|
|
Broken redirects
[edit]I hope this is an uncontroversial addition because we have basically been speedily deleting broken redirects for obvious reasons in the past. However, I hope the wording I added was okay because blindly deleting every page on Special:BrokenRedirects is a bad idea due to MediaWiki's technical shortcomings – does the wording come across clearly? --shb (t | c | m) 07:39, 1 February 2025 (UTC)
User pages of touts
[edit]There is an ongoing discussion relevant to this policy: Wikivoyage talk:Don't tout#When should promotional user pages be deleted? It will probably lead to changes to this policy. Please check the discussion and share your view. –LPfi (talk) 09:39, 17 April 2025 (UTC)
Speedy deleting user pages of touts
[edit]There is a discussion in Wikivoyage talk:Don't tout#When should promotional user pages be deleted?.
It seems that we have a practice of speedy deleting user pages of touts, which isn't documented is Wikivoyage:Deletion policy#Speedy deletions. I assume that means that the policy should be amended. The question is whether any user page with promotional content can be speedily deleted at sight, or whether there needs to be some consideration of the specifics of the user page content or the user's contributions. If there is admin discretion involved, we should probably have some guidance.
At the moment, the discussion is about whether preserving a user page can be required for attribution purposes, or whether attributing the username itself suffices legally (in cases where the user's contributions are copyrightable, and we thus need to respect the licence).
Other issues should also be discussed.
Please share your insights and opinions in the linked thread.
–LPfi (talk) 08:26, 18 April 2025 (UTC)
Speedily deleting abandoned stub articles
[edit]It has come up more recently on Talk:Hồ Tràm / Vietnam and in a few VfD nominations that shouldn't need 14 days of discussion simply to delete, but this proposal in a simple nutshell is allowing any page in Category:Stub articles to be speedily deleted after 7 days of no edits. This is very similar to what enwikibooks does with abandoned stubs – many of which have unclear scopes and are poorly integrated with their respective book structures in a similar way that many stubs aren't integrated well with Wikivoyage's article structure.
For clarity, this does not address abandoned outline articles. This is specific to articles tagged with {{stub}}.
Beyond poor integration with article strucutre, there are a few other benefits to speedily deleting abandoned stubs:
- In the long run, it encourages users to actually use the standard Wikivoyage headings when creating a new article (the wording of the newarticletext header can be reworded if this proposal gains consensus).
- A VfD takes an unreasonably long time – not only does it take 14 days (which is already quite high), but factoring in an undefined grace period (which is usually about a month from the VfDs I've started), such stubs can end up being left as-is for 1–2 months.
- Because of the unreasonably long process it takes for them to be deleted, it leaves the onus of trying to integrate them with other articles up to other users, most of whom are not familiar with the place – all to save minimal, if any, travel content that is very negligible.
- Redirecting such destinations is also unfavourable because it is both:
- a) easier for a new user to create an article from a red link;
- b) less intuitive for readers as they woulld expect for the article where the redirect points to have reasonable content about that destination that was redirected. Stubs, however, don't have reasonable content.
As a first draft, I propose we add this clause to our deletion policy:
Articles tagged with {{stub}} may be speedy deleted if, after a period of seven days, they appear to have been abandoned and contain little to no substantive or salvageable content. Deleting admins are expected to exercise extreme discretion: if alternative actions, such as converting the page to a redirect, merging it into a more comprehensive article, or upgrading its status, would serve the project better, those options should be pursued in preference to deletion.
As for how to implement this, I am reasonably confident that we can use the same category mechanisms that enwikibooks uses quite easily so implementation is not an issue.
//shb (t | c | m) 13:42, 10 July 2025 (UTC)
- Support although I disagree in the case of Hồ Tràm, which took little time to clean up, in the of stubs that have no travel content and were created bya passer-by in 30 seconds, going through a two-week process that takes up contributors' time in discussion is not an efficent way of dealing with such articles. Ground Zero (talk) 15:27, 10 July 2025 (UTC)
- Right, the fact that you cleaned it up like that suggests that the summary deletion of that article without discussion would have been a mistake. So how do we make sure that by supporting the proposed policy, we don't remove usable content that could reasonably be whipped into shape? Should the proposal be tweaked in some way? I don't think we should be too hasty to speedily delete stubs with content in them that are started by new users who don't understand Wikivoyage breadcrumb navigation, article templates or style. Ikan Kekek (talk) 17:41, 10 July 2025 (UTC)
- That article was classified a stub because it had no headings. (It actually did, but they were not formatted as headings.) Perhaps this policy should be targeted to articles that have little or no travel information. I regularly delete on sight article that have the article template and nothing else as page creation vandalism. Ground Zero (talk) 22:12, 10 July 2025 (UTC)
- It was a stub mainly because it lacked important codes like IsPartOf. I agree about abandoned articles that have no or almost no travel information. The article-starter should be warned before we start deleting the articles they started if they are engaging in a pattern or they did this with one or a few articles and we want to find out if they have any plans to add content. Ikan Kekek (talk) 22:36, 10 July 2025 (UTC)
- I mean that is why I added "if alternative actions, such as converting the page to a redirect, merging it into a more comprehensive article, or upgrading its status, would serve the project better, those options should be pursued in preference to deletion." – Ho Tram also had other issues like touting that you wouldn't otherwise see in the stubs created by IP passer-bys. //shb (t | c | m) 23:11, 10 July 2025 (UTC)
- All of that makes sense. Ikan Kekek (talk) 23:26, 10 July 2025 (UTC)
- That article was classified a stub because it had no headings. (It actually did, but they were not formatted as headings.) Perhaps this policy should be targeted to articles that have little or no travel information. I regularly delete on sight article that have the article template and nothing else as page creation vandalism. Ground Zero (talk) 22:12, 10 July 2025 (UTC)
- I Support the option to delete articles of real places. When the issue is user-based (such as a user creating and then abandoning an article with almost no content) and NOT location-based (such as the article covers the location reasonably well but there simply isn't much there), I would much prefer deletion without prejudice over redirecting or merging. It bothers me that we encourage users to wantonly redirect and merge articles for reasons other than that the location cannot sustain its own article and that there is a sensible place to merge it into. Merging and redirecting are options that sometimes makes sense, but I don't think they should be encouraged. ChubbyWimbus (talk) 11:01, 11 July 2025 (UTC)
- Support, with the text advising admins to use discretion.
- The topic has been extensively discussed before; see Wikivoyage_talk:Deletion_policy/Archive_2014-2019#Deleting_NEW_empty_articles & several other places in the archives. Pashley (talk) 14:59, 11 July 2025 (UTC)
- Given there is clear support for this, I will add this tomorrow if they are no objections. //shb (t | c | m) 12:20, 16 July 2025 (UTC)
Done //shb (t | c | m) 12:49, 19 July 2025 (UTC)
Opposea rule on allowing speedy deletions after a week of not reaching outline status, as well as allowing summarily deleting articles with little content after a week. I support ChubbyWimbus' view that merging and redirecting isn't a silver bullet, and I think some stubs and nearly content-less outlines could be deleted without a VFD thread, in preference over redirecting. –LPfi (talk) 07:49, 18 July 2025 (UTC)- Comment. There are many scenarios where stubs or weak outlines are created. Creating a stub instead of an outline is mostly a newbie mistake, which shouldn't effect the decision to delete or not to delete (other than for established users who ignore pleads to add the template). For outlines, there are several scenarios, including placeholder articles created as part of a project and intended to be filled out – at least most of them – as part of that project, but perhaps not immediately. Somebody may have a listing or other fact for which they don't find a better place, perhaps because of lack of time during travel. Sometimes I create outlines on places I think need an article, but about which I know little; I might be able to write a short Understand or, for travel hubs, something in Get in. I think I recently created an outline on a place with a Michelin restaurant, reasonably far from any place for which we have articles.
- Often covering the place in the region article is worse than having the outline, and I think weak outlines aren't a problem other than when there are overwhelmingly many of them in a region, or when somebody directs their energy on creating them rather than using it to contribute real travel information.
- –LPfi (talk) 08:02, 18 July 2025 (UTC)
- This proposal doesn't even cover weak outlines for that matter – only stubs. Like Ikan, I don't get what part of this proposal you oppose either. //shb (t | c | m) 10:27, 18 July 2025 (UTC)
- If you think some contentless outlines could be deleted without a VfD thread, considering that that's the proposal, what part of it do you oppose? Ikan Kekek (talk) 08:10, 18 July 2025 (UTC)
- Sorry. The lead sentence talked about "any page in Category:Stub articles", regardless of content. For some reason I missed the proposal itself when responding. The key in the proposal is "with little to no substantive or salvageable content", and "if alternative actions…". With those wordings, I might agree, but I wonder whether such stubs really are common enough for a specific rule on them. –LPfi (talk) 10:29, 18 July 2025 (UTC)
- That's fair – I wouldn't say they're super common but I see them maybe once or twice every month – which I think for a wiki of this size, is significant enough to warrant a policy. //shb (t | c | m) 10:39, 18 July 2025 (UTC)
- Sorry. The lead sentence talked about "any page in Category:Stub articles", regardless of content. For some reason I missed the proposal itself when responding. The key in the proposal is "with little to no substantive or salvageable content", and "if alternative actions…". With those wordings, I might agree, but I wonder whether such stubs really are common enough for a specific rule on them. –LPfi (talk) 10:29, 18 July 2025 (UTC)
Redirecting typos
[edit]@~2026-19077-76: and @Lesothe: (who I suspect are the same person) have created quite a few redirects for names that seem to be one-letter typos. See Wikivoyage:Votes for deletion/March 2026#Qazaqstan for discussion of one that was nominated for deletion & kept. I want to speedy delete the others at [1] & [2]. However, as I read it nothing on the policy page would justify this.
Other opinions? Pashley (talk) 21:23, 28 March 2026 (UTC)
- I think we'd need to take it on a case-by-case basis. Some of them, like Qazaqstan or Mancester are very plausible redirects, but others like Machester or Grecce seem utterly pointless to me. //shb (t | c | m) 21:26, 28 March 2026 (UTC)
- I see both Lesothe & Botswane are now blocked but Special:Contributions/Lesothe shows many problematic redirects labelled "current". I still want to delete those. Pashley (talk) 14:33, 9 May 2026 (UTC)
AI-generated articles for speedy deletion
[edit]I'm mainly proposing this after User talk:Bigcee007, realizing that there's technically no policy-based grounds on what I did (I suppose a case of ignore all rules?), but would there be any opposition to adding "visibly AI-generated articles" (or something along those lines) to WV:CSD? I think it's a fairly obvious and reasonably uncontroversial addition. Pinging @Ikan Kekek, Ground Zero: as the main two other users who routinely deal with copyright violations on this site. //shb (t | c | m) 10:24, 6 May 2026 (UTC)
- I support this, with the caution that in a few years, it may be impossible to determine what is and is not AI. Ikan Kekek (talk) 10:31, 6 May 2026 (UTC)
- Hopefully AI detectors evolve as time goes on, but I do suppose it's worth reviewing this in about 3 years time. //shb (t | c | m) 12:47, 6 May 2026 (UTC)
- I also support this. Thanks for taking the initiative.Ground Zero (talk) 14:54, 6 May 2026 (UTC)
- I am inclined to support, but there are some cases that aren't clear-cut. I an unsure whether we should speedy delete quality-secured articles made with comprehensive prompting. They may be visibly AI-made but still salvageable with some copy-editing. The line to me is whether all information has been checked. Bot-generated articles are easy to recreate, so deleting those is no problem; they could be recreated in sandboxes and moved to mainspace as they get fact-checked and copy edited. Weak one-offs also aren't an issue: they may be seen as exercises by the creator, which we have no obligation to keep.
- This, of course, relates to whether we want to write an AI policy. Those using AI should be clearly warned about the possibility of speedy deletion, and such a policy would be an obvious place for any guidelines and the warning. Our deletion policy isn't among the first pages a newcomer should need to read.
- –LPfi (talk) 10:23, 7 May 2026 (UTC)
- Good points, though I would say even comprehensive prompting doesn't address the biggest issue with AI articles, which is their ability to hallucinate (thus inserting factually incorrect information). Of course, these articles will be much harder to detect, but I suppose we could make it future precedent to move such articles in userspace. And yes, I do think the time for an AI policy is a bit overdue now, though it does need careful crafting (which is for another discussion). //shb (t | c | m) 10:37, 7 May 2026 (UTC)
- If you find a carefully crafted article with signs of AI authoring, I don't see speedy deletion as the best way forward. If they are published at speed, the user needs to be blocked, but if they weren't, there is good time to talk to them to ensure they have fact-checked the articles before publication, or that they will do it now. If there are too many articles, they can be handled by an ordinary VFD. The phenomenon we need to counter is creating AI articles with little effort, where the risk of one-off socks is evident. –LPfi (talk) 11:29, 7 May 2026 (UTC)
- Good points, though I would say even comprehensive prompting doesn't address the biggest issue with AI articles, which is their ability to hallucinate (thus inserting factually incorrect information). Of course, these articles will be much harder to detect, but I suppose we could make it future precedent to move such articles in userspace. And yes, I do think the time for an AI policy is a bit overdue now, though it does need careful crafting (which is for another discussion). //shb (t | c | m) 10:37, 7 May 2026 (UTC)
- Have added the point under the speedy deletion section. //shb (t | c | m) 12:22, 12 May 2026 (UTC)
- Wikivoyage:Tips for new contributors#Contributing information has:
- Your contributions should mostly be original work, created by you. In general, please don't copy text or images from other web sites to Wikivoyage.
- Should something about not using AI (or perhaps how to use it) be added there? Pashley (talk) 14:39, 9 May 2026 (UTC)
- I added something. Comment or edits solicited Pashley (talk) 16:03, 9 May 2026 (UTC)
- I think we really need to write and discuss an AI guideline before telling what must be done or hint on AI use being accepted. I would say (partly borrowed from your text):
- Be very careful if you use AI. Wikivoyage does not yet have an AI policy, but if you use AI help in writing an article, you should at least check that there are no hallucinations (AI-invented "facts") and copy edit any generated text to suit Wikivoyage's style. Make sure every listing has an url that confirms the added information, and check the existence of any transport service you add. Using AI to find information is OK, but confirm it yourself; the prose is often better written by yourself, for style and to avoid potential copyright and plagiarism issues.
- I wanted to mention the lack of a policy and tried to make the text a bit shorter. I would prefer "if you consider using AI, read our AI guidelines" :-)
- –LPfi (talk) 17:50, 9 May 2026 (UTC)
- I think we really need to write and discuss an AI guideline before telling what must be done or hint on AI use being accepted. I would say (partly borrowed from your text):
- I added something. Comment or edits solicited Pashley (talk) 16:03, 9 May 2026 (UTC)
- I agree, we need an AI guidelines page & if we had one then the text in Tips could be shortened to something like:
- Also, please do not post raw AI output here. For details, see Wikivoyage:AI guidelines.
- Much of the text needed for the guidelines page is above; LPfi's version is definitely an improvement on mine. Pashley (talk) 13:32, 10 May 2026 (UTC)
- I have added a draft policy on Wikivoyage:AI guidelines, as we need to move ahead with a discussion especially as the discussion on a stricter global AI policy is under consideration by mediawiki Globetrotter30 (talk) 13:35, 16 May 2026 (UTC)
- metawiki Globetrotter30 (talk) 13:35, 16 May 2026 (UTC)
- ??? –LPfi (talk) 15:04, 16 May 2026 (UTC)
- metawiki Globetrotter30 (talk) 13:35, 16 May 2026 (UTC)
- A few days ago I made another draft: User:LPfi/AI_guidelines. I did not advertise it right away, as I thought I needed some more thinking about it that I hadn't time for. –LPfi (talk) 15:10, 16 May 2026 (UTC)
- I agree, we need an AI guidelines page & if we had one then the text in Tips could be shortened to something like: