Wikipedia talk:Page Curation/2023 Moderator Tools project

Potential automation

In the past (on Phab), I have requested stats on how many (a)CSDs have been rejected by admins, and (b)were sent to AfD, and passed as keep, redirect, merge or delete or (c) kept and actually deleted at a later date. Redirects are neither cheap nor the answer for eliminating issues because redirects get deleted, and the redirected article ends up in namespace again. Another stat that may prove useful is the number of unsourced articles and stubs in the WP corpus of articles with creation dates and tags that are at least 3 years old still sitting in namespace with no improvement.

I'm of the mind that NPP reviewers who meet certain qualifications, such as DYK/GA/FA promotions under their belt, attended NPP school and passed the course, and did good reviewing work for at least 3 months (the initial trial period after graduation) should be given more tools to work with, some of which are in the admin bundle. To make the unbundling of certain admin tools more acceptable, we could establish a NPP notification system that lists CSD/Prod decisions by reviewers, and requires review/approval of the action by at least 2 other NPP graduates/teachers. That way, we are dealing only with editors who know the system well, do the BEFORE properly, and are less likely to create an unnecessary time sink or mistake (and will also help keep the backlogs at bay).

The tools I have in mind to make the reviewing experience more efficient (while also eliminating some of the frustration for reviewers and at the same time, lightening the load for admins) include:

1. the ability to delete and draft articles (content creators, experienced article reviewers, and editors who are NPP school grads are fine-tuned to the BEFORE process & WP:PAGs, and are qualified to use this tool);
2. the ability to review deleted articles,
3. have full access to WPL TWL resources for the WP:BEFORE process,
4. have access to a specific set of tags (some already incorporated in curation tools) used most often by NPP reviewers (all the page, section, inline improvement tags), with common sense reasons (& ability to customize) for those articles moved to draft space, CSD'd or Prodded.

I believe this will help to eliminate many of the issues we deal with on a daily basis, and may even cut down on UPE, and possibly even WP:POV creep relative to contentious topics. Atsme 💬 📧 18:46, 18 January 2023 (UTC)

@Atsme Thanks for sharing your ideas. I'm curious about the motivation for the data you mentioned, can you elaborate on how that data would be helpful to you? We're also interested in collecting and analysing some more data about NPP so this might be something we could look into.

Unbundling certain admin rights to other user groups is a perennial proposal here. The most common reason for rejecting the suggestion, as I understand it, is that if we trust users to perform these actions we should simply give them administrator rights. Reading between the lines of your comments it sounds like you're seeing a problem where NPP reviewers tag/report articles, but those reports are then incorrectly declined by administrators, and you think that process needs review by patrollers. Is that correct, or am I misreading?

The list of suggestions is definitely helpful and I can see how those features would assist patrollers. I think points 1 and 2 really fall under the perennial community consensus issue I mentioned above - those changes could be made quite easily but would require a community-wide RfC.

In point 3 I assume when you write WPL you mean The Wikipedia Library? I believe all enwiki patrollers should be active enough that they are able to access the library no problem, but if you know of any example users who are patrollers but don't meet that threshold I'd love to know about that.

4 - Can you elaborate what you mean by 'access to a specific set of tags'? I'm not sure I understand the request on this point. Samwalton9 (WMF) (talk) 10:13, 19 January 2023 (UTC)

1 & 2 - left for last. Re: perennial proposals - perhaps it's time to IAR and try something new because our numbers are shrinking

3 - Yes TWL - but I do recall something recent about being granted full access without having to wait for a slot to come open (some sources were restricted and were available only to a limited number). If there is still a limit, I was of the mind that NPP reviewers should get first consideration because of the work they are entrusted to do;

4 - Tags - reviewers don't need a list of every single tag available, as it is time consuming. The tags we need should focus on what we are looking for in articles per the flow chart – but not a top priority.

Back to 1 & 2 - Full administrator rights – not needed because full rights include blocking editors, and why I proposed unbundling only the article-related tools including draftify, move, delete/undelete, review deleted). NPP reviewers need those tools. Consider some of the reasons behind arguments to keep failed articles, and how the stats I've requested will shine a lot more light on the pros/cons. Consider the potential motivation behind anonymous editors dropping loads of unsourced stubs on WP, and inundating us with non-notable articles and hard-to-detect hoaxes. We need to consider cost vs benefit, and right now the cost is a shrinking admin line-up, backlogs at AfD, and too few active (qualified) NPP reviewers.

Surely UPE, COI and advocacies come to mind when considering why anyone would vote to keep an unacceptable article in namespace, especially one that is pure promo, OR, unsourced, or non-notable. Admins have blocking rights so they can control vandalism and highly disruptive behavior, which has nothing to do with content. Sam, community trust in an editor elected to be an admin does not automatically make that admin qualified to judge content as an NPP reviewer. What I am proposing is strictly for content editors who are trained volunteers, some of whom can certainly be admins, but I'm of the mind that having all the tools is not necessary, and may even shade a reviewer's judgment, inadvertently or otherwise. DYK/GA/FA promoters are a good place to start. We should also up the average requirements for new editors who want to be NPP reviewers. There is no denying that AfC & NPP may attract UPE, advocates, etc. Why not let the newer editors start at AfC and graduate to NPP over time, once they have proven their ability to review. NPP is the next step after AfC, so we can see what comes down the pipeline from the new AfC reviewers.

Following are some diffs that further demonstrate the need, and support my position. I want to emphasize that I understand why admins are hesitant to CSD. They neither have the time (and some lack the article review experience/training) to be reviewers which requires BEFORE, so it makes no sense for a single admin to have the last word which often leads to time sinks, & frustration for reviewers forced to AfD obvious speedy deletes. I don't think requiring admins to do a BEFORE review prior to rejecting an AfD will resolve the problem and may result in fewer admins responding to CSD. Some admins approach CSD based on face value (see #3 below), whereas our qualified reviewers (graduates, longterm veterans) already did the BEFORE work so their decision should hold weight. Refer back to my proposal, the training, and the precautionary measures if content decisions are left to content reviewers, rather than conduct/vandalism admins who are already overworked, or may not be trained in or have the time to conduct the full review process:

1. Barbara Dawson
2. Campbell Dickens
3. Only consider article, so no BEFORE is done as with NPP reviewers
4. Frustration
5. Disconcerting
6. Wikipedia_talk:New_pages_patrol/Reviewers/Archive_15#Thoughts_about_AfD Thoughts about AfD
7. Pre-poll
8. Moving to Draft
9. Majority keeps are from socks
10. Good info
11. Article experience requirement

Thank you for taking the time to consider this proposal, Sam. It is much appreciated. Atsme 💬 📧 13:28, 19 January 2023 (UTC)

@Atsme Thanks for elaborating, I have a clearer sense of the problems you're seeing here now. In terms of administrator right unbundling, as I said earlier this would require a community RfC, I don't think there's any way we could make that change and not face widespread pushback, even if (especially if) we invoked IAR :)

Just to unpack the problems you elaborate on, though, am I right in thinking that - in your view - there are a substantial number of administrators who don't have a good grasp of deletion policies and therefore make bad decisions when NPP reviewers tag articles?

In the examples you linked it seems like some of the tension arises when an NPP reviewer has put in a substantial amount of work to review an article, tags it for deletion through one mechanism or another, and then that deletion is declined or opposed based on what the patroller perceives to be a less thorough review. Is that correct? Samwalton9 (WMF) (talk) 14:13, 19 January 2023 (UTC)

Yes, but it's not just what we perceive it to be, Sam. WYSIWYG. In the example I provided, the admin admits to judging the article as it sits; i.e. face value, and did not conduct WP:BEFORE. They do not have the time, which is why we have NPP. It's too easy for an inundated admin to just say no unless it is blatantly obvious promotion or a hoax. This is how all the hoaxes have gotten through - and here is a classic example. It's embarrassing. I am not saying that NPP is perfect, but we are one step above in reviewing and conducting WP:BEFORE. See this AfD which is clearly a snow keep, yet, it is going another 7 days because it was relisted after the first 7. This is typical of an AfD time sink, and it doesn't matter if it is done in GF - the road to hell is paved with good intentions.^[stretch] This is why we have trouble keeping NPP reviewers, and end up with garbage articles which tend to attract more garbage. Atsme 💬 📧 14:48, 19 January 2023 (UTC)

I see what you mean @Atsme, and didn't mean to sound dismissive! Do you think there are any software changes that could help with this? I'm thinking, for example, if speedy deletion tags contained additional contextual information so you could communicate with the administrator who is judging the tag. This might not be a software-solvable problem though, it sounds to me like deletion policy would need to be changed if we wanted admins to spend more time investigating an article before judging a speedy deletion tag. Samwalton9 (WMF) (talk) 12:07, 23 January 2023 (UTC)

Sam, if we changed the deletion policy to the latter, the likely outcome is that the community would support a BEFORE requirement, but admins would not. Even if it passed, the chances an admin will actually conduct BEFORE is impossible to verify. If we had the stats for comparison purposes, you will probably see little to no change in CSD rejections that end up as AfD deletions. I don't see it as a win-win for WP. If qualified NPP reviewers had the deletion tools, you will likely see the same number of deletions but fewer AfDs, reduced backlogs, and far less garbage in the encyclopedia. Over time, we will start seeing higher quality articles at AfC - it's the law of averages. Garbage articles are an incredible drain on almost all of our resources, particularly volunteers ranging from wikiGnomes on up to admins. Atsme 💬 📧 16:50, 23 January 2023 (UTC)

Thanks for the ideas. I don't think the community would have much appetite for giving NPPs administrator-like permissions though. And the proposed checks and balances would add a lot of complexity to existing processes. If the goal is to enable NPPs to help with admin backlogs relevant to NPP, then perhaps the simpler solution is for some NPPs to become admins. Hope this helps. –Novem Linguae (talk) 17:59, 19 January 2023 (UTC)

No, actually, it doesn't help. Did you read what I wrote about deletion? What are you thinking, NL? Are you the only one who is not faced with admin decisions that create extra work? I don't understand your position. Why would you not want to unbundle only the delete/review/undelete? Atsme 💬 📧 19:05, 19 January 2023 (UTC)

Harsh. I'm just trying to be practical. If I don't think something can pass the giant village pump RFC that would be needed to implement it, I try to point it out early. Those RFCs can be rough and eat up a lot of time. –Novem Linguae (talk) 17:25, 20 January 2023 (UTC)

My apologies, NL. That was stress of the day channeling through my keyboard. Just the thought of RfCs and consensus is frustrating, so I won't say anymore because in this particular situation, I view the hegemony of the blankity consensus as useful as an anchor on a sinking boat. I also believe there are times when the WMF should step up and do what is best for the community, even when the community cannot see what is best for themselves. I'm taking a break from NPP. Good luck! Atsme 💬 📧 03:20, 22 January 2023 (UTC)

Comments: @Atsme For the WTL you do not need an NPP right. I had access before I was a NPP member and I am pretty sure any active NPP member will be granted access with a founded argument. But I believe not many NPP members will feel the need for the WTL, as what can be found in the WTL can likely be found on google as well. So access could be granted only to the ones who apply. But I understand your suggestion for some sort of automation.Paradise Chronicle (talk) 19:53, 19 January 2023 (UTC)

Okay, here's my .02 (although that may be due to inflation). When I read Atsme's initial comments, I thought almost along the exact same lines as Sam, that we should simply have NPPers apply for adminship. However, there are several problems with that, not the least of which was exhibited during the most recent RfA. Anyone who does significant work at NPP is going to have ruffled some feathers. And I don't think that many would want to go through the process. But then Atsme expanded on which rights would not be included in the new (what I will call) NPP right. Their 4 points above would certainly be quite helpful.

Regarding 1 & 2 are related. Regarding speedies, the vast majority of the ones I propose are G4, G5, G11, and G12. Occasionally, I nominate an A7 or A9. Very rarely do I use any of the others, some, like A2, G3 and G9 I've never used. The two I have the worst record with are G11 and G4. But G4 is due to the fact that I cannot access deleted articles. The only reason I get denied at G4 is because I cannot compare with the deleted version. G11 is a different issue. For example, there is an admin who has never approved a single G11 that I've proposed. Now, I understand that there will be differences of opinion, but that's simply statistically impossible. Regarding Prods, there are several editors who make it their life's work to remove prods. Most of these prod removals simply add work to many other editors as they go to AfD, yadda yadda, and end up deleted. But there is at least one editor, who is just so irritating, but I have to admit, at least half of the time they remove my prods, they improve the articles the articles to the point where they pass one of the SNG's. So I think that the prod process should remain as it is. Regarding AfD's, about two years ago I began to keep track on AfD's that I nominated. Right now, about 35-40% of the ones I nominate are kept/merged/draftified. That's not a great %, but fully 60-75% of the ones I nominate are simply because that is the only avenue I have left to me. I've usually redirect and/or draftified, and/or prodded those articles before taking them to AfD. Additionally, probably 20% are closed as kept simply on a !vote count, rather than policy based. Wikipedia:Articles for deletion/Ali Hajizade is a perfect example, although there are many others. So I agree that many admins do not look at the policy when deciding these things. I have to admit, I didn't understand what WPL referred to until Sam responded to Atsme. I'm blessed to belong to JSTOR and Newspapers.com, but those are the only 2 I belong to. And the last time my Newspapers.com subscription ran out, it took me over a year to get it re-instated. And I use that a lot. So granting full access to WPL to folks with the NPP right would make a lot of sense.

All that being said, I believe this is an exercise in futility. With the recent RfA, and the sort-of recent discussion regarding WP:DRAFTIFY, I do not see any of these ideas succeeding. But then again, there's always Cervantes. Onel5969 ^{TT me} 02:30, 20 January 2023 (UTC)

I've been looking for the right place to make this observation and I think this is it. I passed RfA because of NPP. TonyBallioni and Joe Roe passed RfA because of NPP. And that's just to name 3 functionaries off the top of my head, but there are others. I'm sad that we lost MB as an editor because of that RfA. And I think that RfA was as much about the disconnect in the community about NPP expectations as it was MB. I also don't think passing RfA necessarily means that person has the skill to be an excellent NPP'er (or an excellent Wikipedian as there are many excellent non-admin Wikipedians and some admin who maybe don't quite reach the level of excellence in my view). But taking what happend to MB and saying "you can't become an admin if you do NPP" means some number of editors who could benefit from the admin toolset but who are committed to NPP may never even try. And that ultimately would hurt the NPP community. Best, Barkeep49 (talk) 02:48, 20 January 2023 (UTC)

I agree with Paradise Chronicle. The third suggestion isn't necessary. The criteria of TWL is 500+ edits, 6+ months editing, 10+ edits in the last month, No active blocks practically IMHO would apply to any active WP:NPPer. Note that NPP has a criteria of 500 edits in the mainspace as well and no blocks in past 6 months, which is more stringent. Of the 500 here (listed by username alphabetical order), 496 are extended confirmed (of the remaining four two are bots) and all have 6+ months editing (first one checked per my command F search for extended confirmed, 2nd one per my check if there are any post July 2022 accounts). So only 2 out of 500 checked are not eligible for TWL. There are lots of suggestions and improvements needed at WT:TWL including some subscriptions that are not working, but practically all NPPs are eligible. Thanks. VickKiang (talk) 06:47, 20 January 2023 (UTC)

Thanks for looking into this @VickKiang - I suspected it was the case :) @Atsme raises a fair point about the collections for which you need to apply, but we're trying to move as many of those as possible out of applications and into the default bundle, which now includes a substantial majority of all the library's content. That includes Newspapers.com (soon, hopefully: T322916). Samwalton9 (WMF) (talk) 10:35, 20 January 2023 (UTC)

And down the drain it goes.

Sam, one more point and I'll move along. Quite some time ago I researched the 500 edit criteria, and discovered that chalking up 500 edits is as easy as adding wikilinks, and fixing punctuation marks. AWB helps chalk-up edits as well. We look to see the number of edits a candidate has racked up, but rarely look to see how they were acquired. PERM is a good safeguard because experienced NPP reviewers who are also admins know what to look for in a candidate. I hold Barkeep and Rosguill in high regard in that department, but we should also consider the weight of the tasks we're putting on the qualified few, and how long will it last–including you. There is not an endless supply. The problems I've mentioned (what I consider problematic) create time sinks that are fixable. When looking at the full picture, well, a time sink = time wasted. Maybe as we grow older, our perspective on time has increased in value compared to our invincibility in our younger years. There is also "out of sight, out of mind" so we are prone to not think about all the unsourced stubs and problem articles. If we did, it would drive us '"out of mind". The stats I mentioned earlier would be quite useful, and will shine some much needed light on these concerns, good or bad. Checking the stats for six million plus articles requires automation, and there are more coming at us as the popularity of AI increases. Atsme 💬 📧 12:15, 20 January 2023 (UTC)

A wiki-agnostic version would be great.

Thanks for these udpates.

Re: the idea of unbundling: the specific point that deletion/undeletion and viewing deleted revisions pages should be accessible to many more people than blocking/unblocking users, seems worth a general RfC. That is relevant to NPP and matches how much more reversible a deletion is. – SJ + 00:58, 20 January 2023 (UTC)

+1, general RfC might face opposition given the local tally at Wikipedia talk:New pages patrol/Reviewers/Archive 46#More user rights for NPP and the WP:UNBUNDLE proposals mostly being unsuccessful (though I agree partially, e.g., regarding viewing deleted versions to evaluate G4s). VickKiang (talk) 10:40, 20 January 2023 (UTC)

I'd support viewing deleted articles. Articles go through the AfD and CSD process all the time and they're sometimes recreated later on. We're aware that they were previously deleted, but we don't know the new article is the exact same as the one that was deleted a few years ago. I think it could be a very useful tool that would help save time and effort. I'm not sure if they're separate rights, but I'm not too keen on the idea of viewing deleted revisions. If a revision is deleted it's usually deleted for a good reason and I don't see it being relevant to the NPP process. I think of it like a Judge ordering a Jury not to consider something that was brought up in court in their decision making process. What could be in that deleted revision

There are a lot of users outside of NPP that could benefit from being able to review deleted pages as well. I think a side effect of this, if implemented, would be some users applying for the NPP perm with no intention of patrolling pages. Hey man im josh (talk) 13:29, 20 January 2023 (UTC)

I meant pages, indeed! Corrected above. – SJ + 16:44, 20 January 2023 (UTC)

Only editors who graduated from NPPSchool will get the right...at least, that is how it should be. And if they become inactive at NPP, they lose those rights.Atsme 💬 📧 15:49, 20 January 2023 (UTC)

If I recall correctly, unbundling viewing deleted content from the admin tools was vetoed by WMF Legal. In court, they want to be able to argue that deleted content, including libel and BLP violations, is truly deleted except for a handful of highly screened individuals that went through a tough community process. This is documented somewhere if someone wants to find it. I remember reading about it. –Novem Linguae (talk) 17:18, 20 January 2023 (UTC)

I'm checking with Legal to see if we have some documentation on this because while I'm pretty sure this is right I'm struggling to find it written down anywhere. Samwalton9 (WMF) (talk) 12:51, 23 January 2023 (UTC)

First, graduating from NPP school IMO is a too low bar. Second, I'm also being realistic here that given the community being already split on NPP practices including WP:DRAFTIFY and WP:BEFORE, and that only modest unbundling proposals have succeeded, drafters of a potential RfC should be prepared of significant opposition, despite some of these proposals being helpful (like what Onel5969 said). Thanks. VickKiang (talk) 20:12, 20 January 2023 (UTC)

The potential harm of looking at deleted content with WP:HARASSMENT is difficult to monitor in general, so while I couldn’t care less about a user looking at REVDELETED copyright content, this gives me pause. A stricter criteria would be in order than merely a successful NPP graduation. Does the candidate meet WP:ADMINCONDUCT like conduct is what I’d look for in supporting such access. ~ 🦝 Shushugah (he/him • talk) 17:23, 21 January 2023 (UTC)

I've posted a notification at WT:NPPR related to this discussion and potential feedback about this moderator tools project in general. Thanks. VickKiang (talk) 02:09, 22 January 2023 (UTC)

Patrolling interviews

Hi all. Although our work on PageTriage isn't going to be getting underway for a little while yet, we are beginning some groundwork. One piece of that is making sure that I understand the NPP process. To that end I'd like to meet with a few folks to ask some questions and to record a screenshare of you patrolling articles. This will give me a chance to learn what it's like patrolling articles on enwiki today using PageTriage, start to see what the pain points are, and ensure that I'm not misrepresenting the process. Although I've done patrolling work myself and have read most of the help and process pages, nothing compares to sitting down and watching someone else edit!

I already asked on the NPP Discord and have a few calls lined up, but I wanted to post here too to see if I can meet with a few more patrollers. If you're interested in joining me for a 30-60 minute call please let me know - I have a release form you'd need to sign if you're happy with screen recording via Google Meet. Alternatively, if you're not comfortable meeting 'face-to-face', as it were, your own screen recordings uploaded to Commons would also be very valuable to watch. Let me know if you have any questions! Samwalton9 (WMF) (talk) 11:35, 20 January 2023 (UTC)

The latest in NPP tools for reviewing articles.

Thanks for taking the time to do this. I'd be very interested in a summary of the things you learn from watching all these folks, both from a PageTriage perspective (how can we improve the software?) and from an NPP perspective (what are people's mental flowcharts/strategy like when patrolling?). There's lots to be learned from this kind of thing. So feel free to post anything you learn. –Novem Linguae (talk) 19:32, 27 January 2023 (UTC)

@Novem Linguae I definitely plan to write up a summary once we've had a chance to talk to a few more folks and spent a little time synthesising what we learned! Samwalton9 (WMF) (talk) 00:35, 28 January 2023 (UTC)

2023 research didn't find any PageTriage bugs

In terms of PageTriage, we found that patrollers generally didn’t run into significant issues using the extension. During our interviews no patrollers encountered flow-breaking bugs Just a quick note so you know the back story: at the time we wrote the open letter, PageTriage had a bunch of super annoying bugs that were driving patrollers to use Twinkle for everything except the "mark as reviewed" button. Early on we envisioned WMF fixing these bugs, but volunteers such as myself and MPGuy2824 ended up stepping up ourselves and fixing them. They included the prod bug, a couple of AFD bugs, and some others. I am happy that these are fixed, since they clear the way for other work. But they were a huge part of our motivation for writing the open letter at the time. I guess I'm writing this so you guys don't feel bamboozled about the software not being very buggy right now. Hope this helps. –Novem Linguae (talk) 16:49, 27 February 2023 (UTC)

@Novem Linguae That totally makes sense, and I'd pieced together that this is probably what happened. I've made a specific note in the research about this. I think this is absolutely credit to you and MPGuy though - there were relatively few complaints about issues with the software now :) Samwalton9 (WMF) (talk) 18:03, 27 February 2023 (UTC)

@Samwalton9: it's also possible that once a reviewer gets some experience on the software, they find (and smoothly use) workarounds. e.g. in "Priority bugs section" of the PT workboard, I face a few regularly and apply the workarounds, without even thinking much about it. It's possible that other reviewers might be avoiding some of the toolbar modules entirely, due to past buggy behaviour. -MPGuy2824 (talk) 03:32, 28 February 2023 (UTC)

Becoming a watcher or member of PageTriage group on Phab

Folks from the Moderator Tools team (Sam?) may want to add themselves as watchers and/or members to the PageTriage group on Phab. For now I am just adding Sam to various tickets I think might be relevant to this project. If folks become watchers, they'd see these tickets automatically. If not, no worries. –Novem Linguae (talk) 16:52, 27 February 2023 (UTC)

@Novem Linguae Thanks for the reminder, I'd been considering doing this. I didn't want to do it too far out from us getting into the weeds with the project but now seems as good a time as any! Watching. Samwalton9 (WMF) (talk) 18:03, 27 February 2023 (UTC)

Another resource for source identification

Given that the discussion of the results includes significant discussion of the difficulties in identifying reliable sources, it may be worth noting that we do have WP:NPPSG as a somewhat obscure resource to log the history of Wikipedia discussions about the reliability of specific sources. I've been the main editor maintaining the page--in distinction from WP:RSP, which is used primarily to head off repetitive and redundant discussions about high profile sources, NPPSG is focused on providing a starting point for discussion based on the community's last recorded position on a source, and is focused on whether a source's coverage can demonstrate notability. _signed, Rosguill ^talk 17:33, 2 March 2023 (UTC)

April 17 project update

I've just posted an update now that our work is underway in earnest. Please let me know if you have any thoughts or questions! Samwalton9 (WMF) (talk) 14:59, 17 April 2023 (UTC)

I'm very excited by the priorities identified in the update. Best, Barkeep49 (talk) 15:23, 17 April 2023 (UTC)

May 1 update

Makes sense and thank you! Even if people strongly dislike the design, standardizing it with MediaWiki makes sense visually but also from technical pov. While refactoring it will be worthwhile to check what bugs are accidentally solved/versus newly created bugs if we’re not emphasizing adding test coverage (I agree doing it twice doesn’t make sense perse). ~ 🦝 Shushugah (he/him • talk) 21:17, 1 May 2023 (UTC)

Help us test a NewPagesFeed update!

As part of our project to improve the technical state of PageTriage, our team has been hard at work upgrading the code that powers PageTriage. Rather than replace the current experience straight away, we've set up a testing venue to gather feedback and squash bugs first.

Per the update I just posted, please use this link to test out a new version of NewPagesFeed! The goal is that although the UI has changed a little, the experience of using the new feed should not be substantially different - full details can be found in the linked update. Bugs and feedback can be reported in this thread or on Phabricator. Thanks for your help with this! Samwalton9 (WMF) (talk) 11:23, 21 July 2023 (UTC)

Pinging folks who have engaged above to make sure you're aware! @Shushugah, Barkeep49, Rosguill, VickKiang, Hey man im josh, Atsme, and Paradise Chronicle: Samwalton9 (WMF) (talk) 11:27, 21 July 2023 (UTC)

Already on it. Paradise Chronicle (talk) 11:29, 21 July 2023 (UTC)

I'm testing it out now and I'm making some notes. Hey man im josh (talk) 13:42, 21 July 2023 (UTC)

Samwalton9 - THANKYOU for your hard work and dedication. I will do my part! Atsme 💬 📧 14:05, 21 July 2023 (UTC)

Not responsive

I'm a bit surprised that the new new pages feed is still completely unusable on mobile. Shouldn't that have been one of the benefits of switching to Codex? – Joe (talk) 06:25, 28 September 2023 (UTC)

@Joe Roe (mobile) Mobile responsiveness is something I personally would have liked to improve. However when we were scoping the project and getting feedback (see subtasks of T324914) a big thing we were hearing was to not change the UI or its behaviour from the current design, including maintaining the layout at arbitrary zoom levels and viewport sizes. As such we spent a fair amount of effort ensuring that the UI did not change despite switching some components to Codex (the main table and filter menu are not using Codex, just migrated to Vue.js, so that wasn't related to Codex anyway). Samwalton9 (WMF) (talk) 12:32, 29 September 2023 (UTC)

That's a shame. The current UI isn't actually very good... – Joe (talk) 12:58, 29 September 2023 (UTC)

I went ahead and created a ticket to explore changing the UI: phab:T347732. –Novem Linguae (talk) 16:03, 29 September 2023 (UTC)

New NewPagesFeed will be the default from Thursday 19th October

A quick heads up that the default NewPagesFeed interface will change from the old version to the new one tomorrow (Thursday 19th October). We've done a lot of testing and the only change should be some minor spacing/styling differences. That said, if you run into any issues please let me know or file a task on Phabricator, and you can access the old version of the UI at Special:NewPagesFeed?pagetriage_ui=old. We'll give it a few weeks and then remove the old interface if all seems well. Samwalton9 (WMF) (talk) 11:42, 18 October 2023 (UTC)

Just a quick update that this is now live. We're already aware of a couple of issues, please let me know if you find more: T349339, T349338. Samwalton9 (WMF) (talk) 18:41, 19 October 2023 (UTC)

Quick question about PageTriage

Hi @Samwalton9 (WMF), PageTriage can now support wikis other than English Wikipedia? (Example) —MdsShakil (talk) 11:36, 24 October 2023 (UTC)

Hi @MdsShakil - unfortunately we didn't have time to work on this side of things, so PageTriage is still English Wikipedia-only at the moment. Work towards this goal is tracked in T50552 and its subtasks. Samwalton9 (WMF) (talk) 11:41, 24 October 2023 (UTC)