Wikipedia talk:Criteria for speedy deletion

 Request withdrawn Apparently, there are cases where a user will say up-front that the content has been put in the main encyclopedia, and it turns out, it was put in by another editor, making it necessary to preserve attribution. Even though the particular case mentioned below[1] was a multi-author page which the proposed bot would ignore, it could just as easily have been a single-author page. davidwr/(talk)/(contribs) 21:28, 26 November 2020 (UTC)[reply]

Many {{

As a simple example, a bot that would auto-delete a page in user space in which ALL of these apply:

• only editor is "the user" in question (I'd want a human to review multi-contributor pages)
• page is a sub-page of User:username/ (not "User talk:", that could be abused to delete user talk page, not main user page, I'd want a human to read it first)
• No move history at all (for keeping the bot simple)
• Short waiting period (1 hour?) to allow for self-reverts ("oops, I didn't mean to do that") or 3rd-party reverts (mentors, education project coaches, or just someone who wants to force this to be looked at by a human)

I'm keeping this simple for now. There are no doubt other "obvious" cases of db-user and db-author that should be "routine tasks that any user can do without an administrator's help." That's what admin-bots are for, to do the "no brainer" tasks so admins can do the things that require human judgement.

I'm asking administrators - would this be worth doing? Would you support this if went to the appropriate channels and asked for a bot to be written? For any bot-writers out there, does this look like something that could be written up and maintained without much effort? davidwr/(talk)/(contribs) 14:10, 26 November 2020 (UTC)[reply]

What problem is this attempting to solve? Do we usually have a backlog of U1 requests?

ping on reply)[reply

]

It's low-hanging fruit and will free up administrators to do other things. That's what bots are for, the "no thought required" things. davidwr/(talk)/(contribs) 16:47, 26 November 2020 (UTC)[reply]

Fair enough. As to the second part of your question, yes, as far as I'm aware this is something that could be written up to meet the specifications to which you describe. Primefac (talk) 16:50, 26 November 2020 (UTC)[reply]

• My impression has been that G7 tends to be seen as an "easy" criterion, and so pages tend to get deleted very quickly, presumably without the sort of checks you would expect to see for more "serious" criteria, like G4 or G11. In these circumstances, a carefully designed bot will actually be an improvement, in term of due diligence, over manual action. – Uanfala (talk) 14:59, 26 November 2020 (UTC)[reply]

  For what it's worth, db-user is

  WP:U1, not G7. Primefac (talk) 15:11, 26 November 2020 (UTC)[reply

  ]

  Ah, my bad. But otherwise, I've got the same experience with U1 as with G7. – Uanfala (talk) 15:19, 26 November 2020 (UTC)[reply]

• Probably a bit cart-before-horse at this point in time, but any bot tasked with this should be checking what links here for, and not acting, if the target page is transcluded/linked from {{Merged-from}}, {{Merged-to}}, {{Copied}}, {{Split article}}, etc. – or possibly, since I'm sure I've missed other relevant templates, hand made notes of attribution, and more I cannot think of, not acting if the target page is linked from the main talk namespace.--Fuhghettaboutit (talk) 17:34, 26 November 2020 (UTC)[reply]

  It goes without saying that the bot should do all the same checks a human would do, or maybe a "superset" of those checks if that makes coding easier. For example "if linked to from anywhere, leave it for a human" would be a much easier check to do. davidwr/(talk)/(contribs) 17:48, 26 November 2020 (UTC)[reply]

• That seems fair enough. U1 deletions do tend to be processed fairly quickly, because they are very easy to do, but there are quite a few of them and a bot would free up some admin time for other things. I'm not sure that merges are likely to be a problem in userspace, if user A writes something in their sandbox, copies it to a mainspace page and then tags the sandbox for deletion under U1 then the edit history of the sandbox doesn't need to be kept for attribution reasons because the content is attributed to user A in mainspace. Hut 8.5 18:04, 26 November 2020 (UTC)[reply]

  @Davidwr: Agreed. That approach is safest. @Hut 8.5: Since this does present the occasional copyright problem, a bot must take the most cautious approach. Your scenario is absolutely correct; that presents no copyright issue. The attribution problem arises when the content (that meets threshold of originality) is taken and used by a different user. I've seen this not just in the mainspace, but in drafts. (Actually, where I've seen this play out most often is: User A copies over content from their dedicated user-subpage containing an article draft; that content was edited in substantive way by user B (or B, C and D); A tags with U1–but that won't be a problem here because the bot is already proposed to only act where the target has only edits by User A).--Fuhghettaboutit (talk) 18:51, 26 November 2020 (UTC)[reply]

  OK - it didn't seem very likely that the content would be copied into mainspace by someone else only for the owner to tag it for U1, but if you've seen that happen then fair enough. I don't think "linked to from anywhere" is a suitable standard though because there are various pages which list all articles with CSD tags on them. You would have to either filter those out, look for certain kinds of links only, or restrict to certain namespaces. Hut 8.5 19:03, 26 November 2020 (UTC)[reply]

  Indeed: it does happen:-)--Fuhghettaboutit (talk) 20:38, 26 November 2020 (UTC)[reply]

• I use db-author a lot and those pages are almost always deleted by @Fastily: so pinging him for his opinion on it. Schazjmd (talk) 18:14, 26 November 2020 (UTC)[reply]

  @

  criterion G7. --Redrose64 🌹 (talk) 23:47, 26 November 2020 (UTC)[reply

  ]

  Oops, thanks for the correction. Schazjmd (talk) 23:53, 26 November 2020 (UTC)[reply]

• We used to have a bot that automatically handled obvious u1/g7 deletion, until the operator was the subject of an ArbCom case (not related to the bot) and desysoped. -FASTILY 02:42, 27 November 2020 (UTC)[reply]

I added some "things to consider" for U2 ([2]) - nonexistent user, and U5 ([3]) - NOTWEBHOST. I think both reflect "common sense."

Since these are recommendations that can be "considered and ignored" by editors and administrators without causing someone to scream "you aren't allowed to delete the page because CSD says such and so," I went ahead and made the changes

Seventyfiveyears, redirects from mainspace to the draft namespace are eligible for R2. What are you proposing to change? – bradv🍁 20:10, 30 November 2020 (UTC)[reply]

We certainly wouldn't want to say that redirects from any namespace to draft should be deleted, as user sandboxes are regularly moved to draftspace as part of the AfC process. But also, redirects in other namespaces really aren't a problem – the point of R2 is to keep this stuff out of mainspace. – bradv🍁 20:15, 30 November 2020 (UTC)[reply]

Except for user sandboxes, according to "WP:Cross-namespace redirects" administrators would also delete redirects from Portal: spaces to Draft: spaces (even when they are caused by moves). Seventyfiveyears (talk) 20:21, 30 November 2020 (UTC)[reply]

That's

WP:G6: accidental creation in the wrong namespace, i.e. routine housekeeping. Primefac (talk) 20:59, 30 November 2020 (UTC)[reply

]

Per

WP:NEWCSD There are four requirements for new and expanded criteria: Objective, Uncontestable, Frequent, Nonredundant. I think what you are proposing is a criteria to speedy-delete cross-namespace redirects to draftspace from any other namespace. That's certainly objective, no problem there. As originally proposed it's mostly nonredundant (with the exception of redirects from the main namespace and redirects created by accidental moves) but that's not uncontestable though - moves of userspace drafts are just one example of something that could be deleted but should not be (you could exclude that, but you need to define your exception objectively). Is every other example something that always should be speedily deleted? I haven't looked, but you should have done this before making your proposal - please can you share your results? As for frequency - how many of these redirects are there? How often do they get created? How many of them are not already speedy deletable under the existing R2 or G6? Thryduulf (talk) 23:17, 30 November 2020 (UTC)[reply

]

It's occasionally been said that CSD#G6 is a catch all that is easily abused. Is this an example? --SmokeyJoe (talk) 00:09, 7 December 2020 (UTC)[reply]

G6 is for uncontroversial maintenance - no more, no less. It is a catch-all, and that's a good thing, but it's also one that should never be used if there's a reasonable objection that could be made. I can totally see that "deleting clearly unhelpful GA reviews" is potentially uncontroversial in that project's context, and if it's genuinely uncontroversial outside of people demanding process for process's sake, then I wouldn't have any particular objection to it. This is probably an example of a poor use of it: it was a low-quality GA review, but well above the threshold of "unambiguously unhelpful". :Ultimately, I'm not seeing a systematic problem here - if a terrible GA review does get shuffled off because it's clearly uncontroversial, then that's a valid G6 to me, but if there's uncertainty, it ending up at MfD is correct. If there's a pattern of certain GA-active admins aggressively deleting stuff in this way, this might be something worth clarifying, but this particular incident doesn't look like one. ~ mazca ^talk 00:32, 7 December 2020 (UTC)[reply]

WP:SPI. Sometimes there's a case opened which is clearly not worth archiving. These routinely get deleted under G6. For example, Wikipedia:Sockpuppet investigations/73.97.254.42. Deleting obviously bogus GA reviews (i.e. Talk:Sarah Cooper/GA1) seems like the same concept. -- RoySmith (talk) 01:05, 7 December 2020 (UTC)[reply

]

Roy, I have long wondered whether that is the case. Bogus SPI and LTA cases disappear, and only weird cases come to MfD, nominated by non-SPI people. I think these deletions are decidely contrary to the wording, spirit and intention of G6. These deletions are being used to for tidying, quiet suppression of trolls, arguably good deletions, but they are not G6. SPI is probably quite competent handling their own deletions of subpages of WP:SPI, and LTA, but they should not be doing it under G6. SPI freedom to delete outside of deletion policy would appear to be providing an example to other groups that they too may delete outside of deletion policy. I see good reason to create a CSD criterion to cover SPI cases, but I do not see a good reason for GAN admins to have near-arbitrary authority to speedy delete substandard GA reviews. Deleting obviously bogus GA reviews is the same concept, that concept being that some groups of people are not bound by deletion policy. --SmokeyJoe (talk) 01:21, 7 December 2020 (UTC)[reply]

SmokeyJoe, Um, people actually nominate SPIs for deletion at MfD? I'll get the popcorn. -- RoySmith (talk) 01:47, 7 December 2020 (UTC)[reply]

On occasion, yes. Usually, ill-advised. --SmokeyJoe (talk) 02:03, 7 December 2020 (UTC)[reply]

SmokeyJoe, Hmmm. I found Wikipedia:Miscellany for deletion/Wikipedia:Sockpuppet investigations/Danielipforsecretary, but that's the only one I can see. -- RoySmith (talk) 02:08, 7 December 2020 (UTC)[reply]

[4] Random MfDs are not the problem or the point. The problem is Speedy Deletion "per G6" for cases where WP:CSD is lacking a line, and for this setting precedent for others to delete broadly "per G6". I think G6 should never be used for a page with a non-trivial history. --SmokeyJoe (talk) 02:24, 7 December 2020 (UTC)[reply]

SmokeyJoe, G6 is also routinely used to delete talk pages of deleted pages, regardless of the history of the talk page. -- RoySmith (talk) 02:27, 7 December 2020 (UTC)[reply]

That would be a mistake.

WP:CSD#G8 should be used. --SmokeyJoe (talk) 02:28, 7 December 2020 (UTC)[reply

]

SmokeyJoe, Ah, my mistake on the talk page thing. You are correct on that. -- RoySmith (talk) 02:31, 7 December 2020 (UTC)[reply]

• I think these deletions of pages with non-trivial histories, including signed comments by other editors, should be invalid under G6.

Past practice, eg

CSD G6, uncontroversial maintenance for SPI. T. Canens (talk

) 22:24, 18 October 2010 (UTC)

implies that these deletions have consensus in practice, and so to make WP:CSD catch up with practice, "Checkuser or SPI clerk maintenance of SPI subpages should be assigned its own CSD criterion.

I am not sure that "uncontroversial maintenance for GAN" would be as easily justified. --SmokeyJoe (talk) 02:39, 7 December 2020 (UTC)[reply]

To note I and other clerks have deleted IP only sockpuppet investigations under G6 in lieu of archiving the case. These IP only cases have always (from my understanding) been deleted when there are no other archived cases for this case page and where its not a deliberate sockpuppetry by the person behind the IPs (i.e.

dynamic IP addresses being dynamic). Dreamy Jazz ^{talk to me | my contributions} 12:57, 11 December 2020 (UTC)[reply

]

That sounds sensible. WP:CSD should catch up with accepted practice. A new CSD criterion for SPI business. —SmokeyJoe (talk) 13:00, 11 December 2020 (UTC)[reply]

Also I more often delete empty sockpuppet categories under G6 too. The idea for both is that they both are uncontroversial actions. For the deleting of case pages, if it meets another criteria (like due to G5, G10 etc.) I will delete under that criteria instead. Sockpuppet categories are deleted early because they are only filled when they have tagged userpages. If the tags have been removed / changed such that the category is empty it is very unlikely that the category will be re-filled with socks.

However, creating a separate criterion for SPI deletion of cases and/or sockpuppet categories seems unnecessary. G6 works well for cases where the deletion is uncontroversial and does not meet any other criteria. If the deletion is questioned and there is good reason for the question, then the case request should be restored as its not controversial. Such a criteria would still be used, but I just think as long as G6 can cover the cases where no other criteria apply it should be fine without an extra one. Dreamy Jazz ^{talk to me | my contributions} 13:09, 11 December 2020 (UTC)[reply]

G6 should not be used to delete others’ non-trivial contributions. It is an abuse of G6. SPI doing it has provided poor example for admins in other areas where they too take a liberal interpretation. —SmokeyJoe (talk) 13:24, 11 December 2020 (UTC)[reply]

I referred Talk:Sarah Cooper/GA1 to MfD because I felt it wasn't an obvious G6. Given the number of conflicting opinions above, I think there should be an RfC to determine whether such pages are actually eligible for G6. -FASTILY 00:19, 10 December 2020 (UTC)[reply]

Frequently in the draftspace (less frequently in articlespace) we see pages created with an incorrect title when the correct title in that namespace is

I would be opposed to it being a G criteria but could see it, perhaps, as an A criteria, in other words only in article space. One effective way to convince an administrator to unsalt would be to have a promising draft. Best, Barkeep49 (talk) 00:08, 13 December 2020 (UTC)[reply]

Salting is largely a legacy feature. Typically, some incident of spam or vandalism from 15 years ago would have rendered a given title protected for perpetuity, and when anyone, at whatever point in the near or distant future, tries to create anything at this title (whether it be an article about an unrelated subject with the same name, or a redirect, or a dab page), they will have to jump through completely unnecessary hoops. I don't think the solution here is to add further hoops to jump through. If something truly terrible has been created, then there will almost always be a relevant CSD. The problem with Draft:Ramy Khodeir (2) is not that it was created at a title similar to one previously salted, the problem is that it's spam. – Uanfala (talk) 00:19, 13 December 2020 (UTC)[reply]

Redundant. Either the new page is already deleteable for the same reason it would have been at the correct title, or the salting was incorrect. —

Cryptic 00:21, 13 December 2020 (UTC)[reply

]

and B-I-N-G-O was his name-o. If another criterion applies (remembering, especially G4 exists if it wasn't speedily deleted the first time), then it wouldn't be necessary. If nothing applies, then not unSALTing and moving the page to the correct place is just a BATTLEGROUND mentality causing you to pointless BITE the new user unfamiliar with Wikipedia bureaucracy and accelerate declining participation. WilyD 05:47, 15 December 2020 (UTC)[reply]

• Stop overquick SALTing of draftspace titles. There is SALT policy documented at
  WP:RfPP, that seems to be overlooked unilaterally by admins annoyed by draftspace recreations. As in mainspace, SALTing too easily has the side effect of re-creations under an variant title, which is a bigger problem. Wait for repeated recreations be multiple accounts before resorting to SALT. —SmokeyJoe (talk) 05:18, 13 December 2020 (UTC)[reply
  ]