Tutorial

Discussion

New page feed Sorted NPP Browser Easy Reports

Reviewers Resources School Top

Curation tool Suggestions

Coordination Talk Awards Newsletter Backlog drives

Archives

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

Wikipedia talk:Page Curation: 1, 2, 3, 4, 5, 6, 7, 8, 9 Wikipedia talk:New pages patrol: 1, 2, 3, 4, 5 Wikipedia:New pages patrol/Noticeboard: 1

This page has archives. Sections older than 30 days may be automatically archived by Lowercase sigmabot III when more than 20 sections are present.

I don't know if anyone can look into this but User:SDZeroBot/NPP sorting/Culture/Biography/Women has not been updated since 24 November. I find this a very useful way of checking out new articles on women rather than looking through the basic list of new articles.--Ipigott (talk) 16:24, 26 November 2024 (UTC)[reply]

Pinging the bot operator, @SD0001. Hey man im josh (talk) 17:40, 26 November 2024 (UTC)[reply]

Toolforge bots have been unstable for the last 24 hours due to phab:T380844. May or not be the root cause here, depending on when this started. –Novem Linguae (talk) 21:31, 26 November 2024 (UTC)[reply]

See it's up any running again now. Thanks for your help.--Ipigott (talk) 08:40, 27 November 2024 (UTC)[reply]

Shahi Jama Masjid - are there any issues? I will like to fix them. Upd Edit (talk) 15:19, 1 December 2024 (UTC)[reply]

@Upd Edit, this article hasn't been marked as patrolled yet, so search engines aren't indexing it. If an article is less than 90 days old and hasn't been marked as patrolled, it won't show up in search engine results. But if a new page reviewer marks it as patrolled within this time, it will be indexed. Articles older than 90 days are automatically indexed, even if they remain unpatrolled. – DreamRimmer (talk) 15:47, 1 December 2024 (UTC)[reply]

Thanks! Upd Edit (talk) 17:48, 1 December 2024 (UTC)[reply]

Hello. I was away from NPP/R for a few years. I am trying to start patrolling again. Today, I went to special:newpagesfeed, and visited some articles. To my surprise, a lot of them were very old articles like Operation Krivaja '95. It was created in May 2009. I went through the history, but that article was not deleted/recreated. I couldnt see any activity that would have included it in the unreviewed list. Why are a few articles like it being listed in the feed? What am I missing? —usernamekiran (talk) 21:36, 4 December 2024 (UTC)[reply]

It was tagged as a "removed redirect" on November 2. Those are counted as new pages. ~WikiOriginal-9~ (talk) 22:05, 4 December 2024 (UTC)[reply]

I first noticed it a couple days ago. In the curation tool, under tagging, the feature of selecting subsections seems to have stopped working. I've seen it on two different PC's and two different browsers. Sincerely, North8000 (talk) 16:17, 8 December 2024 (UTC)[reply]

North8000. All fixed. Thanks for reporting. –Novem Linguae (talk) 21:34, 11 December 2024 (UTC)[reply]

Novem Linguae Cool. Just tried it, works great. North8000 (talk) 21:41, 11 December 2024 (UTC)[reply]

I'm on the edge about marking the above page as reviewed and could use a second pair of eyes on it. The article is heavily reliant on this self-published source, and I'm concerned that it contains original research. There does appear to be some academic coverage of the topic, for example this and this. However, the definitions of the "AI trust paradox" in these sources seem to diverge – one defines it as a situation "in which individuals’ willingness to use AI-enabled technologies exceeds their level of trust in these capabilities" while the other seems to define it as a situation where AI users "ignore all the risks due to the usability of fast intelligent systems". Another online source describes the AI trust paradox as a situation where humans are less willing to trust AI-generated text because they can't have a shared set of experiences with AI like they could with a human author. My impression is that while there are a few matches for the phrase "AI trust paradox" in sources, it isn't a well defined topic yet. It seems like it could be too soon for an article due to the rather hazy definition of the term, but I'd like to get input from some other reviewers. Lord Bolingbroke (talk) 19:38, 15 December 2024 (UTC)[reply]

thanks for post--Ozzie10aaaa (talk) 15:41, 20 December 2024 (UTC)[reply]

January 2025 Backlog Drive | New pages patrol
On 1 January 2025, a one-month backlog drive for new pages patrol will begin in hopes of addressing the growing backlog. Barnstars will be awarded based on the number of articles and redirects patrolled. Each article review will earn 1 point, while each redirect review will earn 0.2 points. Streak awards will be given out based on consistently hitting point thresholds for each week of the drive. Barnstars will also be granted for re-reviewing articles previously reviewed by other patrollers during the drive. Interested in taking part? Sign up here.
You're receiving this message because you are a new page patroller. To opt-out of future mailings, please remove yourself here.

MediaWiki message delivery (talk) 01:54, 18 December 2024 (UTC)[reply]

I reviewed Kasautii Zindagii Kay (2001 series) - Chapter 1 and Kasautii Zindagii Kay (2001 series) - Chapter 2 yesterday and tagged with single-source (each episode was linked back to an Indian streaming service), and left a note for the author saying that the URL for service was returning "Access denied" for me. They have now added a second streaming source to each entry as a reference. My question is - do these count as viable WP:RS? I'm thinking not, just as a spotify "page" for a song wouldn't count as a source. Thoughts? (And if this is in the wrong place, let me know and I'll ask at WP:RS/N. Cheers, Bastun^{Ėġáḍβáś₮ŭŃ!} 11:13, 19 December 2024 (UTC)[reply]

Hello, I am moving the chapters to draft already, the series already exist and having chapters without source is not neccesary. It's weird the editor kept on removing the tags too. I believe if the chapters not nominated for deletion, then dratifying them will be better Tesleemah (talk) 13:32, 19 December 2024 (UTC)[reply]

The second one’s source gave me a pop-up that my iPhone was hacked. Obviously not, and if I clicked on it I might have gotten hacked. Not an RS… UserMemer (chat) Tribs 13:41, 19 December 2024 (UTC)[reply]

Thanks :-) Bastun^{Ėġáḍβáś₮ŭŃ!} 11:52, 20 December 2024 (UTC)[reply]

I reviewed a page and left a message using the Page Curation tool. When it was added to the article's talk page, it worked correctly, but when it was sent to the user's talk page, it got cut off in the middle. My guess is that the vertical bar character somehow confused the program, so it failed to send the rest of the message to the user. – numbermaniac 13:22, 1 January 2025 (UTC)[reply]

WP:NPPHOUR says that we should wait at least one hour before marking a page for deletion unless there are serious content problems. However, WP:A1 and WP:A3 have a shorter delay; namely, 10 minutes. Which one should take priority in the setting of seemingly A1/A3-eligible pages? JJP_Master (she/they) 05:50, 3 January 2025 (UTC)[reply]

10 minutes is the baseline policy requirement. The one-hour standard is more along the lines of informal guidance for reviewers; it doesn't have the same level of community consensus behind it, and anecdotally I'd say it's pretty often not followed. Extraordinary Writ (talk) 06:28, 3 January 2025 (UTC)[reply]

The ten minutes is also just guidance (from the footnote to A1 and A3: there is no set time requirement, a ten-minute delay before tagging under this criterion is suggested as good practice). Use your judgement: the underlying message of both rules is to assess whether the creator could still be working on the article, and give them time to do so, before deleting it. – Joe (talk) 07:55, 3 January 2025 (UTC)[reply]

Yes, you're right. What I should have said is that the ten-minute suggestion is somewhat more widely accepted than NPPHOUR, which is a fairly recent addition. (But judging from the deletion log, many people don't follow either.) Extraordinary Writ (talk) 09:34, 3 January 2025 (UTC)[reply]

I'm not sure I agree regarding the respective levels of consensus. WP:NPPHOUR evolved from a "15 minute rule" that was added to NPP's guidelines in 2009, and before that general advice to patrol "from the bottom" of the log rather than brand new pages which has been there from the beginning. It was increased successively to the current hour following (IIRC unanimous) consensus in talk page discussions. I would say there is broad and long-standing community consensus that it should be followed, at least. – Joe (talk) 09:49, 3 January 2025 (UTC)[reply]

Well, unanimous consensus on a subpage most people don't watch. NPP can give guidance, but ultimately this is a deletion policy issue, and so far the community hasn't wanted to make changes on that front. At any rate, it's a philosophical question as long as no one follows the guidance; 9 of the last 10 A7s in the log were tagged within an hour, for example. Extraordinary Writ (talk) 10:45, 3 January 2025 (UTC)[reply]

To add, I actually do think waiting an hour is generally good advice. I just think it's worth being clear that it's not an idea that's gotten very much traction in practice. Extraordinary Writ (talk) 11:12, 3 January 2025 (UTC)[reply]

Just wanted to post a thing here. Not necessarily a "long discussion", or even if this is in the right location, but still something to get some perspective feedback on from New Page Reviewers who might see these in the queue. How do reviewers treat typos in redirects? I'm happy to patrol typoes in words that are likely to have typoes. But that's not really the focus of this section. The main reason that I came here was to ask "do you all treat typoes in a topic equally with typoes that are in a disambiguator?"

I'm probably going to ramble here but hopefully you all see the vision.

Because I think there's a major difference, generally speaking. I think that typoes in a topic are generally fine. But typoes in a parenthetical disambiguator I can't find a reason why that would ever be likely or okay. The parenthetical disambiguator is not a topic. It is something that Wikipedia adds to a title to give it a unique location in the encyclopedia. By all accounts in the MOS, is a technical entity. Because every page must have a unique title, a parenthesized term is never going to be the "topic". Like lets say there's a redirect titled Apple (fruit). Lets say that the two words were equally easy to spell, and equally plausible to misspell. If we were to compare the helpfulness of the two following redirects, I think that Aplpe (fruit) is more helpful of a redirect than Apple (friut). Misspellings in the topic, well, that's just always going to happen probably. It's something readers want to learn about. Typoes in the disambiguator are not likely. Disambiguators are a technical addition to pages and wouldn't be on a printed encyclopedia if it could be avoided. The topic is "apple", not "apple (fruit)". We should accommodate some likely typoes for "apple" if we really wanted but probably not because in this case "apple" is really easy to spell. Typoes on banana are a bit more likely and those can and should definitely be accommodated. But not typoes in the technical addition that Wikipedia requires is used to disambiguate pages.

I think there's one exception, and that is capitalized disambiguators. I think that Apple (fruit) and Apple (Fruit) are equally likely. My unpopular opinion is that I don't think two of these are useful, but that doesn't mean I don't think it's likely. That's a whole can of worms that I'd like to leave exempt from the discussion of typoes in disambiguators. Because I'm struggling to think of a single misspelled word in a disambiguator that can be helpful. Various disambiguators can be useful, such as Apple (food) if we really wanted it. But not misspelling the technical addition that Wikipedia requires we use to disambiguate different topics. Thoughts on this mindset? What do you all do, and do you patrol all typoes equally? Utopes _{(talk / cont)} 19:21, 8 January 2025 (UTC)[reply]

Is there a way to check which articles are placed into New pages feed the longest? A single page from 2017 days ago does not seem to be reviewed in any way and it was not created on that day. MimirIsSmart (talk) 04:40, 13 January 2025 (UTC)[reply]

Redirects that have been converted to articles are automatically added to the New Pages feed. Most of the "oldest" articles in the feed would be of this kind. -MPGuy2824 (talk) 05:52, 13 January 2025 (UTC)[reply]

Sort by oldest, then scroll backwards until you start seeing a bunch of unreviewed articles from the same month in 2024. That's the "real" back of the queue. –Novem Linguae (talk) 06:21, 13 January 2025 (UTC)[reply]

I already edited said page. Just had to limit newest page to 2017 days ago. MimirIsSmart (talk) 06:22, 13 January 2025 (UTC)[reply]

You can also set your "oldest" boundary date to exclude the very old ones that are always there as the result of redirect wars and/or vandalism. I and others have gotten almost all of January and February 2024 knocked out as part of the backlog drive (lot of AfD material left there) so I welcome fellow patrollers to join me in March 2024 :) Dclemens1971 (talk) 22:58, 13 January 2025 (UTC)[reply]

Deal me in for a March sprint or next one. I couldn't do the January for various work reasons. scope_creep^Talk 08:33, 21 January 2025 (UTC)[reply]

@Scope creep: We're currently on a fourth month cycle for backlog drives, so May is our next expected drive. Ideally we'd prefer only two a year, but we're doing what we can to keep the backlog under 10k. Hey man im josh (talk) 15:15, 22 January 2025 (UTC)[reply]

Hello. I'm Ampil. The backlog of unreviewed redirects is very low. 12 redirects are unreviewed. Please see Special:NewPagesFeed.

• July 2024. 30,000 redirects unreviewed.
• January 2025. 12 redirects unreviewed.

Thanks. ~🌀 Ampil ^{「💬 / 📝」} 07:34, 16 January 2025 (UTC)[reply]

Surely due to our current backlog drive. Woohoo! –Novem Linguae (talk) 09:26, 16 January 2025 (UTC)[reply]

I believe the last time the backlog was cleared entirely was the 2023 May backlog drive, so I'm grateful to see us get it cleared out again.

Now we just need some extra focus on the backlog of articles, which is currently far too large at ~12,500 more than halfway through the drive. We started at 16,260 unreviewed articles, and the change in that backlog is -3,775 (-23.22%), so I'd love to see some of our reviewers shift their focus there now. Hey man im josh (talk) 15:37, 16 January 2025 (UTC)[reply]

We've hit zero a few times; I tried to get a screenshot but can't find it now, so I guess we'll have to do it again for the historical record :). Also, I just realized there's a backlog drive, gotta go sign up and hope it counts retrospectively. Rusalkii (talk) 23:54, 17 January 2025 (UTC)[reply]

@Rusalkii: It does indeed count retroactively! Hey man im josh (talk) 01:15, 18 January 2025 (UTC)[reply]

I keep seeing articles get patrolled while having several fictitious references. Please make sure to click on every URL/ISBN/DOI that a new article cites – if you find multiple 404s, the article is likely AI-generated. If you see other AI indicators (e.g. **Markdown bold** or [Markdown links](https://example.org)), the article should be draftified or tagged with {{AI-generated}}. See WikiProject AI Cleanup for more advice/help. jlwoodwa (talk) 04:24, 19 January 2025 (UTC)[reply]

If not already mentioned, can this disclaimer of sorts be included in NPP guidelines and instructions? ~/Bunnypranav:<ping> 05:12, 19 January 2025 (UTC)[reply]

Might be too much detail for WP:NPP, but would be great for an essay such as Wikipedia:How to spot AI-generated text or something. But then again, that'd have to be counterbalanced with divulging too much information on this topic, the same way that we wouldn't want to necessarily make a list of tells for WP:UPE editing. –Novem Linguae (talk) 05:21, 19 January 2025 (UTC)[reply]

Slightly offtopick of fictuous refs, can completely AI generated looking articles be draftified before 1 hour. NPPHOUR exists as a chance for the author to improve the article, but entire AI articles can barely be improved. Of course the 10 min minimum can stay though ~/Bunnypranav:<ping> 05:31, 19 January 2025 (UTC)[reply]

Is there a common maintenance tag that can be added for when we see fictitious references? They will likely increase, so some clear tag or verbiage may help. It may already exist, though I have not seen it. FULBERT (talk) 08:37, 19 January 2025 (UTC)[reply]

I've seen a couple of these fictitious references as well. I thought it was just one instance. That is a pain. I was really puzzled the first time around. It was bizarre. At the moment I'm servicing the {{Category:Articles containing suspected AI-generated texts}} cat. Sub-cat on that maybe. I was suprised it doesn't take too long to learn to spot the stuff. It seems to be a particular way of writing, well, for the individual. You will spot it. Don't be afraid to remove it once you recognise it. Quality is what we are looking for. scope_creep^Talk 08:31, 21 January 2025 (UTC)[reply]

There is in fact {{AI-generated source}}, and reviewers may find other useful resources at AI cleanup. — ClaudineChionh (she/her · talk · contribs · email · global) 08:40, 21 January 2025 (UTC)[reply]

Useful info, thanks for sharing. Bastun^{Ėġáḍβáś₮ŭŃ!} 15:46, 21 January 2025 (UTC)[reply]

• Multiple 404s also occur in translated articles that have old references so it's not always a sign of AI, Atlantic306 (talk) 21:29, 22 January 2025 (UTC)[reply]

  This happens because to boost their creation count, some users pick older articles from other language Wikipedias, translate them (usually with a machine*), and don't bother to verify if the old foreign references are still valid - if indeed they can even read them. Furthermore, even good faith translators fail to understand that we are far stricter on sources than other Wikis - even the otherwise pragmatic de.Wiki. When these translations - most often unattributed - appear in the feed, one glance will indicate a need to verify an old source. Very often however, on other new articles (esp. BLP) there is a veritable refspam to make a fast working patroller believe the article is notable (it happens). Most of those sources are paraphrasing each other, or they are interviews, and none of them are in-depth biographical treatments. It's often a hallmark of an article desperately wanting to be published. With so many more unreliable websites being created, patrollers just have to check all the sources - and it's a real pain.
  *Machine translations of everyday topics are now extremely close to natural language, but always need post-editing by a bilingual or genuine very near native speaker of both the source and target languages. Such expertise is in short supply among the 822 New Page Patrollers. Kudpung กุดผึ้ง (talk) 10:01, 26 January 2025 (UTC)[reply]

I've not really been doing too much reviewing this month, but I did notice Joaquin Farias, which is a biography for an alternative medicine doctor who has invented their own therapy for a specific condition and are, I guess, getting some testimonials in the news or whatever. Is there like, any tag we can use for something like that? Should we just dump it on WT:MED or WP:FRINGEN's doorstep and let them deal with it (they're also mentioned on Blepharospasm and Focal dystonia so I suppose we'd probably want to do that regardless)? I haven't really looked at it in depth to see if there's anything we'd do through the usual processes (it was already reviewed by the time I was thinking of doing that) but either way I thought it might be worth wile hashing out some best practices for articles that pass the usual checks but might be concerning in other ways. Alpha3031 (t • c) 14:58, 28 January 2025 (UTC)[reply]

IMO the NPP question is "Is he wp:notable" and looking at the sources IMO the answer is yes. Sincerely, North8000 (talk) 20:02, 5 February 2025 (UTC)[reply]

I'm not entirely convinced of that either, but maintenance tagging is a part of the NPP review process even if it is an optional one. Alpha3031 (t • c) 10:10, 6 February 2025 (UTC)[reply]

 You are invited to join the discussion at Wikipedia talk:New pages patrol/Coordination § More frequent backlog drives. ~/Bunnypranav:<ping> 09:46, 1 February 2025 (UTC)[reply]

Could anyone give me guidance on whether I've misunderstood CSD G14? I recently tagged Terrace of the Leper King (disambiguation) for deletion since it only includes one blue link that includes the base page name. Pppery declined the speedy (see our brief discussion) since a play called The Terrace of the Leper King is mentioned at Yukio Mishima, even though it is a red link. Would G14 apply in this case? Lord Bolingbroke (talk) 23:03, 5 February 2025 (UTC)[reply]

No, I don't believe it comes within G14, but it should never have been created and has now been correctly PRODded. PamD 11:04, 6 February 2025 (UTC)[reply]

This might've been asked before, but is it okay if I accept an AfC draft and then mark it as patrolled after I accept? ~ Rusty ^meow ~ 07:02, 14 February 2025 (UTC)[reply]

Yes, that's allowed. Some people prefer to leave it for a second reviewer to check. (This gets asked every once in a while.) SilverLocust 💬 07:22, 14 February 2025 (UTC)[reply]

Thank you! ~ Rusty ^meow ~ 15:48, 14 February 2025 (UTC)[reply]

If you don't mark an accepted AfC draft as patrolled, it will appear under Wikipedia:New pages patrol/Reports/Easy reviews. Reconrabbit 17:03, 14 February 2025 (UTC)[reply]

When I do AFC's (which isn't that often) I almost always also do the NPP. I figure that if I've already spent the time to learn the situation on an article I might as well get that done vs. another NPP'er having to learn it again. North8000 (talk) 03:22, 15 February 2025 (UTC)[reply]

If a draft AfC is accepted by a NPPer they can also mark it as patrolled. If the person accepting the draft is autopatrolled then it is automatically marked as patrolled when moved to mainspace. It seems inconsistent that some AfC articles need one person to review them and others need two. As the criteria to accept an AfC draft and to mark an article as patrolled are pretty much the same, wouldn't it be better to have accepted AfC article automatically marked as patrolled? --John B123 (talk) 08:13, 15 February 2025 (UTC)[reply]

The fact that article reviews involve one patroller taking a look and draft reviews involve two patrollers (the AFC reviewer and an NPP) taking a look has led to a system where it is easier to get AFC reviewer permissions than NPP permissions, making AFC reviewer a good entry level permission for folks interested in this kind of patrolling. If we were to consolidate, and make a bot to mark all accepted drafts as NPP patrolled or something like that, we'd probably have to get rid of the AFC perm entirely and use the higher standards of the NPP perm. That'd be one downside. –Novem Linguae (talk) 12:49, 15 February 2025 (UTC)[reply]

Indeed, AfC reviewer, unlike NPR, is not a user group as set in the granularity of user accesses in the Wikipedia install of MediaWiki and created by official official consensus. AfC reviewers do not undergo the same level of vetting as NPPers who are the gatekeepers of Wikipedia quality. AfC provides the important function of helping the article creator and other eventually interested editors to improve potentially viable new articles to mainspace readiness, ...making AFC reviewer a good entry level permission for folks interested in this kind of patrolling. Kudpung กุดผึ้ง (talk) 17:27, 15 February 2025 (UTC)[reply]

Is there a (hidden) category that is auto-populated by items (articles/redirects) that are not marked as reviewed? I ask because I would like to do some slicing and dicing of the backlog with WP:Petscan, which needs a category as input. Cheers, SunloungerFrog (talk) 06:12, 18 February 2025 (UTC)[reply]

No, SunloungerFrog, there isn't. That information (reviewed or not) is stored in a table (pagetriage_page) that PetScan doesn't use. You should be able to use quarry to slice and dice though. -MPGuy2824 (talk) 09:03, 18 February 2025 (UTC)[reply]

MPGuy2824 thank you for that. Quarry was going to be my next port of call, though my SQL, never very good, is rather rusty. Cheers, SunloungerFrog (talk) 09:43, 18 February 2025 (UTC)[reply]

Here you go: https://quarry.wmcloud.org/query/90941 –Novem Linguae (talk) 17:39, 18 February 2025 (UTC)[reply]

Brilliant, thanks Novem Linguae! Cheers, SunloungerFrog (talk) 17:41, 18 February 2025 (UTC)[reply]

Hello. I finally managed to review Track 143 today - there are a couple of citations to add that I'm looking up at the moment, but I'm pretty happy that it's sufficiently notable given its awards. However, when I was checking one of the sources [1] - through Google Translate, as my Farsi is non-existent! - the publication date came up as Publication date: 22:41 - February 12, 2013 (my emphasis) and everywhere else (well, the article as written and IMDb) places the film being released in 2014. I'd be grateful for any thoughts / experience with quirks of date translation Farsi -> English. At the moment I have put down the source date as 12 February 2014. Cheers, SunloungerFrog (talk) 11:20, 25 February 2025 (UTC)[reply]

thank you for post--Ozzie10aaaa (talk) 12:54, 27 February 2025 (UTC)[reply]

I'd be grateful for a second opinion on Abhishek Anil Kapur. Subject seems to just about meet WP:NDIRECTOR The person has ... played a major role in co-creating a significant or well-known work ... [that is] the primary subject of multiple independent periodical articles or reviews, but it is a very thin article otherwise. I suppose it depends on whether Sky Force (film) is "significant or well-known" - if we think it's notable, because it has its own article, is that sufficient? Cheers SunloungerFrog (talk) 13:36, 3 March 2025 (UTC)[reply]

@SunloungerFrog From what I can see, it seems notable per the above criteria you mentioned. Since the person has been mentioned in a couple headlines, and the movie has good coverage, the director would be notable. I'm not quite sure why it was tagged with AI gen, since the contents seems good enough for a start rating as well. Just check if the first para of Career is true and sourced by the refs included. ~/Bunnypranav:<ping> 13:47, 3 March 2025 (UTC)[reply]

A recent comment in archive #51: Can anyone shed some light on why experienced reviewers seem to often leave pages unreviewed? Am I misunderstanding the criterion/decision-making for when a page should be marked reviewed? I seem to encounter quite a number of unreviewed pages where experienced NPP patrollers have clearly looked at the page, and even made improvements (adding categories, tagging, other improvements) and yet chosen not to mark the page reviewed [...].This is giving me pause, as it is making me questioning my judgement.

Another (an admin) wrote very recently: In my mind at least, a significant part of the problem is that reviewers feel like they will be chewed out for marking something as reviewed when they "shouldn't have". If we want reviewers to feel comfortable "lowering" their standards, we need to protect the reviewers who do so from blowback.[2]

And yet another admin writes: When doing NPP and coming across an article with lack of evidence of notability and/or inadequate referencing, one has the options of draftifying it, requesting speedy, or starting an AfD discussion [...] In my experience, if you draftify, someone will squeal and tell you in no uncertain terms you should have gone for deletion instead. And if you go for deletion, you will get heat for not draftifying. (Also, if you request speedy, you'll be told you should've gone for AfD, and v.v.) I think the technical term for this is "****ed if you do, ****ed if you don't".[3]

Having often received borderline PA flak for some very simple and cautious, friendly comments to new and older creators while reviewing, and having very obvious CSDs reverted I can understand why new and experienced reviewers (myself included) are often reluctant to review any articles that are not an easy pass. The problem is who then is going to review the pages we leave? IMO this is what causes the backlog and my attempts at working from the back of the queue confirms this. This means in effect that those pages that will eventually fall off the cliff are the very ones that should be stringently reviewed.

Over the years we have made enormous efforts to improve the tutorials and curation software, and even created the NPPSCHOOL and the user right, so the problem is not there. What are other reviewers' experiences?

• Have you taken blowback for your reviews?
• Does it discourage you from tagging and/or making for one of the deletion processes?
• Do you leave such articles for another reviewer?
• Do you have any other ideas what causes these backlogs? Do you have any suggestions?

Kudpung กุดผึ้ง (talk) 04:06, 4 March 2025 (UTC)[reply]

A few thoughts:

• During backlog drives, I tend to work from the rear of the queue so I see similar challenges. Those are harder articles, and at least half require a BEFORE search. Sometimes I will "take a break" by working from the front of the queue where I can log some easy passes or clear speedies. When I work from the back of the queue, I spend a lot of time and I usually nominate at least a quarter, maybe a third, for deletion. It takes time.
• However, at the front of the queue, I do sometimes tag an article but leave it unreviewed. The reason is usually that it was created by (a) a good-faith/naive newbie whom I don't want to discourage with an immediate draftification (since I know there are lots of sore feelings in this community about those) or an immediate AfD (which seems bitey), or (b) a very experienced editor who has nonetheless created a problematic article but to my mind has earned a chance to fix it on his or her own. Leaving these unreviewed (with appropriate talk page messages to the creator) hopefully triggers improvements before another reviewer takes a pass. If an editor is new but seems precocious I am much less likely to give extra time before taking action on an article that fails the standards for new page review.
• If there were some way to indicate this in the page curation tool, that might be helpful? Perhaps generate a message to other reviewers within a week of being tagged but seeing improvements that says in effect "a reviewer gave this page a chance to be fixed and nothing happened, so go to it."
• I haven't ever gotten blowback for my reviews (at least not that I recall).
• Admins seem pretty reluctant to accept G11s -- to the point I do virtually none. Early on as a reviewer what seemed uncontrovertibly promotional to me would be declined. If there was a more consistent and widely communicated standard for what admins would accept as a G11, I might do more, which would streamline the review process. Right now I mostly nominate cut and dry G5s and some G4s where I can be confident it's a match to an earlier deleted version (or where an admin undeletes some past revisions for me to check).

Hope this helps. Dclemens1971 (talk) 04:38, 4 March 2025 (UTC)[reply]

Re: G11: you won't get a clearly communicated standard for this because there isn't one. It's inherently subjective, and admins disagree on it amongst themselves. Personally, if I think it's at all plausible that it was created in good faith (as opposed to being intentionally promotional/spam), I strongly prefer draftification. If it's obvious AI slop I have no such hesitation. -- asilvering (talk) 07:57, 4 March 2025 (UTC)[reply]

I agree that "reviewers feel like they will be chewed out for marking something as reviewed when they 'shouldn't have'". When I make improvements to an article but don't review it, my thought process is something like "I'm not sure about this article, but why not improve it while I'm here?" ~ Rusty ^meow ~ 04:49, 4 March 2025 (UTC)[reply]

I am new to NPP, so still learning the ropes. My 2¢:

• I haven't yet had lots of squawking when I've draftified articles or nominated them for deletion. The odd slightly pointed comment but not more than that.
• I do, however, skim through a lot of unreviewed articles and think "gosh, should I do X with this article?" and what would be handy is way of getting the opinion of NPP colleagues. I don't think that the Curation tool readily allows that, and it is not feasible to open up a new thread for each article in question here. That's why I leave articles for another editor.
• Taking that point further, I wonder if there is scope for a script or some such that allows NPPers to
  • Mark an article with a comment (and suggested resolution for the article, be that AfD, draftify, tag and mark reviewed, etc.), that populates a page of "half reviewed" articles somewhere together with the comment.
  • Then others could look down that table and do the other half of the review, as it were.
  • Seems that that would help those half thoughts to be captured, and help other NPPers get the article over the line. Rather than having, say, ten editors look at an article and go "Hmmm...dunno about that one" and moving on, thus wasting the collective time we have spent on the article.
• I suppose I do perceive that enwiki sometimes tends to be quick to blame and slow to praise, which doesn't lead me to stick my neck out on an article on a topic that I personally don't care about. And unfortunately a lot of the newly created articles tend to be on subjects that I don't care about, so if there is doubt I am not inclined to spend time on a subject that is unfulfilling for me.

Cheers, SunloungerFrog (talk) 09:45, 4 March 2025 (UTC)[reply]

@SunloungerFrog, the NPP Discord server is a great place to ask those minor, specific questions that aren't important enough to open a thread about. But if you notice any general point that keeps coming up for you, I'm sure no one would mind if you opened a thread to ask. And you'll probably be helping out another new reviewer who has the same question and is too shy to ask. -- asilvering (talk) 13:22, 4 March 2025 (UTC)[reply]

@Asilvering, thanks, and Bunnypranav mentioned the same thing on my talk page. I'll check it out, and I guess it would be good for synchronous interaction. But I do wonder whether an asynchronous mechanism would also be useful.

For example, one might tag an article and say "I looked at sources 1 & 2, and they both support information in the article. Source 2 is sigcov, so counts towards notability". A future reviewer could then pick that up and look at, e.g. sources 3 & 4, and on that basis say "OK they're both sigcov and decent sources, so ultimately notable" and then mark it as reviewed. I think that that might be useful particularly when sources are in a language that is foreign to the NPPer and it is bothersome to pick through machine translations.

Or, to quote an putative example [4] from @Hey man im josh (and apols if I am taking this out of context) I'm uncomfortable putting my name behind something (aka marking it as reviewed) if I think the page is of low quality and doesn't credibly show that something could reasonably meet any of our notability guidelines. That's why I leave pages unreviewed personally. If I knew that a fellow NPPer thought an article was a bit dodgy, and I thought it was a bit dodgy too, that would help me push the button on sending it to draft or AfD.

Cheers, SunloungerFrog (talk) 13:45, 4 March 2025 (UTC)[reply]

Yes, I agree that would be very useful. My suggestion of the Discord was in no way intended as "we've already fixed that problem, shut up". -- asilvering (talk) 13:55, 4 March 2025 (UTC)[reply]