Wikipedia talk:New pages patrol/Reviewers/Redirect autopatrol

Preliminary discussion

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

I've opened the RfC. --DannyS712 (talk) 01:29, 2 December 2019 (UTC)

---

Before I officially open this proposal, is there anything that should be changed? --DannyS712 (talk) 06:32, 13 November 2019 (UTC)

Please do not open this RfC at the VP. The VP is for notifying of RfCs, although from the number of RfC that get started there by newbies, one could forgive the Community for thinking the VP is the place. The RfC should be on a sub page of its parent project. Kudpung กุดผึ้ง (talk) 11:25, 13 November 2019 (UTC)

Please take the Neelix episode into consideration. Some users here might not remember it.Kudpung กุดผึ้ง (talk) 11:32, 13 November 2019 (UTC)

@Kudpung: I've re-read through the Neelix discussions. What, if anything, would you suggest be done? I was planning to continue logging on-wiki in text form (in addition to the actual patrol logs), but is there anything else? DannyS712 (talk) 06:17, 16 November 2019 (UTC)

I haven't actually given much thought to the technicalities. Getting a new user group created is not easy at all. Many editors oppose based purely on bureaucracy-creep without even considering what is being proposed, while others support because they like collecting hats. Although I realise that redirects need to be patrolled because a lot of superfluous and nonsense redirects are indeed created, I don't believe that they are toxic in the way articles can be. There is something wrong with a significant number redirects - however justified the creation of a redirect may be, the creators often do not follow the guidelines at WP:R properly(if they even botther tomread them) or do not add the correct templates to them. Patrolling an article releases it to to be indexed and found in search engines, and I'm sure that plausible redirects need to be searchable too. However, my main personal concern is that redirects can be turned back into articles for a number of illicit purposes, but that's another story. So I'm sorry if I can't be of more help. Have you asked anyone else to review this RfC draft? I suggest getting some feedback from users such as Barkeep49, Insertcleverphrasehere, DGG, Ymblanter, MER-C, Scope creep, Elmidae, SD0001, Scottywong, Rosguill, Onel5969, PamD, and a couple of users who were major contributors to the Neelix discussions. Kudpung กุดผึ้ง (talk) 01:56, 23 November 2019 (UTC)

I am in favor of this generally. While Neelix was an unfortunate incident they already had true autopatrol so I'm not sure how this would make Neelix 2 more likely. Personally speaking I find redirects to be much less of a priority for patrolling than articles and so methods that can reduce the number of redirects that "must" be patrolled strikes me as a good thing. Best, Barkeep49 (talk) 02:06, 23 November 2019 (UTC)

@Kudpung: Note that converting a redirect into an article adds it to the feed again - see phab:T223828 DannyS712 (talk) 02:14, 23 November 2019 (UTC) (edited --DannyS712 (talk) 02:22, 23 November 2019 (UTC))

I am critically aware of that, but there are nevertheless ways in which illicit conversions can be missed by patrollers. There is precisely a relevent discussion at WT:NPR right now. I also wrote an article in The Signpost last year on black hat UPE Kudpung กุดผึ้ง (talk) 02:19, 23 November 2019 (UTC)

I think the key problem is redirects turned into articles. We need some way of getting these included somewhere (actually, I think there's an edit filter for these, and if not, it would be easy to make one) . Otherwise, I'm not concerned about redirects. I think an additional processfor these would be unconstructive--it would take away from the concentration of effort we need to give new drafts and articles. We must keep from getting further behind for the partthat relaly matters, and every person switching to check redirects is one fewer to check articles. DGG ( talk ) 06:52, 23 November 2019 (UTC)

@DGG: as noted above, redirects turned into articles are added to the queue the same way new articles are - see phab:T223828 --DannyS712 (talk) 08:53, 23 November 2019 (UTC)

yes, I know that, because I was one of the people who asked for this in the first place. We neeed to label them, tho, or give some indication, so they can be spotted more specifically DGG ( talk ) 15:48, 23 November 2019 (UTC)

I added a note at the top of the proposal --DannyS712 (talk) 01:26, 2 December 2019 (UTC)

Also in support of this, as I can't see any negatives. There are several NPPers who appear to spend a lot of time patrolling just the redirects - if anything, this will lift some load off them. While it's true that proposals like this tend to get opposed purely because of the bureaucracy-creep argument, I don't think that would be the case here as most of NPP (appears to be) in favour of this. I agree with Barkeep on the Neelix incident. We shouldn't hold off trusting hundreds of users just because one of them will ultimately break the trust. SD0001 (talk) 13:20, 23 November 2019 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

RfC on autopatrolling redirects

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Should the proposal at Wikipedia:New pages patrol/Reviewers/Redirect autopatrol be adopted? --DannyS712 (talk) 01:29, 2 December 2019 (UTC)

Survey

• Support I have made a habit of patrolling the back end of the backlog of newly created redirects on a daily basis. A majority of uncontroversial redirects are created by a small group of users, many of whom focus on a particular "genre" of redirect (e.g. redirects from songs to albums and artists, redirects from birth names to biographies). Autopatrolling these redirects and removing them from the queue would significantly reduce the amount of work needed to patrol the remaining redirects that actually have a chance of being controversial or inappropriate. _signed, Rosguill ^talk 01:39, 2 December 2019 (UTC)
• Support - sounds like a good plan to me. Masum Reza^📞 01:55, 2 December 2019 (UTC)
• Support Don't see any downside to this, presuming that admins are clueful about who they add to the list. Headbomb {t · c · p · b} 01:59, 2 December 2019 (UTC)
• Support I have no problem with this in principle. However I think we should offer this pseudo-permission with minimal bureaucracy. Having requests for it at WT:NPP/R seems fine, or anywhere that's not WP:PERM. Ideally most additions to the "check page" will be suggested by new page reviewers, since they are the only ones effected. WP:PERM/A has proven enticing to hat-collectors, with requests commonly declined with "this right won't assist you as an editor". I think if redirect-autopatrol is too public-facing you'll see many such requests, especially given it has a lower bar. Finally, a bot of this nature needs to be ran automatically via cron, which I think the operator is capable of using now :) — MusikAnimal ^talk 02:01, 2 December 2019 (UTC)
• Support - Something needs to be done to limit the mountain of redirects that pour in. This seems like a reasonable solution with minimal risk of possible abuse. MusikAnimal's suggestions are good ones. — Insertcleverphrasehere ^{(or here)}_{(click me!)} 02:14, 2 December 2019 (UTC)
• Support I'm in favor of less human time being spent on redirects. I also support Musik's idea of doing this in NPP space. Best, Barkeep49 (talk) 02:49, 2 December 2019 (UTC)
• support seems logical--Ozzie10aaaa (talk) 03:17, 2 December 2019 (UTC)
• Support Makes sense; an editor can be trusted to create redirects unpatrolled without being trusted to create articles unpatrolled . * Pppery * _{it has begun...} 03:42, 2 December 2019 (UTC)
• Support , as part of NPP/R. My earlier doubts about this have been satisfied DGG ( talk ) 04:15, 2 December 2019 (UTC)
• Support - time saver. ^Atsme _Talk 📧 04:27, 2 December 2019 (UTC)
• Support — If it gets rid of the huge backlog at NPP. ~~ CAPTAIN MEDUSA^talk 06:27, 2 December 2019 (UTC)
• Support. It will save time, especially since creation and review of articles is a much more sophisticated process than that of redirects, and allow reviewers to concentrate on assessing content that is more likely to be problematic. ComplexRational (talk) 20:45, 2 December 2019 (UTC)
• Support – makes sense. – Levivich 23:07, 2 December 2019 (UTC)
• Support. and I would like to have this. I create a ton of redirects. –MJL ‐Talk‐^☖ 03:41, 3 December 2019 (UTC)
• Support. Seems like a reasonable idea. · · · Peter Southwood ^(talk): 07:48, 3 December 2019 (UTC)
• Support in furtherance of backlog reduction. I don't fully understand why it's being done as a pseudo-group instead of a real one, but not sure it matters.  — SMcCandlish ☏ ¢ 😼  09:11, 3 December 2019 (UTC)
• Support, but I don't see why requirements for regular Autopatrolled couldn't be adjusted to include prolific creators of good redirects. I was given Autopatrolled status well before I'd created 25 articles; somebody else requested it on my behalf because I'd be creating many redirects that need patrolling. I'd guess I'd created a good 10,0000 redirects before I hit the 25 article mark (at which point I'd been Autopatrolled for about two years). Plantdrew (talk) 17:46, 3 December 2019 (UTC)
• Support Sounds like a good idea to me. Lugnuts ^{Fire Walk with Me} 18:31, 3 December 2019 (UTC)
• Support Good precidents, good plan. ~Kvng (talk) 15:02, 4 December 2019 (UTC)
• Support Good plan. Taewangkorea (talk) 22:01, 5 December 2019 (UTC)
• Support will ease the burden on NPPs who patrol redirects and ease the NPP backlog. In my experience, something like 95% of new redirects created are non-problematic and useful additions to the encyclopedia. buidhe 22:53, 5 December 2019 (UTC)
• Support: Interesting solution to easily deal with at least a part of the new pages backlog. Good proposal. ~ ToBeFree (talk) 16:55, 6 December 2019 (UTC)
• Support: creating redirects is a task with low potential of causing significant problems. I'm happy for a pseudoright to be granted with fairly low barrier to entry. — Bilorv (talk) 15:34, 7 December 2019 (UTC)

Discussion

• Would this pseudo-group (automatically) include editors like myself – of course subject to discretion – who are new page reviewers but neither have nor need autopatrolled for not being prolific article creators? I must comment that not doing so might put us in a rather awkward position: if one criterion to add users to this pseudo-group is discussion among reviewers (point 2 of adding users), would reviewers self-nominate, or assess each other on their track records (last bullet)? ComplexRational (talk) 20:55, 2 December 2019 (UTC)

  The second bullet describes how users are added, the fourth describes expectations for users. New page reviewers wouldn't be special in terms of fast-track or automatic membership - discussion among other reviewers or admin discretion is still needed to add the user --DannyS712 (talk) 21:04, 2 December 2019 (UTC)

I would think that we would handle this in a similar manner to autopatrol where being a reviewer would be a good indicator of competency in asking for the psuedo right but that we wouldn't necessarily automatically include every reviewer. Of note is that anyone (like you) with the autopatrol right won't need access to this pseudogroup as redirects are already patrolled automatically. Similarly, the backend software won't change so for reviewers without autopatrol if they create a redirect the curation toolbar would not let them review their own creation. If that reviewer were in the psuedo group the bot would be able to come along and patrol it though. Does that all make sense? Best, Barkeep49 (talk) 21:20, 2 December 2019 (UTC)

Yes, that makes sense, and should ensure a balance between efficiency and making sure nobody gets a free pass. Thank you, DannyS712 and Barkeep49, for clarifying. (BTW, I do not have the autopatrol right; that is what led me to ask.) ComplexRational (talk) 21:25, 2 December 2019 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Redirect autopatrol: implementation

I have a request: The process of according the right at PERM should please be as automated as possible for the admins who work at PERM. I.e: adding the the user to the white list and whatever else is done to make it work. For example, AWB is too complicated, the script only directs the admin to the white list page and as a consequence AWB requests often have up to a 14 day backlog. Kudpung กุดผึ้ง (talk) 07:40, 8 December 2019 (UTC)

@Kudpung: The bot side was just approved at Wikipedia:Bots/Requests for approval/DannyS712 bot III 66, and for granting, this isn't done at perm. Just add the user to Wikipedia:New pages patrol/Redirect whitelist (view source to see html inline instructions). As for how easy it'll be, I'm not an admin, so I can't tell, but it should be fairly straightforward (just add * Foo to the whitelist page) DannyS712 (talk) 02:00, 11 December 2019 (UTC)