Commons talk:File renaming
This talk page is automatically archived by ArchiveBot. Any sections older than 30 days are automatically archived. Sections without timestamps are not archived. |
![]() |
---|
Participate in the vote & discussion on the proposed File naming guideline
[edit]Participation has been relatively low and it's becoming stale despite of it not being clear whether or not the draft will be adopted. This relates to file renaming in that this would be required less often since people have a guideline on how they and others should name files.
Commons:File naming – discussion and vote on this is here at VillagePump/Proposals. Prototyperspective (talk) 08:28, 17 April 2024 (UTC)
- Please participate. Thanks, Prototyperspective (talk) 11:43, 29 April 2024 (UTC)
Word choice in criterion 3
[edit]Normally I wouldn't be super-picky about a single word, but given that it appears in the templates and is a major policy and is put in bold, figured it'd be better to talk it over first. Currently criterion 3 reads:
- 3. To correct obvious errors in filenames, including misspelled proper nouns, incorrect dates, and misidentified objects or organisms
Now, I think this works in practice. But... why the word "obvious"? Shouldn't it be "major" errors, or perhaps "noncontroversial" errors? If, hypothetically, a picture saying it's one obscure species of beetle is in fact a similar-looking but different species of beetle, this kind of error may be deeply non-obvious. But it should still get fixed, right? Or, in the event of a credible dispute between good-faith editors, moved to a more general term (maybe kicking it up to the genus or family level, say). I brought the issue up in the Wikimedia Discord, and the feedback I got was that it was closer to "noncontroversial" in practice, i.e. undisputed. I'm not sure if that's perfect either - you could imagine a good faith but misguided editor kicking up "controversies" and thus blocking moves - but that would probably be fine 99% of the time.
(This came up because I requested a move that required quite a bit of research to realize that the source was indisputably wrong, but it involved text in Coptic, a language few people read. So a misidentification just stood for a decade+. Now, the image moved after all, so it clearly worked out, but I would not have called the error "obvious".)
Given the above, I'd like to recommend a new wording of:
- 3. To correct noncontroversial errors in filenames, including misspelled proper nouns, incorrect dates, and misidentified objects or organisms.
SnowFire (talk) 05:13, 8 June 2024 (UTC)
- wikt:obvious covers what we are trying to explain. I don't see that your change improves the situation. — billinghurst sDrewth 10:45, 8 June 2024 (UTC)
- Are you actually advocating that only obvious in the sense of "Easily discovered, seen, or understood; self-explanatory" files should be moved, or are you invoking that page in some other way? Because the case of there being a major error that is self-evidently not easily discovered (i.e. it takes 10 years to notice) absolutely comes up. SnowFire (talk) 18:19, 10 June 2024 (UTC)
- wikt:obvious covers what we are trying to explain. I don't see that your change improves the situation. — billinghurst sDrewth 10:45, 8 June 2024 (UTC)
- The way that I've always handled this as a filemover is that it's the requestor's job to make the error obvious to me. If the error isn't obvious from the file then that will usually require providing extra evidence. For the beetle, that might be a link to an authoritative explanation of the differences between the species, or a discussion on a Wikipedia talk page, or a tweet by an eminent entomologist saying we've got it wrong. Once I've seen the evidence, I can decide whether that's enough to make the error obvious and hence justify the renaming.
- "Noncontroversial" is much harder for the filemover because it means that they have to somehow work out whether anyone disagrees with the request. I'm not sure how you'd determine that. Maybe by announcing every criterion-3 rename in advance and having a period for objections? That would certainly make the process more cumbersome. Without an objection period almost every request would appear to be noncontroversial, even those where a small amount of research would demonstrate that the current name is correct.
- "Obvious" is probably not quite the right word, since it doesn't really capture the "given the information at hand" part, but I can't currently think of a better one. --bjh21 (talk) 12:26, 8 June 2024 (UTC)
- agreed.
- even with the current wording, far too often requesters give non-obvious, very brief reasons, and assume other users will understand. RZuo (talk) 16:16, 8 June 2024 (UTC)
- Maybe proven errors, or provable errors, would be a better choice of words, as, in my experience, it is not rare that some people will not accept something as obvious even if it has been logically or empirically proven. Abderitestatos (talk) 11:12, 9 June 2024 (UTC)
- I would be fine with changing to something like "Clearly explained errors" instead, or some variant, if the goal is to encourage requesters to make their case. SnowFire (talk) 18:19, 10 June 2024 (UTC)
- Maybe proven errors, or provable errors, would be a better choice of words, as, in my experience, it is not rare that some people will not accept something as obvious even if it has been logically or empirically proven. Abderitestatos (talk) 11:12, 9 June 2024 (UTC)
Redirects for misidentified people
[edit]File:Georg Andreas Sorge.jpg was a picture of Georg Andreas Reimer that was uploaded in 2015 and corrected in 2016, being renamed to File:Georg Andreas Reimer.jpg. However, the redirect means that the picture still shows up in search results for "Georg Andreas Sorge" (resulting in it being added to Wikipedia biographies of him by inattentive editors since the rename).
This guideline suggests that the redirect shouldn't be removed as the file wasn't "recently uploaded" at the time of the rename. Is there another way to handle this kind of belatedly-realised misidentification so that Commons doesn't perpetuate the confusion? Belbury (talk) 11:08, 1 August 2024 (UTC)
- i often ignore that rule and move files without the erroneous redirect.
- otherwise, com:csd g2. RZuo (talk) 11:17, 1 August 2024 (UTC)
- I don't think G2 applies, since the whole problem is that it is a used redirect. VanIsaac (en.wiki) 00:19, 2 August 2024 (UTC)
- where? RZuo (talk) 06:32, 2 August 2024 (UTC)
- There's a caveat in this guideline that urls to thumbnails of the old file name will also redirect (however, direct links to the full size version of the file will not redirect), meaning that other websites which had embedded the old thumbnail directly would be affected, if we cared about that. But is that statement about redirection actually correct? Trying to access a 75px thumbnail of File:Georg Andreas Sorge.jpg gives an error message rather than a redirect to a thumbnail of the Reimer filename. Belbury (talk) 08:09, 3 August 2024 (UTC)
- I don't think G2 applies, since the whole problem is that it is a used redirect. VanIsaac (en.wiki) 00:19, 2 August 2024 (UTC)
- @Belbury: The guideline only says that the redirect should be created. It doesn't stop you proposing it for deletion immediately after creation, either speedily under G2 (which I think applies in this case) or as a full DR. Essentially, invalidating a long-standing filename is a matter that should be escalated to an administrator. --bjh21 (talk) 14:59, 2 August 2024 (UTC)