Andy Mabbett (talk) 19:45, 30 December 2013 (UTC)

Regarding copyright, I'd say that File:Spirit of the Wild opening, Birmingham - 22 Spetember 2005 - Andy Mabbett - 02.JPG and File:Spirit of the Wild opening, Birmingham - 22 Spetember 2005 - Andy Mabbett - 01.JPG are a bit iffy, but the rest should be fine. -mattbuck (Talk) 21:07, 30 December 2013 (UTC)
✓ Done the first request. --McZusatz (talk) 14:58, 31 December 2013 (UTC)
Thank you. Andy Mabbett (talk) 16:06, 31 December 2013 (UTC)

Featured sounds

Commons:Featured sounds seems not to have got going (last edited in Nov 2012, marked as "forthcoming"; only five flies categorised as such. Would anyone be interested in helping me to kick start it? Andy Mabbett (talk) 12:21, 31 December 2013 (UTC)

a problem i need other eyeballs to check/verify/investigate


for some reason, Category:Days by day is listing dates out of order. it shows a quantity of dates "correctly ordered", then (on the 2nd page, partway through) it jumps back "in time" & shows another batch of "correctly ordered" dated, that SHOULD have been integrated with the first lot of dates.

im not sure how big the problem is (only checked the first few pages), but there are a lot of older dates that USED TO be on the first page, & are now "missing".

& this is a fairly new development. i work on date-cats regularly; it wasn't doing this the last time i visited the parent category (before today), & the total number of date-cats hasn't increased by all that much since the last time i was there.

the incorrectly ordered date-cats that i checked all appear to be edited correctly, they all use the same template, etc. they don't SEEM to be any "different"; at least not in any way that i can see.

so i have no idea what is causing it.

my guesses are: a) media wiki and/or server problem, b) something changed somewhere in the template architecture/infrastructure c) some kind of problem on my end

can some other editors pls go look &

i. confirm that/if there is a problem.

ii. pls help to figure out the cause.

thank-you in advance,

Lx 121 (talk) 14:06, 31 December 2013 (UTC)

Db refreshing lag (which I hate so much); needs mass null-edits. Details see database query result. --Zhuyifei1999 (talk) 14:45, 31 December 2013 (UTC)
Should be now fixed. --Zhuyifei1999 (talk) 15:06, 31 December 2013 (UTC)

Search β-testers for an application linked to Wikipedia

Hello, I am part of a group of 6 students of a French engineering school. We are creating an application for Smartphones which proposes touristic routes illustrated with Wikipedia articles. We have just finished programming and we are actually testing it. So, we need some testers who want to help us by testing and creating routes. If you are interested, please contact us :

Thank you,

Marianne— Preceding unsigned comment added by FatJagm (talk • contribs) 18:03, December 17, 2013‎ (UTC) 18:03, December 17, 2013‎ (UTC)