This page is for New Page Reviewers to discuss the process with each other and to ask for and provide help to fellow reviewers. Discussion also takes place on our Discord server (invite link) For discussions on other matters, such as bugs, etc., please navigate through the tabs, or go to the discussion pages of the relevant policies. For discussion on topics purely relevant to coordination tasks, such as for example - but not only - Backlog Drives, etc., please post at Coordination Talk

Top New Page Reviewers database report (updated by bot 2x daily)

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

This page has archives. Sections older than 30 days may be automatically archived by Lowercase sigmabot III when more than 8 sections are present.

Archives of Page Curation: 1, 2, 3, 4. 5, 6, 7, 8, 9

This page has archives. Sections older than 30 days may be automatically archived by Lowercase sigmabot III when more than 8 sections are present.

From a quick eyeball of the leaderboard it looks like we’ve patrolled 9000 or so articles in November, but over the same period, a nearly equal number of new articles has been created. That’s a lot of paddling for not much progress…… Mccapra (talk) 19:12, 28 November 2021 (UTC)[reply]

Yeah. A bit unexpected considering how successful the AFC backlog drive was. It looks like losing some of our high volume reviewers a few months ago really lowered our velocity. –Novem Linguae (talk) 23:18, 28 November 2021 (UTC)[reply]

on the other hand, without the drive we might be about to hit 18,000 in the queue! Mccapra (talk) 23:35, 28 November 2021 (UTC)[reply]

I know I knocked off some 300 articles more than I would have otherwise. It's not a lot compared to other reviewers but that's 300 articles that aren't in the queue. (t · c) buidhe 23:38, 28 November 2021 (UTC)[reply]

Sounds like a lot....I did 22 :-) North8000 (talk) 01:48, 29 November 2021 (UTC)[reply]

I use the New Pages Feed on the "show me the oldest" setting so that I can patrol kind of slowly and I have seen a lot of what feels like bulk redirect-removals of things like Mad Men episodes and Woody Woodpecker Episodes that used to be all on one page when the individual shows were deemed non-noteworthy but someone is maybe trying again. Do we have any stats on how much of the backlog is newly created pages and how much is older pages that for whatever reason are showing up as unpatrolled (like these old episode articles)? Jessamyn (talk) 04:23, 29 November 2021 (UTC)[reply]

I tend to patrol from the latest articles and mainly check copyvio and notability. What surprises me is how few articles need deletion, at AfC the ratios are reversed. I guess it works at keeping a lot of the crap out of mainspace to begin with. (t · c) buidhe 04:48, 29 November 2021 (UTC)[reply]

That’s interesting if you’re checking thoroughly and systematically. Apart from CSD junk I also don’t find many that clearly and unambiguously need to be deleted. However there is an enormous number where I either have no idea (sports) or where it’s so much work to find out (53 refs in Korean) that I tend to just leave them in the queue. For this reason I suspect there are several thousand deletion candidates in the queue that no reviewer has yet wanted to tackle. Mccapra (talk) 08:14, 29 November 2021 (UTC)[reply]

I used to always work from the back of the queue. I've started using the sorted feed at: User:SDZeroBot/NPP_sorting. It would be useful if we could recruit some NPPs interested in sports and other subjects massively contributing to the backlog, as like you I tend to leave sports articles. Polyamorph (talk) 08:29, 29 November 2021 (UTC)[reply]

I've been spending a bit of time in the sports section during the backlog drive. If I'm being blunt, a lot of articles are low-quality: poorly sourced, only primary sources, dubious notability etc. If I was honest, I would PROD, AfD or draftify a large number of them. I'm worried about being perceived as too agressive though, so until now I've been mostly looking for articles I could mark as reviewed. Any feedback is appreciated. JBchrch talk 16:22, 29 November 2021 (UTC)[reply]

There was a slew of American football stubs that were created recently. Some of them were clear SNG passes, as they had played in the NFL (and were pretty quickly reviewed). Others, of university amateurs, are not. Given how many were created by the same editor, I too am wary of mass-prodding them, potentially overloading their talk page. Sdrqaz (talk) 16:29, 29 November 2021 (UTC)[reply]

Yeah, I don't check over sports articles. It's likely that different topics have different average quality. With sports some editors will argue for keeping at AfD even in absence of GNG coverage so it's tricky. (t · c) buidhe 04:44, 1 December 2021 (UTC)[reply]

I think that another type that are very difficult and where we need to connect a specialist with is where the references are not English language. North8000 (talk) 14:31, 29 November 2021 (UTC)[reply]

Thanks so much to everyone who participated and expect barnstars to be out in the next few days. My only question is how can I access the database cited by MusikBot's NPP chart? I'd like to figure out exactly how much we were able to cut the backlog during the drive. (t · c) buidhe 04:19, 1 December 2021 (UTC)[reply]

My thanks as well to everyone who participated, and especially coordinators, @Elli, Buidhe, and Tol:. I would be interested to see figures on exactly how many more articles than usual were reviewed in the drive. Regards! Usedtobecool ☎️ 15:17, 6 December 2021 (UTC)[reply]

I followed the November backlog drive and noticed that articles created from pages which were started as redirects in previous years come at the start of the Special:NewPagesFeed, for example Welsh republicanism was created as a redirect on 10 May 2006‎, published as an article on 23 November 2021‎ and reviewed on 27 November 2021 (4 days after publication). As a result there are still 662 unreviewed articles in the date range From 23 June 2021 to 31 July 2021, which would have been eligible for indexing by Google at the start of the drive. Would it be possible to prioritise these for review in December? I have to declare an interest as I published an article in late July, but I am not asking for special treatment for it.

buidhe, could you consider mentioning unreviewed articles from June and July in your closing report for the backlog drive? TSventon (talk) 10:59, 3 December 2021 (UTC)[reply]

Reviewers should consider reviewing from the back of the queue, i.e. oldest first. Polyamorph (talk) 16:57, 3 December 2021 (UTC)[reply]

I think articles become automatically indexed by google after 90 days. So your July article should already be indexed, just not formally reviewed. Try searching for it in Google to confirm. Reviewers use various methodologies to pick what they review. Some folks are rockstars and focus on the harder articles at the back of the queue, some folks work the front of the queue, and some folks use topic lists that have them reviewing a variety of date ranges. –Novem Linguae (talk) 07:50, 4 December 2021 (UTC)[reply]

Novem Linguae, Google did index my article shortly after 90 days and I obviously don't think it is a danger to the encyclopedia. And thank you to project members for your important work. TSventon (talk) 13:19, 6 December 2021 (UTC)[reply]

Sometimes when I use the curation tool to leave a message for the creator, it instead leaves a message on my talk page, addressing me as the creator. Is this a malfunction of the tool, or am I doing something wrong? North8000 (talk) 18:16, 3 December 2021 (UTC)[reply]

Could be a bug but more information is needed, under what circumstances does this occur? Are you redirecting the articles so that the system thinks you are the creator? Or is it random? Polyamorph (talk) 09:19, 5 December 2021 (UTC)[reply]

Thanks @Polyamorph:, your post made me take a closer look. For the last two times that occurred, ([1],[2]) I didn't redirect the articles, but they were both articles which were originally created as a redirect. BTW I've been working the oldest end of the cue, which is mostly articles just converted from redirects. Your making me look has solidified something that I've been changing. I probably shouldn't use the "note to creator" tool because that isn't going to go the the person who recently created the article. Thanks. North8000 (talk) 14:30, 5 December 2021 (UTC)[reply]

I am patrolling for AfCs for new and upcoming films that fail to pass WP:NFF guidelines. Am I (as not an NPP reviewer) able to reject such submissions? If not, how should I tag an article to communicate to an NPP-reviewer that a particular article would fail NFF (because film hasn't been released and film production hasn't been notable)? These prematurely created articles are either fancruft or operate as advertising for upcoming films, and are based solely on churnalism of press releases. Platonk (talk) 17:59, 4 December 2021 (UTC)[reply]

Platonk You should decline the article at AfC as a non-notable film. (t · c) buidhe 18:05, 4 December 2021 (UTC)[reply]

@Buidhe: I would love to, but I do not have any page reviewer privileges/rights and I don't see any button/link for me (a regular editor) to 'decline'. Article is Draft:Terror on the Prairie which was draftified per AfD WP:Articles for deletion/Terror on the Prairie. I don't, however, see any bold/obvious notation for an NPP-reviewer to see that it was draftified for failing NFF and to specifically look for whether it now passes. (In this case, it doesn't.)

I don't suppose I would mind becoming a page reviewer, but I never thought about it before. I checked the guidelines; the only experience I don't have is "moving pages", though I suppose I could acquire that. Platonk (talk) 18:27, 4 December 2021 (UTC)[reply]

Platonk If you're already patrolling drafts why not apply for the AfC perm? I think it's granted more easily than NPP and we can always use more reviewers! (t · c) buidhe 18:35, 4 December 2021 (UTC)[reply]

@Buidhe: Wow, more as yet undiscovered corners of Wikipedia! I'll look into that. I thought that was NPP's bailiwick. Platonk (talk) 18:54, 4 December 2021 (UTC)[reply]

One more question

I don't have 'Afc reviewer' rights, but have NPP rights. Can I accept/decline draft at Afc? --Gazal world (talk) 18:48, 4 December 2021 (UTC)[reply]

Gazal world No, you have to apply for AfC permissions separately. But they will almost certainly be granted if you are already reviewing at NPP and are a trusted editor. (t · c) buidhe 18:59, 4 December 2021 (UTC)[reply]

Not officially. You're supposed to apply for the right at WP:AFCP, get added to the list, then install and use the WP:AFCH helper script. –Novem Linguae (talk) 19:04, 4 December 2021 (UTC)[reply]

I see. 'll apply if needed. Thanks. --Gazal world (talk) 19:47, 4 December 2021 (UTC)[reply]

@Gazal world I went ahead and added you to the AFC list. Best, Barkeep49 (talk) 03:39, 5 December 2021 (UTC)[reply]

Thanks, Barkeep49. --Gazal world (talk) 09:25, 5 December 2021 (UTC)[reply]

The +sysop user group (aka Admins) is about to lose autopatrol out of WP:RFA2021/P. Admins will be able to grant it to themselves. But several have already indicated that they are likely to not grant it to themselves. As such don't be surprised when you start to see new articles created by admins appearing in the feeds. Best, Barkeep49 (talk) 03:37, 5 December 2021 (UTC)[reply]

good to know, thanks--Ozzie10aaaa (talk) 14:51, 5 December 2021 (UTC)[reply]

I have noticed a trend recently: When an new article is being moved to a draft because a patroller thinks it is not ready for the main space, the creator goes and movis it back without improving it. Today, I noticed that Khivabad and a few other articles of the same creator (and the articles, to be honest, are problematic, there are doubts as of whether these localities currently exist); the creator just move them back to the main space without any comment. Recently, I have seen Tammy Flores Garman Schoenen of another creator moved to the draft, the creator added a couple of sentences and moved it back. I am not sure it would survive an AfD. Should we do something about this practice?--Ymblanter (talk) 11:14, 6 December 2021 (UTC)[reply]

I believe there's nothing you can do about it because draftification (unless ordered by consensus at an AfD) is supposed to be voluntary. If the article content is very bad other solutions can be tried, such as AfD, stubbing etc. (t · c) buidhe 11:17, 6 December 2021 (UTC)[reply]

WP:DRAFTOBJECT doesn't really advise re-draftifying. I think we're supposed to AFD these if they're bad enough. –Novem Linguae (talk) 12:28, 6 December 2021 (UTC)[reply]

Once you draftify, even if they don't move it themselves, but simply object to the draftification, you must return it to mainspace. At that point, you can make any necessary edits (e.g. remove any and all uncited material, which then can't be re-added without valid citations as per WP:BURDEN, or add appropriate tags) or take it to AfD. I just had the same issue with a list of weapons article. It's aggravating and a headache, and imho shows a lack of intent to build an encyclopedia on the article's creator's behalf, but that's what it is. Onel5969 ^{TT me} 15:44, 6 December 2021 (UTC)[reply]

I'd like to think that AfC is not optional for paid editors and editors with a conflict of interest. For the others; yes, they can do that, but they ought to understand that draftification is an alternative to deletion: Either the article is improved or all non-verifiable content is removed. Vexations (talk) 21:14, 6 December 2021 (UTC)[reply]

If you find the right admins, you can get a paid/COI editor sanctioned for refusing to use AFC or edit requests. But community has rejected efforts to change the "should" to "must" in the COI editing guidelines. So, if such an editor reads the guidelines, says it only says "should", then takes you to ANI, what the outcome will be is a matter of who participate in the discussion. It's one of those areas where common sense and policy are at odds and community has deliberately left it that way. Usedtobecool ☎️ 08:48, 13 December 2021 (UTC)[reply]

I concur with others -- even for COI editors, AfD is the way to go here. Move-warring between draft- and article- space is counterproductive and not a good reflection on our project. Elli (talk | contribs) 21:44, 6 December 2021 (UTC)[reply]

WP:DBLDRAFT. Usedtobecool ☎️ 08:51, 13 December 2021 (UTC)[reply]

The last couple days I have been going through and patrolling a number of redirects (using a script I wrote which opens all visible links at Special:NewPagesFeed). One thing I've noticed is that there are a lot of unreviewed userpage redirects clogging up the queue. Currently, there are 941 of them between November 12 and now (I can't link to this view of the queue directly, but it is "unreviewed pages" in the "user" namespace). After going through about a hundred of these, it seems like the vast majority are simply left behind from users being renamed, users changing the names of their own subpages, and drafts being moved into draftspace/mainspace. Additionally, a very large number of them have ((R from move)) on them; this one is a good example. With this in mind, I was able to come up with a list of twenty (manually reviewed beforehand) and run a script on them to automatically mark the batch as patrolled. This took a little under a minute.

However, it seems to me that manual review of each redirect left behind when a user moves a draft into mainspace is quite unnecessary. I can't think of a reason why these need to be patrolled -- what would we be patrolling them for? If they are being created abusively (someone creating "User:Example/Gigantic stupid piece of shit" and redirecting it to a BLP's article), this seems like something that could be picked out at a glance from scrolling down a list of pages -- would there be any objection to me writing a script or bot to process these automatically, similar to DannyS712 bot III's Task 66 (approved to patrol redirects automatically following this RfC)? jp×g 02:34, 9 December 2021 (UTC)[reply]

• Potential scopes for autopatrolling include:

1. Only redirects created as part of a user renaming, moves within a user's own userspace, or a move to draft/mainspace page at the same title (i.e. "User:Example/John H. Smith" → "John H. Smith" would be automatically patrolled but "User:Example/John Smith" → "John H. Smith" would not)
2. All redirects within userspace targeted to draftspace or userspace
3. All redirects within userspace with ((R from move))

I don't know which of these would make the most sense -- would be interested in hearing some perspectives on this. jp×g 02:42, 9 December 2021 (UTC)[reply]

Also worth mentioning is the fact that there are currently 9,266 pages in the NPP queue, of which it appears that 7,070 are redirects; I am giving some consideration to ideas regarding autopatrolling redirects (or at least writing tools which make patrolling them less time-consuming). jp×g 02:45, 9 December 2021 (UTC)[reply]

@JPxG, you don't need to patrol any pages outside of the article space at all. Getting their numbers down doesn't help NPP goals, and they are not what we count under our backlog. At this moment, there are 9,164 non-reviewed non-redirect articles in mainspace. Those are our number one priority. Then there are 7050 more unreviewed redirects in mainspace which are not as urgent as the articles but have to eventually be looked anyway. The userspace stuff we can leave to be discovered in the course of normal editing and patrolling. Regards! Usedtobecool ☎️ 09:01, 13 December 2021 (UTC)[reply]

Honestly, I don't know how to do NPP for redirects; I'm also not sure why we do NPP for redirects. It is surely useful to have someone checking for Neelix-style bad redirects en masse, but clicking for every redirect feels like a bit much. That goes triple for userspace redirects; unless there's obscenity or BLP violations I'm not sure why anyone could possibly care about a redirect created by a user in their own userspace. User:力 (powera, π, ν) 02:50, 9 December 2021 (UTC)[reply]

力, I've been doing a day's worth of backlog pretty much daily for redirects for the past year or so and feel like I run into enough redirects that need correction or deletion for it to have been worthwhile (it also doesn't take too long, ~20 min a day). I agree about the non-mainspace redirects though, those are not worth our time. _signed, Rosguill ^talk 06:34, 9 December 2021 (UTC)[reply]

If you're finding several issues per day, and it only takes one person 20 minutes to clear the daily backlog, I stand corrected regarding mainspace. User:力 (powera, π, ν) 16:44, 9 December 2021 (UTC)[reply]

is now done. Some have given themselves the right again, others who had Autopatrol before they were admins have been given the right again. There are nevertheless some admins who have abused this right in the past. Not assuming bad faith, but pragmatism dictates that a watchfull eye on new articles is required. Kudpung กุดผึ้ง (talk) 11:30, 9 December 2021 (UTC)[reply]

@Kudpung, After the original announcement, my mind came to you. If I recollect correctly it appears some years ago you advocated for this and spoke extensively on why admins should not have automatic Autopatrol rights amongst other rights that are granted to sysops after they haves passed an RFA. If I’m reading all the entries correctly, it appears sysops can still give themselves the rights. Furthermore I agree that certain sysops have abused this right in the past. Celestina007 (talk) 09:02, 10 December 2021 (UTC)[reply]

Already 248 admins have either given themselves autopatrolled or had it assigned to them because they had it before, which is about a quarter of all admins and just over half of active admins, so unfortunately it seems the RfC will have a limited effect in practice. – Joe (talk) 11:56, 9 December 2021 (UTC)[reply]

I don't understand why they think they need it so badly. Polyamorph (talk) 13:22, 9 December 2021 (UTC)[reply]

Do you really want to patrol all my new articles? I am not a high-volume creator, but I still make about a dozen per week.--Ymblanter (talk) 13:29, 9 December 2021 (UTC)[reply]

Autopatrolled is meant for trusted users who regularly create articles. You would fall into that category so it seems appropriate for you to have it. But not all admins are content creators, it seems strange that 1/2 all active admins are bothered enough to give it to themselves. Admins make mistakes too. Polyamorph (talk) 14:00, 9 December 2021 (UTC)[reply]

I do not think it is a correct description of what happened. Most admins who are now autopatrolled were autopatrolled before getting the admin flag, and had the autopatrolled flag restored by bot. Some (including myself) had autopatrol before and re-assigned themselves the flag, and some decided to go through the request. I guess it is only a handful of admins who have never been autoparolled and who assigned themselves the flag.--Ymblanter (talk) 17:11, 9 December 2021 (UTC)[reply]

Not to split hairs, but I don't think Barkeep49 is a bot Sdrqaz (talk) 17:31, 9 December 2021 (UTC)[reply]

Joe Roe didn't specify the fraction of those who held the perm previously. So I guess that would be useful to know. Polyamorph (talk) 17:54, 9 December 2021 (UTC)[reply]

Approximately 129 admins held the usergroup at one time. It was this group that I restored the permission to. Best, Barkeep49 (talk) 18:15, 9 December 2021 (UTC)[reply]

Thanks Barkeep49. In other words 119 admins decided to grant themselves the perm for reasons unknown. Polyamorph (talk) 18:29, 9 December 2021 (UTC)[reply]

I think there's a variety of reasons admins may have made this choice. I know some admins want to do it for a reasons talked about by Xaosflux as they may create new pages (rather than new articles as per autopatrol) and don't want those pages to need reviewed. Another example would be, I know of one admin who got admin before autopatrol existed but who has been enough of a content creator that they'd have had autopatrol if it had existed. Best, Barkeep49 (talk) 19:16, 9 December 2021 (UTC)[reply]

I'm pretty much going to drop this note and not look back here unless asked, but there are various perspectives where this flagging is at play - for example, on this page, despite the being called "New pages patrol" doesn't seem to be frequented heavily by editors primarily concerned with new pages, only by those actually doing "New article patrol". Patrol marks are useful elsewhere as well, for example when patrolling Special:NewFiles. — xaosflux ^Talk 20:10, 9 December 2021 (UTC)[reply]

I am not on the list which means those who had autopartol but granted it to themselves are not included.--Ymblanter (talk) 19:40, 9 December 2021 (UTC)[reply]

I'll just note that this change was made as part of an RfA change not some broader look at adminship. The change does allow candidates who are worried about content creation criticism to declare they will not take the flag. In that way it's going to have the intended effect, which I agree will not be substantial. Best, Barkeep49 (talk) 18:17, 9 December 2021 (UTC)[reply]

This is not correct. It is not done. It will be done next week. See this phab task if you want to know more or follow the actual implementation. Best, Barkeep49 (talk) 18:11, 9 December 2021 (UTC)[reply]

• I did support it in fact, and Celestina007 is not wrong, but it's actually a toothless new piece of policy because it won't prevent admins from abusing it again as they have done in the past. It might however ensure that new RfA candidates have created sufficient content to avoid arbitrary opposes on the lines of 'Not enough content work'.

That said, in the feed there are plenty of list display options to select in the 'Set filters' panel . I'll ask Phab to include a new one: 'Created by Autopatrolled users'. I didn't think of it at the time of its development because the extent of the abuse had not come to light. There are currently 4,252 users with the autopatrolled right, it might be an idea to run a Quarry to see exactly how many articles are being created by Autopatrollers and what level of burden it would be on NPP if the right were scrapped altogether. Kudpung กุดผึ้ง (talk) 11:22, 10 December 2021 (UTC)[reply]

There are ways admins create new pages other than writing articles. For example, if I'm working an SPI case and block a bunch of socks, I usually tag them, which usually means creating user pages for them all. An active SPI clerk might easily create dozens of new user pages in a single session - don't need reviewing. I'm in the 'autopatrolled before I was an admin' camp, which is why I reinstated it, but I wouldn't assume someone else reinstating it on their own account is doing it without good reason. Girth Summit _(blether) 13:14, 10 December 2021 (UTC)[reply]

But there's no real need to mark pages outside the mainspace as patrolled – NPP has better things to do. A candidate who goes to WP:PERM/A with the primary reason of non-mainspace creation would probably be laughed out of there. I know that autoreviewer is sometimes given to SPI and Committee clerks, but there's no real demand to do so. Sdrqaz (talk) 13:20, 10 December 2021 (UTC)[reply]

(edit conflict)NPP is (or was) primarily concerned with mainspace. New articles are the pages that appear in the feed. Kudpung กุดผึ้ง (talk) 13:25, 10 December 2021 (UTC)[reply]

Kudpung, I agree a quarry report showing how many articles are being created by autopatrollers would be useful. Would that need to include drafts created by other users and moved to mainspace by autopatrollers, as part of AfC or otherwise? The number of articles being created by non autopatrollers in the same period would be an interesting comparison. TSventon (talk) 17:19, 12 December 2021 (UTC)[reply]

It would be extremely useful TSventon. The two other types of New Pages to include are, as you say, drafts created by other users and moved to mainspace by autopatrollers, and of course redirects 'undireted' and turned into new articles. This has often been a method to usurp a redirect for the purpose of evading scrutiny at NPP. Kudpung กุดผึ้ง (talk) 21:49, 12 December 2021 (UTC)[reply]

I'm not sure if you're aware, but pages are often marked as Reviewed when they're converted from redirect to article. Bahçeköy, Düzce would be an example. –dlthewave ☎ 02:04, 15 December 2021 (UTC)[reply]

Autopatrol stats in quarry

Here is the # of autopatrolled articles created in or sent to mainspace per month. Looks like it's around 25,000-35,000 per month. We can refresh this query in a month or two to see if this goes down. Let me know if you need the data in a different format. We get 12 months of data before the PageCuration extension deletes the old data. –Novem Linguae (talk) 02:43, 13 December 2021 (UTC)[reply]

Thank you, Novem Linguae, does that mean that twice or three times as many autopatrolled articles are created as non autopatrolled articles? TSventon (talk) 00:49, 15 December 2021 (UTC)[reply]

@TSventon. Good question. According to this slightly adjusted query, NPP marks around 25,000-31,000 mainspace articles as reviewed per month. So it looks like half of all new articles in mainspace are autopatrolled, and half of mainspace articles are manually reviewed by an NPP. –Novem Linguae (talk) 02:54, 15 December 2021 (UTC)[reply]

Novem Linguae (and Kudpung), those figures seem to include redirects, e.g. 40,022 November NPP reviews on the report compares to the latest Wikipedia:Database_reports/Top new article reviewers 30 day figures of 43,111, comprising DannyS712 bot III 24,147, Rosguill 5,980 and others 12,984. Could redirects be excluded from the reports you linked? TSventon (talk) 15:54, 19 December 2021 (UTC)[reply]

Good catch. Updated. Looks like it's about 10,000 per month marked as reviewed, and 10,000 per month autopatrolled. –Novem Linguae (talk) 16:12, 19 December 2021 (UTC)[reply]

Thanks Novem Linguae. The monthly NPP figures show what you would expect from the backlog chart: around 11,000 reviews per month until August, falling to around 8,000 in September and October, while the backlog increased from 5,000 to almost 10,000. TSventon (talk) 17:16, 19 December 2021 (UTC)[reply]

We lost our two highest volume patrollers around then I think. They were unsung heroes, doing a ton of work. –Novem Linguae (talk) 17:20, 19 December 2021 (UTC)[reply]

Novem Linguae, I had read about the lost patrollers. I wonder if it would be possible to improve the visibility of some of this information, for example could there be a leaderboard for autopatrols, with the same format as for Top new article reviewers? TSventon (talk) 18:18, 19 December 2021 (UTC)[reply]

Hey @TSventon. Here's some queries. Most autopatrols in the last month. Most autopatrols in the last year. If there is enough demand I could turn this into a bot that posts daily to one of the database report pages, but that would take some work. –Novem Linguae (talk) 07:01, 20 December 2021 (UTC)[reply]

I'm frankly gobsmacked by this revelation. After having been involved in NPP for more years than most other users have been registered, It never previously crossed my mind until recently to have such a set of stats extracted. I have absolutely no clue how to quarry for stats, but is it possible to refine that search into:

• Number of creations by admins
• Number of creations by relative new users
• Autopatrolled users who created more than 200 (for example) mainspace articles in those periods?

Sorry to say, but this is looking as if the Autopatrolled right is wortheless and all new mainspace articles should be patrolled. Totally impossible of course with only a tiny handful of the 750 New page reviewers prepared to use the right they asked for. Kudpung กุดผึ้ง (talk) 04:20, 19 December 2021 (UTC)[reply]

Yeah, wow, I would have guessed maybe 10–20% are being autopatrolled at most... but half?? That's frightening. Having been one of the main admins working WP:PERM/A for a while now, our standards for granting autopatrolled are extremely low, and that's not counting all the people given it on the whim of some admin friend way back when. We really need to look into this further and consider making changes to the right. I tried to do this before (see Wikipedia_talk:Autopatrolled#Reviewing_and_removing_autopatrolled and Wikipedia_talk:Autopatrolled#Reviewing_and_removing_autopatrolled) but didn't get much interest, and I think the problem is that we need to show hard data to convince editors not involved in NPP/COIN/etc. that autopatrolled is a risk and should not be handed out willy-nilly. – Joe (talk) 15:52, 19 December 2021 (UTC)[reply]

Let me suggest something from a different angle. We have X amount of editor capacity to review articles. Novem's great stats suggests it's roughly 10,000 articles a month. Are we allocating that capacity in the best, or if not the best a good, way? Now I am always in favor of increasing this capacity... assuming we can do it with qualified people but I also think, as noted, our review capacity is somewhat fragile and top heavy so even doing that doesn't change the question. While I think Joe is right that our standard for autopatrol might be lower than ideal, we're not operating in an ideal environment and I am not so sure we're mis-allocating our reviewer capacity/focus. Best, Barkeep49 (talk) 20:17, 19 December 2021 (UTC)[reply]

Maybe we should consider lowering adjusting our standards. I often find myself looking over a day's worth of new additions to the queue to see if there's something that's within my area of expertise, and as a I look over the list, I notice nothing that really stands out as particularly problematic. I don't really want to mark them as reviewed without a better look, but they're probably just fine. If that happens to other reviewers as well, that's a lot of lost capacity. So I decided to see what happens when I do review at maximum speed, choosing a similar set of articles to work on. I got through 137 in a single day, and after a few days, found that I made two mistakes where an article was AfD'd that I should have nominated, but didn't. I typically average about 5 articles per day, so that's a big change in my productivity, but it comes at a cost in quality. Is 98.5% accuracy good enough? Probably. If I had an easier way to measure that, I might be able to make some adjustments to my reviewing, increase productivity while maintaining a level of accuracy that is acceptable. Vexations (talk) 21:33, 19 December 2021 (UTC)[reply]

I think this is a really interesting train of thought. What can be done to speed up reviewing? User scripts is a big one, scripts like rater and stubsort speed things up. As Vexations mentions above, a less thorough check of sources greatly increases speed while slightly reducing quality. Another idea might be to simplify the NPP flowchart, for example by focusing on CSD and notability, and leaving all the gnoming at the end of the flowchart (stub, tags, wikiprojects) to others. Sometimes I've wondered if we should have 2 "marked as revieweds": one for notability, and one for gnome tags (easier perm to get, or give to all ecp). The shorter the checklist for each queue, the faster the velocity. Oh, and another big velocity booster idea: an RFC to get rid of WP:BEFORE and require GNG and SNG sources to be in the articles for all articles created after the RFC date. AFC does this and I find AFC reviewing to be easier as a result. –Novem Linguae (talk) 00:13, 20 December 2021 (UTC)[reply]

Splitting the review process could have a lot of bureaucratic overhead...I would, however, support reorienting the emphasis of new page reviewing to focus strictly on identifying threats to the encyclopedia (hoaxes, UPE, BLP violations, POV-pushing on sensitive topics, copyright) rather than the strict enforcement of notability guidelines that forms the bulk of actual review time (and which also causes the most disputes with contributors and thus burnout). Quite frankly, if the editors who write sports articles want there to be articles on every decent NCAA prospect or video game fans want to write a 30k word character bio largely relying on the designer's own commentary about the character's development (to name two genres of article filling up the queue yesterday), that's their problem and I don't think it's a productive use of our time to fight them on it. I think that any solution that is premised around us being more reckless in sending things to AFD, however, is a non-starter as it just kicks the backlog can down the road (not to mention that it's going to receive a LOT more pushback from the rest of the editing community). _signed, Rosguill ^talk 00:31, 20 December 2021 (UTC)[reply]

I think that's an intriguing idea, not sure how I feel but I think it's worth thinking about. Is it speedy deletable, UPE, BLP vio, etc? NPP should be doing something about it. Is it potentially not notable? Tag and up to someone else to do the BEFORE to decide whether or not to send to AfD. AfD scars also tends to be one of the leading causes of NPP burnout so in addition to cutting down on review time it could cut down on reviewer fatigue. Curious what the pre-ACPERM people have to say. Best, Barkeep49 (talk) 01:20, 20 December 2021 (UTC)[reply]

One nuance that's probably worth bearing in mind is that while I think we could do to deemphasize the centrality of notability checks in reviewing, notability will nonetheless continue to be an important tool in fighting hoaxes, UPE and BLP issues. Companies, BLPs, recent media works and any other topic with a potential for self-promotion or abuse should still get the full notability evaluation. _signed, Rosguill ^talk 02:09, 20 December 2021 (UTC)[reply]

I'm a pre-ACPERM person, I suppose? We've had a fixation on the question of delete-or-not for as long as I've been involved, but my sense is that, over time, NPPers have become more and more hesitant to use AfD and have instead slowly expanded the scope of alternative routes to deletion – most notably the massive creep of draftification over the past five years combined with the advent of WP:G13. I think I've made it clear enough elsewhere that I think this is a regrettable trend. The things that make AfD (understandably) unpleasant from an NPPer's perspective—the slow process, the onerous BEFORE requirements, that being 'wrong' is a black mark if you ever go for RfA—are the same things that make it good at preserving good content and more importantly, the motivation of article writers. In other words, they're there by design. Conversely, the increasingly bureacraticized draft process is terrible for inexperienced editors, who often (understandably) feel like their work is subject to the whim of a single faceless and unaccountable 'moderator'. It's a bit of a stalemate but I agree that expecting NPPers to use AfD more is a less plausible solution than getting them to use alternative deletion methods less.

I like Rosguill's idea too. In principle there's no reason why NPPers shouldn't be content to slap an article with ((notability)) and then mark it as reviewed, as long as it doesn't have serious content problems. But one rub: what's to stop the creator then simply removing ((notability))? And who would notice? – Joe (talk) 12:12, 20 December 2021 (UTC)[reply]

Nothing obviously. But in this theoretical NPP it would be out of scope. Best, Barkeep49 (talk) 12:45, 20 December 2021 (UTC)[reply]

Inactive autopatrollers

At the moment, 630 users who have the autoreviewer (Autopatrolled) user right have not made any edits to Wikipedia in the last five years. Vexations (talk) 21:58, 19 December 2021 (UTC)[reply]

I'd pull all of those and have them reapply if they become active again. It could be dangerous in the case of hijacked accounts or changing notability standards. (t · c) buidhe 22:14, 19 December 2021 (UTC)[reply]

Be my guest:

Vexations (talk) 23:23, 19 December 2021 (UTC)[reply]

Sorry, not an admin. (t · c) buidhe 23:32, 19 December 2021 (UTC)[reply]

I don't think it's a great idea to pull autopatrol perms from these folks, for motivation reasons (never feels good to get a perm taken away), and also because of our huge backlog. –Novem Linguae (talk) 23:58, 19 December 2021 (UTC)[reply]

@Novem Linguae: Per WP:HATSHOP, I'd like to think that these editors listed above wouldn't let their egos get in the way of the project. A few of those listed above have already been blocked, so I don't think it's unreasonable to insist that tools are for getting work done, not awards for longterm gameplayers. Surely you agree. Chris Troutman (talk) 00:21, 20 December 2021 (UTC)[reply]

Just like removing autopatrol from admins, what demonstrable problem does this solve? Is there evidence that these folks cause greater than average disruption with their perm? –Novem Linguae (talk) 00:48, 20 December 2021 (UTC)[reply]

It's not a demonstrable problem nor do I have evidence for you. It's a principle. We don't give userrights as political favors. They have tools in order to edit the wiki just as we expect those with "teh bit" to be actively swinging a mop. No edits in the past five years tells me those editors don't need tools, at all. Chris Troutman (talk) 01:19, 20 December 2021 (UTC)[reply]

Someone should start an RfC on (i) what are conditions fpor removing autopatrolled flag for inactivity; (ii) what are the conditions for a flag restoration. I do not think unilateral removal of a flag in the absence of misconduct is a good solution.--Ymblanter (talk) 11:12, 20 December 2021 (UTC)[reply]

Graphs are unavailable due to technical issues. There is more info on Phabricator and on MediaWiki.org.

NPP backlog, October 2017–present

@Novem Linguae: The principle of least privilege says that we should not assign advanced permissions to accounts that do not need them. We already apply this logic to practically every permission from sysop up, and I would argue that autopatrolled carries a substantial amount of risk compared to other minor perms. As for the backlog, I understand that NPP always feels put-upon, but putting it in larger perspective: we had a perrenial backlog of around 20,000 for many years until 2017. In the last year, we have gone from a historic low back to the level it was last autumn. These cycles are natural and since, as a baseline, NPP handles hundreds of articles a day, pulling autopatrolled from small numbers of people will not have a significant effect on the backlog. In this case, it will have no effect, because these editors have created zero articles in the last five years. – Joe (talk) 11:50, 20 December 2021 (UTC)[reply]

Clarification; it's not just that they haven't created any new articles; these users have not made any edits at all in the past five years. Vexations (talk) 14:21, 20 December 2021 (UTC)[reply]

Whilst I agree with Ymblanter that prior consensus is needed to remove these en masse, there seem to be plenty of individual cases in this list that should be looked into: indef blocked users, sockpuppets, an account that has never edited enwiki, etc. I'd encourage other admins to scan through it and see if there are any that should be pulled immediately. – Joe (talk) 14:36, 20 December 2021 (UTC)[reply]

On that note, @Vexations: is there a quarry query behind this list? And/or is it easy for you to reorder by edit count? – Joe (talk) 14:50, 20 December 2021 (UTC)[reply]

yes, https://quarry.wmcloud.org/query/60773 I’d have to look into edit count. Vexations (talk) 16:23, 20 December 2021 (UTC)[reply]

Joe Roe This query yields a list of autopatrolled users, sorted by editcount, who have not made any edits in the past five years: https://quarry.wmcloud.org/query/60800 It's not very efficient, my apologies for that. If anyone has suggestions for improvement, I'd love to know what I could do better. Vexations (talk) 21:52, 21 December 2021 (UTC)[reply]

@Joe Roe, Vexations, Ymblanter, Chris troutman, Novem Linguae, and Barkeep49: I don't believe that a broader removal of the Autopatroll right would impact on NPP. Indeed, IMO the threshold for obtaining it has been reduced by steps to a dangerously easy level. I would advocate returning it to its origial, 100 'near perfect' articles.

BTW, has anyone though to open a Phab ticket to get 'Articles by Autopatrollers' added to the search prefs of the New Pages Feed?

There is enough consensus in this short sub-thread to get that RfC for the rights removal started. The sucess of an RfC depends on its carefully worded proposition statement, and holding it in the correct venue. The hat collectors would naturally object to losing their rights, they always do, and the same problem exists for stripping the NPP right from the 600 people who don't use it.

It's interesting to note however that the monumental backlog began to fall once the (with Wikipedia's traditional lethargy) effects of the new NPR right kicked in, and then when ACPERM was rolled out in April 2018. It's most likely that the new high backlogs are far more due to a lack of interest on the part of the vast majoity of rights holders than any huge increases in New Articles. Naturally the burn out of some of the most prolific reviewers is to be taken into consideration, but that's to be expected after leaving them to all the work, and the lack of consequential coordination since other coords retired from Wikipedia or put their talents to work in venues at higher levels of equal importance and which also require urgent attention. All these milestones are easily recognisable in the graphic.

There is no point in appealing to the hat collectors to do some work - they won't. OTOH, recruitment and according the NPR right must be done carefully and with a high degree of due diligence. Kudpung กุดผึ้ง (talk) 01:39, 24 December 2021 (UTC)[reply]

Redirects

I understand there were reasons to remove autopatrol from admins generally, but perhaps we can automatically review redirects created by admins? Surely there wouldn't be issues with those, right? If yes, it should be fairly simple to add to DannyS712 bot III. Thoughts? --DannyS712 (talk) 00:25, 2 January 2022 (UTC)[reply]

Since the whitelist is fully protected, administrators can add themselves to it if they wish (just as they can self-assign autoreviewer). Alternatively, they can request it on the talk page if they want greater scrutiny (just as some administrators requested at WP:PERM/A). As many of the thousand-ish administrators don't frequently create redirects, I'm more in favour of a case-by-case basis being used rather than the pre-emptive addition of all of them. Sdrqaz (talk) 00:34, 2 January 2022 (UTC)[reply]

I was going to have a dedicated query instead of just using that list - even if they don't frequently create redirects, do we expect there to be any inappropriate ones? DannyS712 (talk) 00:40, 2 January 2022 (UTC)[reply]

I guess the stock answer to that question (because someone is going to bring it up) is to invoke the case of Neelix and its accompanying speedy deletion criterion. I'm curious to see what other patrollers think. Sdrqaz (talk) 00:50, 2 January 2022 (UTC)[reply]

I just came across Kheo17's creations which seem to have made it past NPP. As I'm sure many of you are aware, this type of geostub mass creation based on tabular data is problematic and probably should have raised some red flags. Just a reminder of an editing pattern that folks should look out for and put a stop to if they see it. –dlthewave ☎ 03:20, 11 December 2021 (UTC)[reply]

@Dlthewave, I’m sorry are you concerned because the article is a stub or because it comprises tabular data? If references are in the article and the article clearly passes GNG or the relevant SNG, insofar as it meets WP:PSA I don’t think there is anything inherently wrong. Of course that is just my own understanding. Celestina007 (talk) 19:56, 11 December 2021 (UTC)[reply]

Celestina007 is correct. The relevant SNG here is WP:GEOLAND. Populated, legally recognized places are presumed notable. The stubs should be tagged for improvement if needed, otherwise if there are no other issues these articles should be marked as reviewed. Polyamorph (talk) 20:05, 11 December 2021 (UTC)[reply]

Since the headline is "poorly sourced" I assume the issue is that the source is not reliable and/or does not support the claim to passing GEOLAND—an especially important issue to evaluate in the case of mass creation. (t · c) buidhe 21:31, 11 December 2021 (UTC)[reply]

While I agree with Celestina007 and Polyamorph, I do not think these articles are sourced well enough to pass GEOLAND, let alone GNG. Unlike Geostubs which are solely sourced to government censuses, these appear to be solely sourced by this link... although I only looked at about 10 of the articles. This is not good enough to establish GEOLAND. A similar case came up a couple of years ago regarding US stubs which relied on GNIS entries to establish notability, and I think this is a similar instance. An alternative to prodding them would be to draftify them, to allow Kheo17 to provide better sourcing. Onel5969 ^{TT me} 22:10, 11 December 2021 (UTC)[reply]

That's correct, Buidhe. The two main issues I see are that A) the source does not show incorporation or other legal recognition for these villages and B) WP:NGEO specifically excludes maps and tables from being used to establish notability. When similar mass creation efforts have been discussed at ANI and other venues, the consensus has been that these types of stubs should not be created en masse from databases without secondary sourcing. They often end up being redirected or deleted which requires massive amounts of time and energy from other editors who must do the legwork of establishing notability (or lack thereof) which was never done by the article creator. –dlthewave ☎ 22:35, 11 December 2021 (UTC)[reply]

@Onel5969, Buidhe, My interpretation or rather what I idiosyncratically deduced from the entry was Dlthewave was unhappy with the article being a stub thus the statement I made above which reads “are you concerned because the article is a stub or because it comprises tabular data?” to which I further stated that if the said article or articles as the case might be met WP:PSA then there wasn’t any inherent problem with that. But as Buidhe, Onel5969 pointed above, it seems Dlthewave's major concern was that the articles were poorly sourced. If that is the case then it raises one question, who reviewed the articles? especially when it is a stub that doesn’t meet the relevant SNG? From what I can see from here, Dlthewave mentioned that this articles has passed through NPP, which means from my understanding’ means they have been reviewed. This then brings to my mind the grievance of Doomsdayer520 who stated that most new page reviewers are too quick to use the green button rather than take up the time to perform due diligence before marking as reviewed. Celestina007 (talk) 15:09, 13 December 2021 (UTC)[reply]

To find out who marked the geo stubs as reviewed, just check the log of the articles concerned Polyamorph (talk) 18:45, 13 December 2021 (UTC)[reply]

@Polyamorph, thanks Poly, I did know how to check to see the creator of an article I was just too tired and fatigued to actually bother to check who. Celestina007 (talk) 14:56, 14 December 2021 (UTC)[reply]

If it is clear that stubs that fail GNG, SNG and/or WP:GEOLAND are being passed, then they should be either redirected to the nearest sttlement that has an article that complies with notability, or sent to draft and the reviewer should be cautioned. At least the drafts will expire by G13 so they are not actually increasing the workload at AfC where they can be declined with a mouseclick. Imprecise patrolling is the very reason the NPR right was created. There may be backlogs but filling the encyclopedia up with needless junk is a waste of everyone's time and patience sorting it out. Noting BTW, that the November backlog drive wasn't exactly a roaring success. Have the year-end barnstars been prepared? Kudpung กุดผึ้ง (talk) 15:35, 13 December 2021 (UTC)[reply]

I had the same issue with Belanov87's geostubs. I initially objected to them because they seemed to be created by a unapproved bot. The user denied it, which I took at face value. Then I saw other reviewers were just accepting these articles, so I capitulated. In retrospect, since sources were tables and some autogenerated web profiles, I guess they could have been PROD'd or draftified? MarioGom (talk) 17:25, 13 December 2021 (UTC)[reply]

I think you have to be familiar with the specific issues surrounding mass geostub creation in order to pick up on these patterns; it's understandable that someone could take a good look at the source, recognize it as an Official Government Document™ and give it the green light with the assumption that it will be expanded in the near future. That's why I wanted to educate folks on a red flag to look out for that they might not be aware of. However, it seems like the NPP process is set up to review the content of individual articles rather than identify and address larger editing patterns. Folks should certainly be using draft space more and sending incomplete stubs there as well; I think it's fair to expect all editors to develop articles in draft space until they have enough sources to establish notability (SNG or GNG) at the very least. –dlthewave ☎ 05:04, 14 December 2021 (UTC)[reply]

I'm with Dlthewave in spirit but not in blaming NPP. According to the SNG. populated, legally recognized places are presumed notable. They don't need sources to meet the SNG and so the exclusions of source types for establishment of notability are moot. Also our well-supported effort to add wording to discourage mass ceation of stubs got marked as "no consensus" (Wikipedia talk:Notability/Archive 73#Adding one new thing to the current SNG text) This isn't the way I want it, but it's the reality at AFD. If a NPP person takes it to AFD, they'll get verbally beat up with things like "Didn't you read the SNG?" North8000 (talk) 19:09, 13 December 2021 (UTC)[reply]

Except that at minimum you need a source that supports the assertion that the place is legally recognized in order to pass the SNG. (t · c) buidhe 19:54, 13 December 2021 (UTC)[reply]

Yes, but but that's not the normal meaning of a source establishing notability. North8000 (talk) 20:34, 13 December 2021 (UTC)[reply]

@North8000, whilst I agree that I’m not familiar with GEOLAND and whatnot, Semantics and linguists ambiguities aside, I totally agree with Buidhe that an article must have at least one reliable source present, if there are SNG's that give leeway for certain types of articles to be unsourced then that SNG needs to be re-written as it is not in agreement with WP:V. Celestina007 (talk) 15:02, 14 December 2021 (UTC)[reply]

@Celestina007: I agree with you 100% that that is how is SHOULD be. I'd probably start with the "did it for a living for one day" sports SNG and then move on to geostubs. But my previous post was about how it IS rather than what it should be.North8000 (talk) 20:48, 14 December 2021 (UTC)[reply]

I'm pretty new to the NPP, but I think this relates to a broader debate about the level of "agressivity" that reviewers should have. While I'm inclined to agree with Kudpung's philosophy, it's not easy to go ahead and draftify/PROD 20+ articles of a series in one go, unless you're a very experienced reviewer and you know that you can trust your judgment. I've already expressed this. JBchrch talk 20:01, 13 December 2021 (UTC)[reply]

• We need to be clear that tabular data sourcing refers to sources that have the village only on a list that gives minimal information. It does not mean that settlement specific entries on a national government census do not count for notability when they have significant coverage such as 1000-2000 word + detailed statistics for the specific location. I had a recent conversation at RSN about this with a member of WP:GEO who explained this, imv Atlantic306 (talk) 02:58, 14 December 2021 (UTC)[reply]
• This reminds me of my previous question I asked here about what to do with poorly sourced professors and politicians. SNG passes with poor sourcing appear to be one of those gray areas where you can't get a consistently straight answer. Some people say draftify, some people say mark as reviewed, etc. –Novem Linguae (talk) 03:06, 15 December 2021 (UTC)[reply]
• I think the key is does a stub, even though poorly, or sparsely sourced, show that the article's subject passes the SNG? For example, an article on a geostub, which is sourced to that government's census data, clearly indicates it is a legally recognized populated place, and therefore passes GEOLAND. While a geostub which is simply referenced to a source which shows that it exists, but doesn't not categorically state it is a legally recognized populated place, does not pass the SNG. In the former, I would mark it reviewed, and tag it for a single source. In addition, if there was other unsourced material in the I would either tag it as needing more refs, or if there were only 1 or two lines, I might simply place CN tags instead. In the latter instance, I would make a judgement call about whether to draft, redirect (if appropriate, and a logical target is available), or PROD/CSD if I don't find any info on the subject. For foreign country articles, unless it's obviously a hoax, or a one line unsourced stub like, "XXXXX is a village in Vietnam", I would rarely go the PROD/CSD route. Politicians are similar. If there is a state senator from Nebraska with a single ref, but that ref is to the official Nebraska legislature page showing they are a member, then that person satisfies the SNG. Of late, I'm seeing a spate of MP articles for Indian politicians which fall into this category. But if the only source is a newspaper article saying, State Senator XXXX officiated at the grand opening...", then again, draftify or redirect. While the newspaper may be a reliable source, it's not an official source. Some reviewers may disagree with that, but there was a stub about 7 years ago, where this was exactly the case. The newspaper said "State senator", but that was an error, they were running for State Senator, but they were only a local councilman at the time. NACADEMIC is even dicier. If there is a person who due to their position (president of a college, for instance), and there is a single cite, they pass the SNG. Similarly, if there is a line in the stub which says that the person is cited frequently, and then adds a google scholar link which shows high citation counts, they pass the SNG. But often, neither of those is present. That's when the redirect/draft/tag option comes into play. However, since a google scholar search is relatively easy, if they are an associate professor with a high citation count of 10, I'm going to PROD them. Sorry to blather on, but I thought a look into reviewing thought processes might be beneficial. Onel5969 ^{TT me} 15:08, 15 December 2021 (UTC)[reply]

• Your argument about the newspaper is the caution about using secondary sources or third-party sources. The ultimate solution to the incorrect info would be utilizing multiple secondary or third-party sources so the statements could be verified. I think it is merely understanding if there are lower quality sources about a geostub at its article creation - it most likely won't be expanded. – The Grid (talk) 15:41, 15 December 2021 (UTC)[reply]

  Ultimately, what matters is whether an AFD will delete the article. If the topic is notable, it almost always survives AFD regardless of its state. Any article that would survive an AFD should be marked as reviewed, per current NPP guidelines. If you look at the NPP flowchart, you are supposed to pass articles on notable topics even if they are completely unsourced. Usedtobecool ☎️ 16:28, 15 December 2021 (UTC)[reply]

"... an article on a geostub, which is sourced to that government's census data, clearly indicates it is a legally recognized populated place, and therefore passes GEOLAND." It's not that simple though. Census tracts are not inherently notable under the SNG even though they're sourced to official government documents; this was the root of the whole Iranian "village" debacle where articles were created for the lowest-level census areas which sometimes turned out to be individual farms. I fear that reviewers may see a government source and assume that it establishes notability without even clicking on it to see what it actually says. –dlthewave ☎ 16:42, 15 December 2021 (UTC)[reply]

As I mentioned before, I'm with you in spirit but I've been talking about the current realities. With that in mind in comments we need to be clear which of these we are talking about:

1. Where it appears established that they a populated, legally recognized place (but still questioning wp:notability)
2. Where it is not been established that it is a populated, legally recognized place

Sincerely, North8000 (talk) 18:51, 15 December 2021 (UTC)[reply]

To be clear, I'm talking about when an editor finds a government source that they believe is a list of "populated, legally recognized" places but turns out to be something else, for example census tracts. The source is in the article and appears at first glance to meet GEOLAND #1 which is why these are getting past NPP. –dlthewave ☎ 21:22, 16 December 2021 (UTC)[reply]

In many countries census tracts are in fact "lists of populated, legally recognized places". Not every country is like the US. This is already the third time I am trying to point this out to you, but you do not seem to be interested in listening.--Ymblanter (talk) 22:21, 16 December 2021 (UTC)[reply]

Please if you can take up little time to read this it would serve as a bedrock/prerequisite to what I’m about to say pertaining the sudden awakening of Sleeper accounts. Now, On December 14th I draftified this article created by the editor named Jefak007 (they are now blocked). Upon further observation I noticed this was a sleeper account who prior creating this article was inactive since 2017, at the time they were 10 years old with less than 90 edits, who basically re-awakened to create the promotional article. Now, today whilst patrolling new articles I stumbled upon another account Hazelnike who apparently has been here for 14 years with 540 edits, they created this article which I draftified on the basis of a strong COI suspicion, & as is required, I initiated a conversation with them over 40 minutes ago and they are yet to to respond. In summary, I think we should be on the lookout for sleepers I’m seeing a lot of well written articles by WP:SLEEPER accounts which I believe are alternate accounts of blocked editors, or, for bad faith editors still editing here. Coincidentally it appears as though the first editor I speak about who is now blocked was in communication with the (in my opinion) LTA editor named Oluwa2Chainz, see this, this & this. I believe we need to take this seriously by being on the lookout for Sleepers. Celestina007 (talk) 23:28, 18 December 2021 (UTC)[reply]

In many cases of article topics not from the English speaking world, the article is a translation from another language Wikipedia but is not marked as such. Always check to see if the article exists on another language wiki and if so, whether the new enwiki article is original. If not, it is a copyright violation and breaking Wikipedia licensing. Just today I uncovered cross-wiki sockpuppetry by a user blocked on another language wiki for source fabrication, who was translating his articles deleted on German wikipedia into English. Several of these articles were passed by NPPers who never checked German wiki to see the history. (t · c) buidhe 01:16, 19 December 2021 (UTC)[reply]

::sigh:: another case of poor reviewing. It's being suggested in other places already that if NPR is not being done thoroughly and/or too little interest from the 750 reviewers to reduce the backlog : 'Why bother reviewing new articles at all?' particularly ones that are hardly likely to be the source of a search. Kudpung กุดผึ้ง (talk) 03:59, 19 December 2021 (UTC)[reply]

• It would be nice if File:NPP flowchart.svg had a final step to check for reuse attributions (translations, splits from existing articles). MarioGom (talk) 15:25, 20 December 2021 (UTC)[reply]

The file is here. The creator is semi-retired and currently has no intentions of returning to regular editing. He won't object. Kudpung กุดผึ้ง (talk) 14:41, 21 December 2021 (UTC)[reply]

We should probably get consensus for changes to the flowchart, especially if adding work. Also, the original files are lost, so updates may not be as easy as they appear. –Novem Linguae (talk) 15:41, 21 December 2021 (UTC)[reply]

Novem Linguae, There is no need to get consensus to improve it, any more than there would be to change my earlier, simpler, chart. It's a private offering and it's freely licenced. A flow chart like that can be put together with patience in a very short time if one already knows thre content for it. ICPH already did all the hard work with his conceptual design. There are dozens of free online flowchart builders out there. ICPH used Draw.io, now known as Diagrams.net, but Inkscape is another, and one could just as easily use Apple Pages, part of the default Mac suite of apps. Bear in mind however, that while that flowchart is excellent, we did NPP for years with just my simple one and there is no substitute for experience, and learning all the deletion and notability policies -and that's the steep side of the learning curve. Kudpung กุดผึ้ง (talk) 13:04, 23 December 2021 (UTC)[reply]

@Kudpung. I think the flowchart is a summary of and codification of correct NPP reviewing, and changing its steps should get consensus, since it affects a lot of people's workflows. As to the SVG thing, I am proficient in Inkscape, just now I downloaded a couple SVG versions of the flowchart to investigate modifying it, and the file appears to be a c:Template:Fake SVG, which means that it would be difficult to modify. I suspect an SVG file in the proper vector format would also be difficult to modify without its original Microsoft Visio-style underlying generating program, since you'd have to individually move around boxes and lines, instead of having them auto generated for you. I have been interested in flowchart updates in the past so that is why I am commenting on this issue. I left a list of 8 flowchart update ideas here in July 2021, and I have also color coded the flowchart before, see here. –Novem Linguae (talk) 02:12, 24 December 2021 (UTC)[reply]

Page views at the tutorial are quite good over the last five years at a monthly average of 2,178, but over the last 12 months page views have dropped to 1,692. This is still fairly good for a relatively obscure and niche help page, but the hits are probably due to new reviewers referring back to the page or its flowchart many times. However, as I mentioned above, the knowledge of deletions and notability are the hard bit. The flowcharts are quick reference of what to do next. Inkscape is a very elementary but nice piece of free software, it does all you want in just a few minutes. It's very intuitive and doesn't even need a user manual. It just depends how much time you want to spend on it. Kudpung กุดผึ้ง (talk) 10:46, 26 December 2021 (UTC)[reply]

Gaviscon was redirected without discussion in 2016 to Antacid. Gaviscon is possibly the most widely used over-the-counter and prescribed antacid/heartburn/reflux remedy in the UK. If Maalox is allowed an article, why not Gaviscon which seems to me to be a highly likely target for Wikipedia searches? Gaviscon is Wikilinked from 72 pages including GERD. In addition to its existing sources, admittedly many of which are primary, reliable sources are at least available about it by the UK government NHS, and the US government articles and the Thai government's FDA site, and peer reviewed papers, and in Le VIDAL, the official French pharmacopeia. At one stage (around 2005) , according to the UK Govt. Office of Fair Trading 49% of the sales of Gaviscon were through NHS prescritpions]. So, see WP:MEDRS:

Ideal sources for biomedical information include: review articles (especially systematic reviews) published in reputable medical journals; academic and professional books written by experts in the relevant fields and from respected publishers; and guidelines or position statements from national or international expert bodies. – (The bolding is mine)

If Maalox is allowed an article, why not Gaviscon which seems to me to be a highly likely target for Wikipedia searches? If there is no response within 7 days I will add the NHS and NCBI sources and republish it in mainspace. Kudpung กุดผึ้ง (talk) 06:31, 19 December 2021 (UTC)[reply]

@Kudpung: I don't see why you're asking here? The user who redirected it is now blocked and the redirection was never due to a strong community consensus. If you can write a decent article on the subject, I don't see why anyone here would object... Elli (talk | contribs) 06:33, 19 December 2021 (UTC)[reply]

Elli I thought someone would like to do it. You'd be surprised what people object to when I do something! I'm trying to retire from this circus ;) Kudpung กุดผึ้ง (talk) 06:36, 19 December 2021 (UTC)[reply]

This is copy pasted from This AN REPORT. I have moved most of the articles back to the new pages feed and would be proceeding to nominating them for deletion, I believe our collective effort would be required here. Celestina007 (talk) 15:18, 19 December 2021 (UTC)[reply]

In response to this serious mess I have mass nominated most of their spam articles see here can anyone be kind enough to in cleaning this serious mess? Celestina007 (talk) 16:05, 19 December 2021 (UTC)[reply]

Note that if the account gets blocked, G5 will apply to most creations that haven't gone through AfD yet. I would just hold off a bit. MarioGom (talk) 11:40, 21 December 2021 (UTC)[reply]

Could a Russian speaker help assessing reliability of sources at NUTSon? Thank you. MarioGom (talk) 15:57, 20 December 2021 (UTC)[reply]

MarioGom I don't recognize any of the sources on-sight, which is a bad sign. A cursory glance at content, bylines and editorial policy pages leaves me with the impression that these sources are churnalism or churnalism-adjacent. _signed, Rosguill ^talk 16:06, 20 December 2021 (UTC)[reply]

I will have a look later today, in three to four hours.--Ymblanter (talk) 16:32, 20 December 2021 (UTC)[reply]

First, there are not 16 sources as one could think but only 5. Second, none of them is a mainstream established source (which in Russia often also have issues, but still). As Rosguill correctly says, they are churnalism, of the same type as the sources in the articles about our Nigerian friends, though prersumably not affiliated and not direct advertisement. Probably AfD is the safest route.--Ymblanter (talk) 19:06, 20 December 2021 (UTC)[reply]

And the credentials of the creator are not inspiring either.--Ymblanter (talk) 19:07, 20 December 2021 (UTC)[reply]

The only other articles I could find where any of the sources from NUTson have been used are YooMoney and Nanosemantics. Vexations (talk) 13:36, 21 December 2021 (UTC)[reply]

This is a Çelebicihan article. Deleted, sock blocked. MER-C 17:11, 21 December 2021 (UTC)[reply]

Thanks! MarioGom (talk) 12:47, 23 December 2021 (UTC)[reply]

A page created about 10 days ago is pending patrol at: RobOps Please do the needful. Wickedwiki2 (talk) 15:45, 24 December 2021 (UTC)[reply]

Thanks for the quick review User:Chris_troutman Happy Holidays.Wickedwiki2 (talk) 16:03, 24 December 2021 (UTC)[reply]

@Wickedwiki2: Yes, harassing us volunteers to patrol a page you made is a quick way to get the thing nominated for deletion. I did "the needful". Chris Troutman (talk) 16:08, 24 December 2021 (UTC)[reply]

Currently there is a discussion over at Wikipedia talk:WikiProject College football#Notability of several CFB players which NPPer's might have some interest in. Onel5969 ^{TT me} 15:54, 24 December 2021 (UTC)[reply]

thank you for post--Ozzie10aaaa (talk) 17:33, 28 December 2021 (UTC)[reply]

Are disambiguation pages converted to articles relevant to this project? I believe that they do not feature on the feed, but they can be entirely new articles. An example is Environmental defender, written by Larataguera, which I saw at DYK. TSventon (talk) 22:51, 27 December 2021 (UTC)[reply]

buidhe, the history of the disambiguation page is now split to Environmental defender (disambiguation) so should Environmental defender now go through NPP? TSventon (talk) 11:09, 28 December 2021 (UTC)[reply]

@TSventon: The page shows as unreviewed, so yes it's relevant and will be reviewed in due course. Elli (talk | contribs) 17:02, 28 December 2021 (UTC)[reply]

@Elli: the page didn't show as unreviewed on 27 December. @GeoffreyT2000: requested a history split today and then added the new page to the page feed today, which probably shows that the article was of interest to the project. TSventon (talk) 17:16, 28 December 2021 (UTC)[reply]

Hi -- not sure this is the right place to ask, but could someone look over the above article, and hopefully give it the NPP seal of approval, please? The subject recently died, article was created by an experienced editor from Indonesian press coverage, and it is now up for In the News (Wikipedia:In the news/Candidates#RD:_Caliadi). It's already been ok'd by a couple of reviewers who read Indonesian. I found no obvious problems but I don't read Indonesian so hesitated to tick it myself. Thanks! Espresso Addict (talk) 07:31, 31 December 2021 (UTC)[reply]

 Done I was reading the ITN entry (mainly because the term was new to me). No obvious issues stand out. The review that's happened at ITNC is better than most NPP reviews, as we don't have reviewers by language/region. Regards! Usedtobecool ☎️ 07:35, 31 December 2021 (UTC)[reply]

Thanks very much! I think it's had a fairly comprehensive review, all told! Cheers, Espresso Addict (talk) 07:38, 31 December 2021 (UTC)[reply]

There is a current AFD ongoing; Wikipedia:Articles for deletion/Liquorose, The individual in question is only notable for one event, and it seems the current participants in the AFD do not grasp to satisfaction the concept of BLP1E, I would like for unbiased experience new page reviewers to keep a close eye on this one, Infact in the past prior the article being nominated for deletion by a different editor I have long suspected the article to be a potential undeclared COI. I am concerned because the current keep !votes are by editors who have had a history of undeclared COI, Infact one of the keep !voters is currently under a topic ban and the other editor (the creator of the article) has been strongly suspected of creating possible undeclared COI articles. Celestina007 (talk) 21:07, 2 January 2022 (UTC)[reply]