Wikipedia talk:Deletion policy

This is the talk page for discussing improvements to the Deletion policy page.

Put new text under old text. Click here to start a new topic.
New to Wikipedia? Welcome! Learn to edit; get help.

Archives: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52: 6 months

The project page associated with this talk page is an official policy on Wikipedia. Policies have wide acceptance among editors and are considered a standard for all users to follow. Please review policy editing recommendations before making any substantive change to this page. Always remember to keep cool when editing, and don't panic.

Archives

Non-numbered archives from 2003

This page has archives. Sections older than 180 days may be automatically archived by when more than 3 sections are present.

Storage of deleted articles

Currently, the section Wikipedia:Deletion policy § Access to deleted pages includes a sentence stating that deleted articles remain in the database (at least temporarily) - my emphasis. WikiBlame tells me that this was inserted in 2008 with Special:Diff/241376527.

The qualifier at least temporarily can be read to imply that deleted articles will be permanently erased after a retention period, which is contrary to my understanding that deleted pages/revisions are kept in the database indefinitely. I’m therefore proposing to remove that qualifier (my reason for starting this discussion rather than making the edit boldly is because I wanted to make sure that my understanding is correct/that there wouldn’t be any other problems with making this edit).

All the best, ^user:A smart kitten_meow 12:39, 8 December 2023 (UTC)[reply]

My understanding is that the WMF do not guarantee that deleted revisions will remain available in perpetuity. The likelihood of deleted revisions ever being permanently deleted is massively lower in 2023 than it was in 2008, but at least theoretically still possible. The original version of Oversight (pre 2009) also permanently deleted the relevant revisions, although it's unlikely that was what was being referred to. Thryduulf (talk) 20:05, 8 December 2023 (UTC)[reply]

Since it would be an extraordinary event, maybe we should still remove that part. Aaron Liu (talk) 21:22, 8 December 2023 (UTC)[reply]

Meh. It's still accurate, and I see no reason to increase the level of expected retention. We know it's probably sticking around... but do we want to promise that? Don't think that's our place as a community: we don't own the infrastructure. Jclemens (talk) 00:57, 9 December 2023 (UTC)[reply]

Maybe we should just replace it with footnote [c] from Wikipedia:Viewing and restoring deleted pages Aaron Liu (talk) 02:32, 9 December 2023 (UTC)[reply]

I think it would be better to copy that footnote here and add it to the end of the text quoted above, giving it context, rather than replace it. Thryduulf (talk) 02:44, 9 December 2023 (UTC)[reply]

Possibly also it might be worth speaking to the devs to confirm that statement is still accurate nearly 17 years later. Thryduulf (talk) 02:49, 9 December 2023 (UTC)[reply]

I’ve emailed answers

wikimedia.org with the query and a link to this discussion, so hopefully someone from the WMF will be able to provide the latest information. Best, ^user:A smart kitten_meow 09:43, 9 December 2023 (UTC)[reply]

Deletion means deletion. The deleted page archives ARE TEMPORARY TO FACILITATE UNDELETION OF PAGES WHICH SHOULD NOT HAVE BEEN DELETED and are subject to being cleared or removed AT ANY TIME WITHOUT WARNING. --brion 00:50, 19 January 2007 (UTC)

Wikipedia:Village pump (technical) 19 January 2007

Emphasis and ALLCAPS as per the original.

- SmokeyJoe (talk) 10:30, 9 December 2023 (UTC)[reply]

This is what we are discussing directly above. A smart kitten has emailed to see if this 17-year-old statement is still accurate. It would seem foolish to do anything before we get an answer. Thryduulf (talk) 10:36, 9 December 2023 (UTC)[reply]

I think this is a reference to edit histories being quite unreliable in the very early days of the encyclopedia (I believe before around 2003) see WP:UuU. It may be technically correct but unnecessary in practice. --Trialpears (talk) 14:29, 9 December 2023 (UTC)[reply]

I vaguely remember there was a policy decision to never flush deleted revisions due to the CC-by-SA attribution requirement. But that could just be leaky neurons conflating different discussions. RoySmith (talk) 15:40, 9 December 2023 (UTC)[reply]

You find all sorts of weird crap if you look through the primordial database. For example, WP:VPT#* in comment table? that I found yesterday. What's really weird is that revision_ids aren't (weren't?) assigned in monotonically increasing order. Step through the earliest history of the WP:UuU in chronological order. The revision ids go:

291430
385544927
302608
13692247
15927838

My brain hurts. RoySmith (talk) 15:56, 9 December 2023 (UTC)[reply]

Heh, turns out this is documented

Note that while rev_id almost always increases monotonically for successive revisions of a page, this is not strictly guaranteed as importing from another wiki can cause revisions to be created out of order.

RoySmith (talk) 16:19, 9 December 2023 (UTC)[reply]

Which is exactly what happened. —Cryptic 01:25, 10 December 2023 (UTC)[reply]

I've had an email response from the WMF as follows:

I've heard back from Legal, who noted that they do not see a need to change the wording in that section of the deletion policy at the moment.

All the best, ‍—‍a smart kitten_[meow] 08:06, 4 May 2024 (UTC)[reply]

Striking blocked users at AfD?

Isn't it customary to strike blocked users, such as sockpuppets, within AfD discussions? I understand that the AfD discussion can continue even if they are the nominator, but we generally strike their comments regardless. @Beccaynr @Another Believer Cielquiparle (talk) 20:35, 24 March 2024 (UTC)[reply]

@Cielquiparle: I feel like I see comments by blocked editors crossed out often, but maybe that's something I should leave to admins. I didn't mean to overstep, and I gave permission for Beccaynr to remove the strike. Makes no difference to me. I also asked at User_talk:MER-C#AfD_comment, since I saw MER-C comment on the block in other AfD discussions. ---Another Believer _(Talk) 20:40, 24 March 2024 (UTC)[reply]

According to what appears to be the relevant part of the Talk page guidelines, that I linked to in that discussion, Removing or striking through comments made by blocked sock puppets of users editing in violation of a block or ban. Comments made by a sock with no replies may simply be removed with an appropriate edit summary. If comments are part of an active discussion, they should be struck instead of removed, along with a short explanation following the stricken text or at the bottom of the thread. Previously, when I have attempted to clean up after sock-related !votes, I recall having strikes unstruck because I did not follow this precisely. Beccaynr (talk) 20:44, 24 March 2024 (UTC)[reply]

@Beccaynr That's what AB did – strike the comment instead of removing it, with a short explanation following. Can you please restore the strike? Cielquiparle (talk) 20:49, 24 March 2024 (UTC)[reply]

There is no indication that I can find that the comment was made by [a] blocked [sock puppet] of [a] [user] editing in violation of a block or ban. Perhaps MER-C can offer some guidance here; based on my past experience with having strikes unstruck when I thought the TPO provision applied to editors socking generally (and there being no indication socking is involved here), I would prefer to rely on my understanding of the guideline and experience, and wait for additional guidance. Beccaynr (talk) 20:59, 24 March 2024 (UTC)[reply]

It's customary to strike people who are using more than one account in the discussion, so they don't get counted twice, and users who are sockpuppets of blocked/banned users, since they are not allowed to edit. A common mistake in this context is to strike users who are subsequently blocked for a reason other than having a prior account. It is a common mistake, but it's still a mistake. -- zzuuzz ^(talk) 21:00, 24 March 2024 (UTC)[reply]

Thanks for the clarification @Zzuuzz. Perhaps it's enough then to just add a comment after their comment making it clear that they were subsequently blocked? Cielquiparle (talk) 21:23, 24 March 2024 (UTC)[reply]

That is also customary, perhaps (optionally) explaining why they were blocked. -- zzuuzz ^(talk) 21:25, 24 March 2024 (UTC)[reply]

Commenting on this case - I think this user is a UPE spammer, that's why I blocked them. The problem I described here still exists and probably has become worse. I don't particularly mind whether their comments are struck, but UPE spamming elsewhere does have a negative impact on whether the vote/comment is in good faith. UPE spammers are more likely to be socks too. MER-C 17:21, 25 March 2024 (UTC)[reply]

If thje user's blocked as a LTA or a sock or a UPE then yes; if they're just a normal user who got blocked for a totally unrelated thing I would say no. jp×g 🗯️ 02:58, 29 April 2024 (UTC)[reply]

If the sockpuppet has started an AfD and there are no other delete comments it can be speedily closed as per WP:Speedy keep Applicability criteria 4, imv Atlantic306 (talk) 21:01, 24 March 2024 (UTC)[reply]

Discussion at Wikipedia talk:Revision deletion § Process for requesting revision undeletion

You are invited to join the discussion at Wikipedia talk:Revision deletion § Process for requesting revision undeletion. ‍—‍a smart kitten_[meow] 12:45, 28 April 2024 (UTC)[reply]

Comuni

I propose the deletion of most of the Italian comuni. One thing I have never understood about this encyclopaedia is this: what's the point of creating so many pages (over 8,000 pages of Italian comuni) and then leaving them to their own devices? This isn't the way to treat pages. I do my best to improve them, but not even in forty years would I manage to improve 8,000 pages of comuni. JacktheBrown (talk) 19:16, 3 June 2024 (UTC)[reply]

Sadly, the problem is not specific to Italy. Many other countries have numerous articles about places with a handful of residents (probably one or two ordinary houses) which are apparently notable. Certes (talk) 22:55, 3 June 2024 (UTC)[reply]

Be ruthless and bold. There was a set of village articles I looked at about a year or so ago, and after checking all of them I realised that a) they were mass-created, and b) didn't have anything other than a name and location (and maybe a population count). Redirected them all to the district they were found, and (as far as I know) they haven't been rewritten. I'm all for improvement but for some things it just doesn't make sense to waste time doing the research for such a little improvement. Primefac (talk) 19:41, 6 June 2024 (UTC)[reply]

No, I can't spend my whole life on Wikipedia. I'm already "ruthless and bold" about Italian cuisine. In September 2023, I started to improve the pages on Italian cuisine and now, after a long time, I'm very satisfied with the great, enormous results; and I'm not finished yet. I'm sorry, but I'm one, not thousands. JacktheBrown (talk) 19:57, 6 June 2024 (UTC)[reply]

That's fair enough. I suppose that 8k pages is a bit much to go through, even with something like AWB. Batch nominations for deletion would be possible as well, but again, even if only 10% of pages fall into the AFDable category, that's still 800 pages... Maybe the best option would be to start an RFC at WP:ITALY to see if there's a general consensus to just redirect them all (at which point a bot could take care of the actual editing). Primefac (talk) 20:08, 6 June 2024 (UTC)[reply]

@Primefac: self-correction: 7,904 comuni in 2021, but nothing changes. JacktheBrown (talk) 20:13, 6 June 2024 (UTC)[reply]

New users creating articles in talk pages: G8 or draftify?

I am seeing a lot of new users attempt to sidestep restrictions on article creation by putting material in a talk page for a non-existent article. The material is usually too poor to be a useful draft, or violates what Wikipedia is not. The pages often do not meet any CSD criteria besides G8, although they might meet article-specific criteria if they had been created in articlespace. When should these talk pages be draftified instead of tagged for deletion? Helpful Raccoon (talk) 20:19, 8 August 2024 (UTC)[reply]

Isn't that just what G8 is meant for? See Wikipedia:Criteria for speedy deletion#G8. Pages dependent on a non-existent or deleted page:"G8. Pages dependent on a non-existent or deleted page: Examples include, but are not limited to: Talk pages with no corresponding subject page ...". Straightforward. PamD 20:43, 8 August 2024 (UTC)[reply]

One time someone removed my G8 tag and draftified the page. Draftifying might be useful at least for plausible drafts in talkspace. Helpful Raccoon (talk) 20:52, 8 August 2024 (UTC)[reply]

Although, the page that was draftified was essentially an essay. Helpful Raccoon (talk) 20:55, 8 August 2024 (UTC)[reply]

Sometimes the page is in good enough shape to be moved to article or draftspace. Otherwise, G8 might apply. Jo-Jo Eumerus (talk) 06:42, 9 August 2024 (UTC)[reply]

Agreed; if it's a clear non-starter I'll nuke it with G8. If it looks like it might be worth salvaging I might move it to the Draft space, but I could probably number those with two hands. Primefac (talk) 13:03, 15 August 2024 (UTC)[reply]

Amending ATD-R

The footnote to ATD-R cites RfCs from 2018 and 2021; in the latter discussion, [m]ost users believe that AfD should be used to settle controversial or contested cases of blanking and redirecting.

In a discussion at WT:R, it was pointed out that the current language of ATD-R presents AfD as merely an option, rather than the preferred venue, for contested BLARs. Thus, I propose rewording ATD-R as follows:

voorts (talk/contributions) 01:26, 9 October 2024 (UTC)[reply]

The current wording is better. If the status quo is an article, then AfD is the preferred venue. If the status quo is a redirect, then RfD is the preferred venue. If a template were to be BLAR'd (current example), then TfD would be the preferred venue. -- Tavix ^(talk) 01:31, 9 October 2024 (UTC)[reply]

I'd support this; BLAR'd articles are generally low-profile enough that talk page discussions don't usually get sufficient participation to resolve the disagreement. Maybe a more generic reference to deletion discussions would resolve Tavix's concern. Extraordinary Writ (talk) 01:44, 9 October 2024 (UTC)[reply]

Substitute "Wikipedia:Articles for deletion" with "appropriate deletion discussion venue" and I would support. -- Tavix ^(talk) 01:49, 9 October 2024 (UTC)[reply]

We should avoid giving the impression (explicitly or implicitly) that RfD is the appropriate deletion venue (because it almost never is). Thryduulf (talk) 01:55, 9 October 2024 (UTC)[reply]

RfD is the appropriate venue way more often then you give it credit for. For example, RfD is the appropriate XfD for Grood even though it was BLAR'd back in 2005. -- Tavix ^(talk) 02:02, 9 October 2024 (UTC)[reply]

Grood is an example of something that would be better at AfD but is not completely implausible at RfD (the content would be an A7 speedy if it is topically in scope, I haven't checked). The length of time something has been a redirect is not relevant to almost anybody except you - what matters is the pre-BLAR content. Thryduulf (talk) 02:10, 9 October 2024 (UTC)[reply]

"The preferred venue for doing so is the appropriate deletion discussion venue" is singularly unhelpful; it's like saying "The preferred venue for doing so is in the right place to do it". The "deletion discussion venue for the page's stable version" gets the idea across, but it's awkward to the point of absurdity. —Cryptic 02:09, 9 October 2024 (UTC)[reply]

"The preferred venue for doing so is the appropriate deletion discussion venue for the pre-redirect content" is simple and unambiguous. Thryduulf (talk) 02:11, 9 October 2024 (UTC)[reply]

Including "for the pre-redirect content" would have major unintended consequences. RfD would have to reject all redirects that were previously articles, which is further than even you support. Hopefully you can agree than even if you would prefer that Grood be at AfD, it is something that RfD is capable of handling. -- Tavix ^(talk) 02:21, 9 October 2024 (UTC)[reply]

We're talking about the preferred venue for the majority of cases, not the only venue for all cases. RfD should reject the majority of undiscussed BLARs brought there, but that doesn't mean it has to reject all nominations of undiscussed BLARs and it definitely doesn't mean sending them to RfD in the first place should be encouraged. Thryduulf (talk) 02:27, 9 October 2024 (UTC)[reply]

How about "In most cases, the preferred venue for doing so is the deletion discussion venue appropriate to the pre-redirect content, although a discussion on the talk page (or at an appropriate WikiProject) can sometimes resolve the disagreement, especially if the redirection is recent. Exceptionally, for example if the pre-redirect content would be subject to speedy deletion if restored, it can be nominated at RfD." Covers all the bases with appropriate weighting, but it is a lot more complicated and I'm not sure that either the parenthetical or second sentence are necessary. Thryduulf (talk) 02:34, 9 October 2024 (UTC)[reply]

RfD should be the preferred venue for pages like Grood. Redirects that contain article content from Wikipedia's wild west days constitute the vast majority of BLAR's that get sent to RfD, and they're usually the ones that get uncontroversially deleted. It's silly to restore article content from a long-standing redirect just because it was an article for a couple months back in 2005 that was redirected without any fuss. -- Tavix ^(talk) 02:39, 9 October 2024 (UTC)[reply]

We've already established that Grood is one of the exceptions. It can't be both an exception and an example of the vast majority. Once again, the age of the content is irrelevant. Thryduulf (talk) 02:45, 9 October 2024 (UTC)[reply]

No, Grood is not "one of the exceptions". My point is that standards for articles were much different in Wikipedia's early days than it is now. As Wikipedia's standards increased, the easiest way to handle them was to BLAR them and that's usually the kind of redirects that get nominated at RfD with with article content in their histories. -- Tavix ^(talk) 02:53, 9 October 2024 (UTC)[reply]

How about: "The appropriate deletion discussion venue is preferable, though sometimes the dispute may be resolved on the article's talk page." -- Tavix ^(talk) 02:26, 9 October 2024 (UTC)[reply]

That's saying "The preferred venue for doing so is in the right place to do it, but sometimes it isn't" which is even worse than your previous suggestion. Thryduulf (talk) 02:28, 9 October 2024 (UTC)[reply]

No, it says that XfD is preferable to the talk page. For what it's worth, I still think the current wording is best. -- Tavix ^(talk) 02:41, 9 October 2024 (UTC)[reply]

Ok, being more precise you suggestion says "the preferred venue for doing so is the right deletion venue for doing it, but sometimes the talk page is the right venue for doing it", which is a minor improvement to the status quo, but other suggestions are significantly better still. Thryduulf (talk) 02:51, 9 October 2024 (UTC)[reply]

(edit conflict) I support the proposed change. The relevant XfD for the pre-redirect content simply is the most appropriate venue in the majority of cases (AfD for BLARed articles, TfD BLARed templates, etc), RfD is almost never appropriate and even when it is the other XfD is very nearly always equally appropriate. Talk page and/or WikiProject discussion are fine in some situations and shouldn't be excluded, but they are the minority. Thryduulf (talk) 01:54, 9 October 2024 (UTC)[reply]

on second thought, forget everything i said about my suggestion at wikipedia talk:redirect being a maybe, both discussions have convinced me that naming more than one xfd venue is necessary if any venue is to be named in the first place. the idea that other venues just can't is at best an assumption of incompetence, and at worst an easily avoidable misinterpretation of both the 2018 and 2021 consensus(es) cogsan _{(nag me) (stalk me)} 12:01, 9 October 2024 (UTC)[reply]