Commons:Deletion requests/Files on User:Faebot/SandboxT

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search
This deletion discussion is now closed. Please do not make any edits to this archive. You can read the deletion policy or ask a question at the Village pump. If the circumstances surrounding this file have changed in a notable manner, you may re-nominate this file or ask for it to be undeleted.

These frames from YouTube videos have been correctly marked with a {{YouTube CC-BY}} license of CC-BY-SA-3.0, however the EXIF data added by Jan Arkesteijn states "Online copyright statement https://creativecommons.org/licenses/by/4.0/", which is false. The wiki text additions of the YouTube CC-BY and changes of a CC-BY-4.0 template to CC-BY-3.0 have been done in good faith by others, but it is not reasonable "housekeeping" to expect other volunteers to start making factual copyright corrections to misleading EXIF data.

I have discovered 258 files with this problem using wiki database queries, which verges on having to use forensic techniques, there are many other uploads by Jan Arkesteijn with similar contradictory or directly misleading copyright statements hidden in EXIF data that are both hard to find and potentially misleading for reusers. Deletion seems far preferable to using up volunteer time, though Jan Arkesteijn is always free to propose fixing their own work.

(talk) 11:56, 10 November 2018 (UTC)[reply]

Discussion

[edit]
Technically and practically, no, slapping a warning on these files is insufficient. Reusers will not always have our templates in front of them (the media viewer will not even display them), and any time the file is downloaded and shared, the only information reusers will have will be the legally false EXIF claim about copyright. To be fixed, someone needs to process these off-wiki, overwrite the files and delete/revdel the Jan Arkesteijn versions. Let's save on volunteer time.
However if someone wants to individually reupload especially useful frame grabs, now is the time for them to do so. Look, be practical, we have a 2 year backlog of over 3,000 Jan Arkesteijn fake colour overwrites that need to be reviewed, and even that does not deal with all the fake colour and fake EXIF creations that were not overwrites, but in those 2 years the backlog has barely been started on; mainly by me and honestly my volunteer time is much more useful spent on content creation. -- (talk) 12:29, 10 November 2018 (UTC)[reply]
 Oppose  Oppose  Oppose  Oppose  Oppose  Oppose  Oppose do NOT delete this for heaven's sake! Create a maintenance cat! Many of these were taken from the DWDD YouTube channel. Some of those videos later changed license on YouTube due to copyrighted music, which doesn't bother us anyway. Deletion also means all the license reviews will be lost. Also many of these are used for visual identification of notable individuals on nlwiki for which no alternative exists. As much as I hate the blue and yellow from Arkesteijn, even an image with wrong colors is preferable over no picture. - Alexis Jazz ping plz 12:37, 10 November 2018 (UTC)[reply]
I would support a longer time limit than 7 days on this DR, like a month, if someone is personally taking responsibility for getting the false copyright claims fixed. However the project should not be wishy washy about firmly enforcing our standards on copyright. Nobody should be able to upload false copyright claims and have them persist on this project indefinitely as if the mess they made does not matter.
By the way, the license reviews were done in good faith, but they are wrong so long as the false copyright statement is right there, displayed on every Commons image page above. -- (talk) 12:50, 10 November 2018 (UTC)[reply]
 Keep Useful images, but EXIF tags need fixing. I am willing to process 50 of these images. (50-258 =1 down, 4 to go.) What is the best procedure for these images? Vysotsky (talk) 13:34, 10 November 2018 (UTC)[reply]
@Vysotsky: wait 24 hours. Probably less. - Alexis Jazz ping plz 14:19, 10 November 2018 (UTC)[reply]
  •  Speedy Keep - Anyone who takes images from Commons probably never looks at the licence info but if they do I would imagine they only look at the licencing heading not the EXIF data, EXIF data can be fixed as I recall this being done for someones image before so it is possible,
Also as Alexis notes if some of the YouTube licences have been changed then that means we lose out all because of an honest mistake (probably a typo),
You could technically say "Well the licence info and the EXIF data say 2 different things so this invalidates the copyright" - I would say no it doesn't as the EXIF data doesn't really mean anything to the average editor here, I guess you could include some sort of note on each image that states the licence is 3.0 and not 4.0 and then something like it was a typo by the uploader. –Davey2010Talk 14:03, 10 November 2018 (UTC)[reply]
 Keep all Commons hoax Ymnes (talk) 14:31, 10 November 2018 (UTC)[reply]
  •  Keep There is a misunderstanding by . See CC Compatible licences. Note that YouTube videos use CC BY 3.0. A screen grab and crop is an adaptation of the video, and it is valid per CC to offer that with "a later version of the BY-SA license". The CC licences are designed to be upwardly compatible. I see no reason to delete the images nor to trash the EXIF data as Alexis has been doing today. -- Colin (talk) 16:57, 10 November 2018 (UTC)[reply]
 Keep This is not necessary, see above. The intensions of Jan Arkesteijn are good. Just keep all files and find another way to solve this. - Richardkiwi (talk) (talk) 17:02, 10 November 2018 (UTC)[reply]
 Comment @: I can change the EXIF on all the files. I've been there, so I've done five so far:
And all I got was this lousy thread on ANU. Sorry for the lack of edit comment - don't know how to set that. - Alexis Jazz ping plz 19:34, 10 November 2018 (UTC)[reply]
 Comment I would propose to park this issue, give Alexis the chance to fix as many images as possible, and revisit in, say, a month? Effeietsanders (talk) 20:39, 10 November 2018 (UTC)[reply]
Hi Effeietsanders, Personally I would say this doesn't need revisiting .... if Alexis changes all the EXIFs then Faes nomination/rationale would be considered moot as their only concerns are the licence (under desc) and the EXIF data. –Davey2010Talk 20:56, 10 November 2018 (UTC)[reply]
@Effeietsanders and Davey2010: I can fix every single one before midnight. But I should probably wait for an admin to greenlight this so I won't be blocked for helping. Most countries have a good Samaritan law. On Commons, the good Samaritan gets dragged to COM:ANU by Colin. - Alexis Jazz ping plz 21:06, 10 November 2018 (UTC)[reply]
Alexis Jazz, Exactly you could've been long done by now but instead everything's been put on hold thanks to those few people, Shame really. –Davey2010Talk 21:23, 10 November 2018 (UTC)[reply]
If Fae's nominations are moot after fixing, revisiting will be easy enough. Fae, can you confirm this would fix the main issue (assuming it's done well)? Effeietsanders (talk) 21:30, 10 November 2018 (UTC)[reply]
I pretty much said yes earlier.
I have no idea why people would want to waste time arguing that any of the faked EXIF data has educational value, I would be perfectly happy to see the EXIFs fully removed as YouTube frame grabs do not "natively" come with any EXIF data, nor does any Commons guideline encourage non-original EXIF data to get slapped on images before upload. As the evidence shows, it just gets in the way of digital verification. The only consistent thing we see is that Jan Arkesteijn can not resist playing with colours to suit their personal tastes, rather than sticking to the creative choices of the original credited, copyright holding, photographer or videographer.
As for taking someone to AN/U who stated here that they were acting in good faith and working collegiately to repair the files; toxic. -- (talk) 06:02, 11 November 2018 (UTC)[reply]
  •  Keep - per procedural reasons and above. From what I understand (although i could very well be wrong) is that this is due to a technicality malfunction rather than a direct violation of commons' policies. Inter&anthro (talk) 05:28, 11 November 2018 (UTC)[reply]
  •  Keep - issue that easily can be fixed and Alexis Jazz is willing to do that; so allow him to fix it and problem is over - Robotje (talk) 08:10, 11 November 2018 (UTC)[reply]
  •  Comment I have asked at the VP/Copyright about the CC4/3 issue. Of course the EXIF data has educational value -- it is exactly the data we add and insit on in our file descriptions. If someone said that a licence template on Commons had a 4 rather than a 3 in it, you wouldn't be happy if someone went round deleting all the text on all the file description pages. That's what happened here to the EXIF. Alexis has not behaved collegiately, but rudely, rashly and incompetently. -- Colin (talk) 09:40, 11 November 2018 (UTC)[reply]
  •  Comment This discussion could do with a few people learning what EXIF data is and what it is for. Information isn't "hidden" in EXIF as Fae claims, nor is the exif data below "fake" as Fae claims:
  • Title: Bert Meerstadt (2015)
  • Caption: Openbaar verhoor Bert Meerstadt - enquêtecommissie Fyra 28 mei 2015
  • Credit/Artist/Copyright: Tweede Kamer der Staten-Generaal
  • Terms of Use URL: https://creativecommons.org/licenses/by/4.0/
  • Embedded colour profile :sRGB
  • Source: https://www.youtube.com/watch?v=z7zdVpOxAyo
  • Dozens of fields related to the Photoshop CS2 that was used to create the JPG and which are themselves are record that the image has been Photoshopped.
The source video stream from Youtube has been screengrabbed. There was no JPG at all until Jan created one by saving his screen grab with Photoshop. So this isn't a case where Jan has altered existing EXIF on a JPG created by a museum, but one where a brand new JPG has been created just like when a photographer takes a picture. Nobody claims that when I add geolocation, licence, credit and title to my photographs that any of this is "fake". The same with a screen grab. EXIF data is useful to the world and is not hidden but quite visible on many quality image viewing/editing programs, and many of the fields are reproduced at the bottom of our file description pages. Professionals embed data into EXIF and indeed get very very upset if someone tries to remove that data because it is how the world indicates credit, licence, subject, title, etc on a JPG. What Jan did here was absolutely standard practice, absolutely desirable. I suspect the 4/3 licence digit was merely a mistake rather than some evil faking scheme that Fae wants to claim. Anyone who goes around removing accurate information from EXIFs or removing all EXIF data on Commons will face exactly the same threat of OVERWRITE sanctions that Jan faced. And anyone editing or offering to edit 258 JPGs needs to display some comptence with the tools first. -- Colin (talk) 09:56, 11 November 2018 (UTC)[reply]
  •  Comment I feel this DR is based on the assumption that coherence between the license statement and the EXIF data is a key element of the validity of the license itself. Based on this page of the Creative Commons wiki, I would consider it is not. The page says a correct statement "can boost confidence in embedded metadata", but it does not say an contradiction between the license explicit statement and the license field of the metadata compromises the license statement itself. As a side note, our current template for EXIF data says: "If the file has been modified from its original state, some details such as the timestamp may not fully reflect those of the original file". This may be sufficient to cover mistakes as those we are discussing or it might be better to try to improve the wording of this template. — Racconish💬 09:45, 11 November 2018 (UTC)[reply]
  •  Keep: They are free under either the youtube or EXIF license. Perhaps the image files can be multi-licensed or another solution can be found here? This is a technical mualfunction and not a deliberate error by the uploader. The copyright owner is still saying their images are free. So, we can assume good faith and keep them. Alexis Jazz's solution seems the most promising here. --Leoboudv (talk) 22:45, 11 November 2018 (UTC)[reply]

Kept: no real copyright issue - the EXIF may need or may not need some minor fix, but this is to be carried out only based on consensus. --Jcb (talk) 23:16, 11 November 2018 (UTC)[reply]