This is an archive of past discussions on Wikipedia:New pages patrol. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
Latest comment: 6 years ago6 comments4 people in discussion
I have come across a few list articles that only contain a couple of redlinked entries no bluelinked ones and no sources. I originally prodded them as for me they we're in contradiction with the WP:CSC for stand alone lists. At the top of the SETINDEX page it states that the guidelines for the articles are the stand alone list guidelines. They were unprodded and I entered into a lengthy discussion where I was told that it was perfectly acceptable for a particular project to have these kind of articles. I wanted to know if it is common practice to allow all Set index articles to follow this format or not. I do not want to get into another heated discussion with project members for nothing and get told that I should not have prodded them without asking the project if they were ok with it first. Could anyone advise me ?Dom from Paris (talk) 23:57, 24 April 2018 (UTC)
Hi, can you link the articles and the discussion? Red-link lists are almost always nuked. As a side note, there's no point in debating that every editor including NPRs can unilateraly act/(propose deletion of) on any page, without taking advice from concerned projects etc., though such advice may be sought.~ Winged BladesGodric13:16, 27 April 2018 (UTC)
Red links have a purpose (I'm currently working on a list at Cardinal electors for the papal conclave, March 1605 that has plenty of red links that should be blue, but that has more to do with historical coverage in English sourcing than anything else...) The issue is that if someone creates an article that is entirely redlinks, it'd often be better to have it in project space as a list that a project wants to work on, etc. in my opinion. TonyBallioni (talk) 13:23, 27 April 2018 (UTC)
WP:CSC is really not a prescriptive or exhaustive guideline — "Lists are commonly written to satisfy one of the following sets of criteria" — however what appears to be the SIA in question now meets all criteria of WP:CSC #3, and one of its entries is now bluelinked. ~Hydronium~Hydroxide~(Talk)~13:57, 27 April 2018 (UTC)
Thanks for the advice. Here is the talk page Talk:Italian_ship_Uragano, the discussion got a bit ad hominen concerning what was a good faith but with hindsight ill advised deletion nomination that I made a few months ago and then walked away from. Seeing as the person who brought that up was an admin and was also the one who told me I should have asked if the project felt it was OK to keep it and if they said yes leave it at that and also told me to "drop the stick" I thought it best to seek advice. As it happens some of the project members turned up and added the sources to the page and then created a stub for one of the articles and the article now meets CSC. What really posed a problem for me was the fact that there was no way of seeing if the redlinked entries were legit and that a very experienced editor answered the question "So unsourced lists of redlinked entries are ok by you?" with "So long as they're ship index pages, sure." Dom from Paris (talk) 17:21, 27 April 2018 (UTC)
Permanent autoconfirmed page creation limitation is now live
Latest comment: 6 years ago8 comments8 people in discussion
Mwahaha. One step closer to dominating the global information trade by suppressing the editorial masses. - Hey, it's nice and quiet all of a sudden... :) --Elmidae (talk · contribs) 07:52, 27 April 2018 (UTC)
Daniel Sirera (blp sources, blp unreferenced, 4 cn tags)
I have only gone back about 6 days in checking pages, so there's probably more tha I didn't see yet. I'd also like to note that many of their additions of tags occur in seperate edits just a few seconds apart, which I found odd. Before commenting here, please see the discussion on their talk page. I am not here to explicitly recommend anything be done, as this is not my area of knowledge; I'd just like to bring this to the attention of involved sysops and coordinators. Pinging primefac who accepted them to the AfC WikiProject. Thank you, and happy editing. Vermont (talk) 10:24, 26 April 2018 (UTC)
Notified of this via IRC, and having poked through the diffs over coffee, I'd say that overtagging drafts I'm generally okay with, especially unsourced BLP info, since well, that's kindof the point of draft space, to fix everything up before publishing. But I'm not sure if you realize or not Cass, but a single unreferenced tag at the top of the article covers the whole article and should generally not be used in conjunction with other tags. Similarly, an unreferenced section tag covers the whole section, when the whole section is unreferenced, and shouldn't be used in conjunction with the next lower level tag, {{cn}}. Then, cn should only be used when there is usually one or two lines that are unsourced but everything else is sourced. It seems that if we can clear up that bit it solves most of the problem here unless I'm mistaken. GMGtalk10:55, 26 April 2018 (UTC)
I think for things like Draft:Adam Payne (Cyclist), where the awards are what seem to be defining the notability, a proliferation of cn tags might be okay (ish), but a simple comment/note of "these awards really need to be sourced or they'll be removed and notability won't be demonstrated" will work just as well (plus the creator will be notified). Primefac (talk) 15:30, 26 April 2018 (UTC)
I'd say in general that there is too much tagging going on here without going too in-depth to any specific case. I'd echo GMG's comments. Less is often more in these cases. TonyBallioni (talk) 00:48, 27 April 2018 (UTC)
How can you review a page in 15 seconds??? When I review a page I open all the links to make sure they say what they are supposed to be supporting in the article. Even redirects unless it is an obvious possible mispelling you have to open the target page and check the title is in there somewhere. Dom from Paris (talk) 11:58, 27 April 2018 (UTC)
I have also requested this on their talk page, in case they are not receiving pings, but they have not edited since I made said post. Primefac (talk) 13:58, 27 April 2018 (UTC)
@Vermont, GreenMeansGo, Primefac, TonyBallioni, Kudpung, and Domdeparis:. Greetings to you. Before I start my explanation/respond to the above, I would like to seek your patient on reading my message as writing eloquently is not my strong suit, as metaphorically speaking it might put glass eyes to sleep, for this might be the disadvantage of knowing multiple languages and master of none syndrome. Firstly, I didnt know I was supposed to respond to this message as the Vermont requested an audience of sysops and coordinator and I am not one of those and I hope not the case was raised because of my question to Vermont on how a pattroller review a page without a single source provided that prompt such displease since Vermont and Mz7 and I (Nicolas Moussiopoulos) already had a conversation prior. I have said I might have overtagged and thanked Mz7 advice which means I accepted the recommendation/suggestion even though I acted out of good faith (and know now it might not be aligned with Wikipedia guidelines) by not only placed the tag but also (7 Nicolas Moussiopoulos - inline citation needed) explained to creator the reasons, instructions of citing and example how a ref cite look like and by tag citation needed is to let the creator know where to place the possible inline citations. All this is to provide a "specificity" to creator instead for the new editors to read all texts and links in Wikipedia and find how to cite source which would be a little daunting for the amount of information need to thought to know/find to know what to do (we all have been there before to know it was like when we first joined Wikipedia family). Secondly, I did not review the article but Vermont did as there are only 2 refs and one of them need a password to login and the other is just a paragraph of text. For such I place the tag to hopefully request for more ref from creator. I only started tagging with both bioplates and cn for the last (I think - not 100% sure there, so dont bury me on this) 10 days out of my NNP review since mid Nov. last year which I have reviewed total of 848 pages I patrolled accordingly to Xtools findings and if I am familiar/have knowledge with the subjects I would find and cite additional independent sources to strengthen the article's notability in hope to avoid the article might be PROD in the future. I have explained my actions and have accepted the recommendation of refrain from doing such and if you find it is not satisfactory, do what you may for I have nothing to refute as I volunteer my service is to give back what Wikipedia has given me all the info/knowledge for years and since I am now taking a hiatus from work and have the free time to do so while I am moving around/living in different countries. Thank you. CASSIOPEIA(talk)00:17, 28 April 2018 (UTC)
Hey Cass. No one is trying to crucify you. As long as you're open to constructive criticism I think we'll be okay. You just need to be open to it. GMGtalk00:25, 28 April 2018 (UTC)
I'd like to clarify that no, I did not start this discussion because of your question to me a few weeks ago. I started this discussion based off of the 19 pages with excessive tags I linked above. As GMG put it, we're not here to crucify you. We're here to discuss this issue. Vermont (talk) 00:33, 28 April 2018 (UTC)
It could be that the rapid edits are due to one edit triggering several automated edits. Some spot checks of AfC reviews seem to show that they know what they are doing, but needs second opinion. The user's English is unusual - at first I thought machine translation, but there are some unusually apt uses of less common idiom in the text. Kudpung กุดผึ้ง (talk) 01:30, 28 April 2018 (UTC)
Latest comment: 6 years ago4 comments3 people in discussion
I'm interested to know if anyone has any opinion on Global Music Awards. The article and the award do not smell right to me, and I'm wondering if it is promotional. The judging process appears to be dubious and self-perpetuating (the judges are largely past winners [2]), what its entry criteria are (it appears that you can enter it yourself, but may have to pay [3]), and the winners are apparently largely unknown apart from some recognizable names. Seems like a self-promotional award site to me. Hzh (talk) 14:08, 28 April 2018 (UTC)
Definitely a pay to enter competition. As it says on their page "Global Music Awards opens doors. Our goal is to be your partner in your journey to the top. If you win our honors, our credibility may make it easier for you to lock-in bookings, be discovered by A&R executives and be signed by labels." Not enough secondary sources tishow it is notable. Dom from Paris (talk) 14:34, 28 April 2018 (UTC)
It is odd, I shouldn't imagine someone like Esperanza Spalding would pay to enter the award herself, which makes me suspect they sometimes throw in a random artist to make it seems somehow legitimate. A candidate for deletion then? Hzh (talk) 14:41, 28 April 2018 (UTC)
The Page Curation tool reports 'unable to find target location' i.e. Redirects for Discussion, when sending redirects to RfD. Could somebody please stat a bug report at Bugzilla. Maybe Kaldari could address this issue directly. Thanks. Kudpung กุดผึ้ง (talk) 22:48, 22 April 2018 (UTC)
@Kudpung: I think I've fixed this. The fix should go out next week. Unfortunately, the deletion discussion systems are so complicated and quirky that the software that interacts with them is extremely fragile. For example, for Redirects for Discussion, it has to find the current log page (e.g. Wikipedia:Redirects for discussion/Log/2018 May 2), and then instead of just posting to the bottom or creating a new section like you do for any normal wiki page, you have to find the comment that says "<!-- Add new entries directly below this line. -->" and post the nomination directly under that. Unfortunately, someone changed the daily log generation script so that the comment says "<!-- Add new entries directly below this line. -->" instead of "<!-- Add new entries directly below this line -->". The period broke Page Curation! Is there any chance that we could introduce some sanity into this system so that it isn't so difficult for WMF software to interact with? Ryan Kaldari (WMF) (talk) 01:21, 2 May 2018 (UTC)
Latest comment: 6 years ago6 comments4 people in discussion
Because so many of the articles we review are moved to draftspace leaving a redirect automatically tagged for CSD, I propose NPP automatically grant the right "suppressredirect". Thoughts? Jjjjjjdddddd (talk) 15:46, 2 May 2018 (UTC)
Suppress Redirect flag can enable NPPers to swap page histories and do complicated moves, which can lead to gargantuan messes, if they are unaware of our RM guidelines, which ought to be often the case.Our PM grant rules are a lot stricter than that of NPR grant and trends show that PM is disbursed for the aforesaid purposes of easy draftification to only those long-standing editors who don't have a track record of disputed moves and/or can be certainly believed to stay away from RMs etc. (in case they are non-experienced).~ Winged BladesGodric16:59, 2 May 2018 (UTC)
NPP curation toolbar
Latest comment: 6 years ago7 comments3 people in discussion
hello, from 6 hours ago the curatiom toolbar just don't pop up when I am aboutto take reviewing of a page. Tried to check SQL the script is still there. any remedies ?Quek157 (talk) 01:45, 23 May 2018 (UTC)
yes, I changed browsers as well as deleted the SQL and re imputed it, still can't. I can see review button on new page feed but click on it just a line below stating mark this page as patrolled. clear all caches already. Quek157 (talk) 08:19, 23 May 2018 (UTC)
I guess I'll use twinkle in the mid time and do content creation till the things are sorted up, likely script error. Quek157 (talk) 08:43, 23 May 2018 (UTC)
Tried edge browser, chrome still didn't work. Will file a phabricator report soon once I had time. I think I reviewed enough for this month but will still use twinkle to CSD or XfD and manually patrol without the curation toolbar. --Quek157 (talk) 10:59, 23 May 2018 (UTC)
Resolved, apparently the toolbar hide itself and ended up a small link at "tools". --Quek157 (talk) 11:30, 23 May 2018 (UTC) further confirmed by marking a page as reviewed. --Quek157 (talk) 11:37, 23 May 2018 (UTC) thanks for all helps received --Quek157 (talk) 11:37, 23 May 2018 (UTC)
Ling Dong
Latest comment: 6 years ago8 comments4 people in discussion
I've been trying without success to nominate this article for PROD via BLP notice board and on the article's head. It is a BLP with the only source being a link to the baidu search engine - i.e. it has no sources. I'm feeling like I'm getting a bit militant because I have started reverting people who remove the PROD notice without fixing the issue. Please advise. Cheers. Edaham (talk) 04:11, 9 May 2018 (UTC)
It was removed by two different people. The next best thing is for you to start AfD if you believe the subject is not notable, or to find sources yourself if you believe otherwise. But constant reverting is never the good option. –Ammarpad (talk) 04:40, 9 May 2018 (UTC)
I only reverted once so far. It's a BLP sitting there with no sources. it came up on NPP - The subject may or may not be notable. The text is all unverified. I can't make a search to verify any of it. Just trying to follow the book and maintain BLP standards. AFD is an optionEdaham (talk) 07:34, 9 May 2018 (UTC)
Good tag, will watch the page. You're right, a generic image search link is not a "reference", nor is it anything other than a link. Primefac (talk) 12:21, 9 May 2018 (UTC)
Hi I just checked out the page and backed up the message left on Aspects' talk page by Edaham here User_talk:Aspects#Ling_Dong and then just by curiosity I clicked on the link again and this time it came back with a load of images of a casette tape written in chinois, this may be sufficient to support the statement that she released a casette tape if the name on the image of the tape is indeed Ling Dong. The trouble is that the link is not stable so often it defaults to the home search page. Dom from Paris (talk) 15:24, 9 May 2018 (UTC)
thanks for taking the time to check. Aside from verifying notability, sources on BLPs need to verify what’s being said about the subject and withstand a greater degree of scrutiny than articles which aren’t about living persons. Unverified text (all of it in this case) can be challenged and removed. Since removing the unverified text would mean blanking the entire article, a more appropriate action would be prod, especially since the authors of the article have made no effort to provide adequate sourcing despite numerous notifications. Regarding the paltry source given, there’s not even verification that the person being talked about in the article and the assortment of photos from that search are the same. If this were AFC.... etc etc. I don’t need to spell this out. Edaham (talk) 17:38, 9 May 2018 (UTC)
From what I understand if any source of any kind supports any statement in the article then a blp prod is not valid. As I said the statement that she released a cassette would be backed up by a photo of this cassette. I do not read Chinese so I don't know if what I saw said Ling Dong or not. Dom from Paris (talk) 18:25, 9 May 2018 (UTC)
I see, yes that's also the way I understand it. What I couldn't fathom is how an image search constitutes a "source of any kind" :). I won't argue (again) about any decision which is made though. Thanks again for looking into it. Edaham (talk) 00:37, 10 May 2018 (UTC)
Nothing But an Infobox
Latest comment: 6 years ago2 comments2 people in discussion
Chips, cheese and gravy had been a redirect to Poutine. A user has (twice) removed this redirect. At the moment all that exists is an infobox (see this [4]). It seems like this user is still developing the page but for future reference would I review this as a blank page or treat it like it has content? Looking for some guidance about this article but also concept more generally in case I were to see it again. Best, Barkeep49 (talk) 23:12, 8 May 2018 (UTC)
Barkeep49, personally I'd pass it by until it was at least 2 hours old. If it was a few hours old and still like this, I'd probably move it to draftspace and send a message to the creator thanking them for their work, and asking them to move it back to mainspace when the article is finished. Best wishes, Boleyn (talk) 19:20, 10 May 2018 (UTC)
RfC: Deletion of drafts
Latest comment: 6 years ago1 comment1 person in discussion
I would mark as unref and send the creator a message saying 'Please add your references. External links just means suggestions for further reading. We really need to know what your sources were so we know the information is accurate. Could you please add them clearly, preferably WP:INLINECITED?' On the whole, people do respond by adding them. Boleyn (talk) 08:56, 17 May 2018 (UTC)
CoI editor
Latest comment: 6 years ago5 comments4 people in discussion
How does it come about that editor Azerbanjo, who on May 9th created the new article Bionix Radiation Therapy, has only two actions in their contribution history whereas looking at the editor's talk page it is apparent they have made other contributions? I see I asked them to disclose whether they had a conflict of interest a month ago and they have not done so. Cwmhiraeth (talk) 06:43, 18 May 2018 (UTC)
It has implications for new page reviewers. It would be much more helpful if non-admins were able to see that contributions had been made even if they could not view the details. Cwmhiraeth (talk) 09:27, 18 May 2018 (UTC)
Thank you! I've wished we were able to see deleted contribs. Now if only there were some way to tell if a deleted article recreation is the same as it was when deleted... Natureium (talk) 16:44, 18 May 2018 (UTC)
Backlog still growing
Latest comment: 6 years ago4 comments3 people in discussion
Note that despite the fact that ACTRIAL has now become permanent the backlog is still growing and currently approaching 5K articles.--Ymblanter (talk) 07:18, 16 May 2018 (UTC)
There are a ton of almost identical pages about Indian politicians that all meet notabilty that have just been created and also loads of music redirects that can be dealt with really quickly. I managed to about 100 today just by selecting the editors that created these pages. Of cource this doesn't add to my edit count by I'm not in it for the glory...well maybe just a little bit. Dom from Paris (talk) 14:18, 16 May 2018 (UTC)
General sanctions in effect for blockchain and cryptocurrencies
Latest comment: 6 years ago3 comments3 people in discussion
A proposal for the community to impose General Sanctions on cyrptocurrencies and blockchain has been posted and might be of interest to New Page Reviewers. Best, Barkeep49 (talk) 07:34, 18 May 2018 (UTC)
Women and GLAM...those are going to be hard to deal with. Every time I tag PROD or Afd a GLAM subject there seems to be a little army of members of the GLAM community that jump out to defend the article tooth and nail... exhausting. Dom from Paris (talk) 13:28, 24 May 2018 (UTC)
Latest comment: 6 years ago30 comments7 people in discussion
NPP backlog, number of unreviewed articles by creation date. Green is less than 30 days old, orange is between 30 and 90 days old. Red arrow indicates the 90 day index point.
The backlog is currently at 4800 articles and dates back to very early March. We have been generally keeping the numbers steady, but the start of ACREQ has not resulted in an appreciable reduction of the backlog as hoped (due to a drop in reviewing activity following ACREQ). I can't fault anyone for taking a well deserved break, as I myself have also been on an extended wikibreak, but special thanks go to those that have continued chipping away at the coal face. Thank you!
We have only about two weeks buffer from the Index Point at present and the shape of the backlog is also a little less healthy than it was in my last update: there is a pretty big bulge around the middle of March that is headed our way, so we need to be prepared to make sure that we can deal with it and ensure that it doesn't push past the index point. Review from the back if you can, our long term target is still to keep the backlog below 30 days, so we still have a long way to go. Cheers all and keep up the good work. — Insertcleverphrasehere(or here)23:09, 21 May 2018 (UTC)
We could ask, but likely wouldn't happen, and not really needed right now since we are under control. I don't really consider us backlogged as we're before the 90 day point. Yes, there are pages to review, but we are able to have quality control on every page before it hits Google. That's the most important thing. It'd be nice to get it into Google before 90 days, but that's less important than making sure it's of sufficient quality for Google TonyBallioni (talk) 23:17, 21 May 2018 (UTC)
@TonyBallioni: There is no reason why the backlog needs to be more than 30 days. 5,000 articles in the backlog is also still unacceptably high IMO, especially when we can get bulges in the graph like the one above that can make it difficult for us to deal with when they approach the Index Point. When we had a month buffer it was a lot better, but we have slowly been losing ground since then as the backlog has spread out. It isn't as high at the front any more, due to less articles coming in, but it has spread out toward the tail a lot more, which is the opposite of what we want. We should be targeting a 30 day max backlog, and <2000 I think. We are still in a situation where half of all reviews are done by the top few 10% of reviewers, and if any of them decides to quit, we could be in real trouble again. That's how the backlog spiraled out of control last time, and I'd like so have a decent buffer so that we can ensure that it doesn't happen again. — Insertcleverphrasehere(or here)23:30, 21 May 2018 (UTC)
I was responsible for setting the 90 days period on a discussion with Kaldari because I thought we would be able to cope with it. If there is going to be a crisis, I can probably get it extended without a fuss, but we wouldn't want it to become an excuse for reviewers to do even less patrolling. That said, with many UPE coming in now since ACPERM, it's probably what we'll have to do. And sorry to keep harping on it, but ~630 reviewers should be more than enough, but y'all know my views on that. Kudpung กุดผึ้ง (talk) 02:31, 23 May 2018 (UTC)
I'd be more keen on reducing the index point rather than extending it. Extending it just puts us in a position to let it get large enough to get out of control. If the Index point was at one month we would have an incentive to keep the backlog small enough that it would not be daunting for even one or two people to crush back down if it approaches the index point. I realise that this view isn't particularly popular, but it is basic human psychology; if it was hitting the index point and the numbers are low enough that you think you personally can make a difference, you are more likely to do a bunch of reviewing compared to if the backlog is at many thousands of articles. Honestly, not blaming anyone, but extending the backlog to 90 days didn't help last time and probably did more harm than good. Extending the backlog further is a recipe for disaster. — Insertcleverphrasehere(or here)03:12, 23 May 2018 (UTC)
I'm afraid I don't share your opinion, and using words like 'disaster' is over dramatising the situation. The real issue which I'm now forced to bring up, is that 630 reviewers is a ridiculous number to not be capable of doing 630 patrols a day between them. I don't even consider my own patrolling to be a participation but time and time again I keep cropping up in the top 40 who are doing 95% of the work, and I think I've done my bit for NPP after all these years. I think the other 590 are an insult to what we are doing and it's time to introduce a feature I forgot about when I created the user right. If the AfD people can do it, then I very sure we can. Unfortunately fur us volunteers there are no carrots, only sticks. That's the actual psychology, while the WMF are sitting on their laurels and earning their salaries thinking they've appeased the hungry masses with an oblation for AfC. To be fair to them though, what they are doing will raise the quality of reviewing, but it won't increase the speed or make it a more attractive task. Your and Legacypac's initiative to increase the numbers of reviewers is certainly not without merit but now is perhaps the time to audit the result.Kudpung กุดผึ้ง (talk) 08:45, 23 May 2018 (UTC)
@Kudpung: The fact is: people go in and out of reviewing, some out for good. If user rights should be about trust and not a status symbol, then why stigmatise and take rights away from trusted editors that don't use them? You keep harping on about '630', and saying "if everyone XXX" like it matters. Only a small fraction of our reviewers will ever be active at any given point, and taking rights away from inactive users won't change anything. By all means go through and audit the numbers and remove the flag from everyone who hasn't reviewed anything in the last 6 months or whatever if it makes you feel better, but how will this improve anything at NPP? If even one of the reviewers who recently joins becomes an active member of NPP, our 'initiative' will have been worth it. I DID audit the result of my first round of invitations, you know the results. You seem to want to have your cake and eat it too, you want the Cwmhiraeth and Babymissfortune and Elmidae (thank you guys very much) without getting any of the users that didn't end up doing any reviewing (I won't name them but you have seen my list). I'll say it again but I sound like a broken record; it just isn't possible. Statistics doesn't work that way, and the only way to find those editors that might devote real time and energy to the project is to cast a wide net even if you get a lot of bycatch.
Extending the index point is basically saying "let someone else do it, later". How is that not inviting a future disaster? I vehemently oppose extending it, even if we fail to meet a deadline and articles spill over into the index zone (Kaldari, please take note of that). Basic human psychology: if you extend a deadline beyond what is actually needed, most people will put off the work until later. It solves exactly zero problems to extend the index point or to have an index point that is far longer than it needs to be. It should really be reduced back to 30 days as soon as we can get the backlog back down to that level. — Insertcleverphrasehere(or here)09:34, 23 May 2018 (UTC)
I have always tried to emphasise on increasing the number of reviewers. Thats why i had created the invitation message template. But unfortunately, lately my overall activity has decreased a lot. If in the next newsletter we can say something like "reviewing 5 pages a day" would cripple the backlog, that'd be great. I mean, if we consider that only 100 reviewers would review 7 pages a day (a combined average), that would be 700 pages per day. Then within 12-13 days, we will have the backlog on the same levels of pending changes, that is, "as they come-in". That would still be like 400 per day though. @Kudpung: I am not sure what you are referring to with "If the AfD people can do it, then I very sure we can." Would you elaborate please? —usernamekiran(talk)09:14, 23 May 2018 (UTC)
I felt sure it would have been obvious that I was referring to 'If you don't use it, you lose it.' We can play with the math as much as we like but it takes 20 minutes to patrol 7 articles. That's 1 a day, or 7 on a Sunday afternoon. You're good at providing stats, could you come up with an audit of the performance of the rights holders that were created as a result of your campaign? Just to see if it did any good. Kudpung กุดผึ้ง (talk) 09:28, 23 May 2018 (UTC)
See my reply above. I have no interest in doing another audit. My previous efforts proved that casting a wide net worked well, and irregardless, these users that were granted the NPP useright have been deemed competent and trusted, there is no reason to take it away unless that is no longer the case, regardless of activity level. — Insertcleverphrasehere(or here)09:37, 23 May 2018 (UTC)
(edit conflict) @Kudpung: about the stats comment, I am guessing it was meant for ICPH, based on your edit summary. I am good with stats too, but all i can provide is numbers. No graphs/charts. Also, ICPH was keeping track of newly joined reviewers in his userspace. I will search for that with prefix tool. —usernamekiran(talk)09:38, 23 May 2018 (UTC)
@Usernamekiran: Yeah that is the list. I stopped updating it around the end of February as the point had been made and it had become a time sink. It is rather obvious that the best way to get reviews done in the long run is to invite new editors to the project. Despite the (by now) tens of thousands of reviews that have occurred as a result of that invitation drive, I seem to get nothing but flak from Kudpung for the users who were granted the right but haven't done a lot of reviewing. — Insertcleverphrasehere(or here)09:48, 23 May 2018 (UTC)
Oi, I didn't say that. I was commenting in general on the 630 of whom only about 40 are doing the bulk of the work. I'm just curious to know not only about the competency of the users you recruited, but more importantly to know what impact their actual patrolling has had on the backlog. That would be the true metric of the efficacy of your initiative. If it worked, well, fine - do it again. I'd rather see that than a backlog drive. OTOH, psychologically it can work counter-productivitly: 'Aw, we've got over 600 reviewers, I don't need to do anything...'Kudpung กุดผึ้ง (talk) 09:57, 23 May 2018 (UTC)
This tells me most of what I need to know. And it's very telling. The 40 or so regulars are not going to be put under pressure to do more - nor will I. The only way to get more done is to encourage the inactive users to do something. If that makes sense... Kudpung กุดผึ้ง (talk) 10:13, 23 May 2018 (UTC)
Just a take that AFC as what Kudpung always said is less complex than NPP, and we have so much trash there anyway, so I need sometime 2 minutes to review per page there than the 15 I need at NPP. Hence, the backlog can be reduced faster there than here. Just pray hard that less nonsense submission there is good enough already. So I think any comparison between the 2 backlogs are insignificant. Anyway the worst of the backlog is never these 2, is the Good Article Review or the reassessment. We still have sometime IMO. ---Quek157 (talk) 11:03, 23 May 2018 (UTC)
@Kudpung: This power law curve of New Page Patrollers that I prepared at the end of 2017 will always be a thing. If you graph the editors that I invited that joined, the same shape curve is seen. i.e. if you invite 100 people, a half dozen of them will do half of the work, and the other half will be done mostly by about 30 people. This isn't going to change, but if we want to get more of those half dozens, we need to get more groups of 100 to join. The return rate of invites is somewhere between 10-15%, so that means inviting another 1000. The place to continue is where I found the last cohort: Wikipedia:List_of_Wikipedians_by_number_of_edits#1001–2000. I finished the top 1000 last time. Template:NPR invite is also still there waiting to be used. I'll go back and invite more users at some point, but it is rather frustrating to get a huge backlash onto me every time one of my invitees fail. I got a lot of flak from you for invitees that failed due to stuff that I couldn't have known about (the one you brought up specifically who I won't name here), or for user activity level, which wasn't actually a requirement of the user right at the time (this has been subsequently clarified). Carefully vetting every user is something that needs to be done before they are granted the user right, but doing so before inviting quickly becomes prohibitively time consuming, which is why I stopped inviting people in the first place. — Insertcleverphrasehere(or here)11:33, 23 May 2018 (UTC)
I understand you power law argument only too well and I can't dispute it, it's exactly the problem we're faced with. We also have over 1,000 admins but only about 30 of us are doing the bulk of the work but one thing is for sure, if they don't use their tools, here comes a time when they wake up without them - and that's what happens at AfC. but if you want 5,000 patrollers to get enough to do the work of 500, bear in mind quis custodiet ipsos custodes? at the moment it's only me and DGG and we've both got other things we would prefer to be doing. There's a scary undertermined number of reviewers out there who are offering to write articles for money, do you want to put the foxes in charge of the henhouse? It's all happened before.
It's not time consuming if you organise your regex or quarry or whatever you folks do for datamining, and enter enough criteria like we did when we grandfathered the first batch of reviewers. You'll certainly not get a lot of thanks from admins if you swamp PERM with a lot of extra work - particularly when you leave them with the unsavoury job of researching and declining 15-25% of them, we don't just read the user names and click a button. I don't even work there as much as I used to but when I take a look I see applicants yelling because their requests haven't been processed within 24 hours or getting really nasty when we decline. Anyone in that much of a hurry is only wanting to fill the top shelf of their wardrobe. If you want me to give you a set of criteria for you to compile a search string, let me know. It might keep your computer running all night, but those are things Scott and I had to do in the past for NPP for less urgent matters than we are discussing today. Kudpung กุดผึ้ง (talk) 14:10, 23 May 2018 (UTC)
I think Kudpung is making two separate arguments here and the interplay between them is worth considering. There is the "we need to be careful who we give this to" argument. I hope we can all get behind that for all the reasons mentioned (fox in the henhouse, overburdened admin, etc). The second argument is that we should take away the permission from people (who presumably are trusted) for inactivity. There is some logic behind this - the threat of removal will likely cause some number of reviewers to use the tool rather than lose it (people like their hats). However, is more gained from this coercion of a few editors into activity than loss caused by the ill will of removing a permission and whatever likelihood that someone returns to active NPP activity after some period of dormancy? There is also the consideration that at what point does someone's inactivity cause them to no longer be trusted with it? To me this would be an inactivity with Wikipedia rather than the tool itself - if someone is active around here they'll know policies and can still accurately CSD (or not, but being active with the tool doesn't mean they'll get better with this). As someone who wasn't here for the "bad old days" I do think that we're in a position now where NPP needs to focus on supply rather than demand and so a sustained drive to eliminate the backlog (as below) combined with a change of noindex to 30 days for ongoing review pressure would benefit the encyclopedia. Best, Barkeep49 (talk) 15:12, 23 May 2018 (UTC)
@Kudpung: I'd be very interested in a set of Quarry-able searchable criteria that might enable us to narrow down the editor pool. This is an idea with a lot of merit. I previously did it manually, going from the list of top wikipedians and checking each user in Xtools for number of edits (N/A when starting from the top wikipedians page) >10000 ideally, >5000 min for my personal criteria, activity level (must meet NPR requirements), block log (clean 6 mo+), deleted creations in the xtools page creation log (no deleted submissions for non-notable any time recently, and very low % deleted main space articles), and I also checked their talk page for other worrying signs. While some of these can be automated (the first three), I'm not sure about the others. What other searchable criteria would you think are useful? — Insertcleverphrasehere(or here)20:28, 23 May 2018 (UTC)
edit count >5000??? or 10000?? minor, major, automated?? Anyway, for this I can't be a NPP anymore, just saying from a 2900+ edit account, others seems fine --Quek157 (talk) 21:44, 23 May 2018 (UTC)
@Quek157: Don't worry, these were just criteria that I was using to easily identify users that were highly and obviously qualified for NPR. Many users with much lower edit counts may be great New Page Patrollers, but it becomes a bit hit-and-miss the lower the edit count goes. These numbers were chosen from a practicality standpoint, as there are a ton of editors with >5000 edits that are active and ready to invite and at less than this I would run into a lot of editors that I would review and then decline to invite for various reasons. Don't take these numbers as a minimum required, it was a rule of thumb I developed to save me time; e.g. The vast majority of still-active editors with counts >5000 will meet the other criteria, so it makes sense to target the higher count editors first. I should have explained this better, sorry. — Insertcleverphrasehere(or here)21:53, 23 May 2018 (UTC)
not to worry, just intented to be humour but the edit part is true, minor, major is important, I can make 10000 small minor addition (like correction of typo) to an article and then I can be in the list, so be careful. Also mainspace is important, I won't trust an editor who have 5000 edits but only 10% to mainspace / wikipedia spaces(but then some editors can do 1 GA with 1 edit) either. --Quek157 (talk) 21:58, 23 May 2018 (UTC)
If you are determined to organise a backlog drive, let me know the dates in the next 24 hours before I publish the The Signpost - I have an article about NPP in it. Kudpung กุดผึ้ง (talk) 14:10, 23 May 2018 (UTC)
Done. publication probably withn the next 48 hours. I'll also take a look at the NPR newsletter draft and discuss it with Tony, maybe one should be sent out soon. Kudpung กุดผึ้ง (talk) 01:32, 24 May 2018 (UTC)
Insertcleverphrasehere, I have a busy day in the office today. I'll do the search criteria list this evening. Signpost notice for the backlog drive is done including an appeal for more reviewers. Publishing in about 10 hours. Newsletter is ready but I'll send it nearer the end of the month in case there are new developments - especially to the WMF work on the feed, by which time I might also have news about the upcoming coord election. Kudpung กุดผึ้ง (talk) 04:13, 24 May 2018 (UTC)
InsertcleverphrasehereThe news letter has been sent already in order to coincide with the publication of The Signpost. It met withTonyBallioni's approval last night[5]. Whatever direction the discussions on this page take, we need to instil in our reviewers the need for their active participation, and to recruit for more reviewers. I'll get that criteria list done for you asap. I'll be issuing another, short newsletter on the eve of your backlog drive. By then we may have more updates from Marshall. Kudpung กุดผึ้ง (talk) 21:02, 24 May 2018 (UTC)
Speedy deletion problem
Latest comment: 6 years ago6 comments3 people in discussion
I have just nominated Barima Sidney for speedy deletion for G11 and G12 using page curation. A notice has appeared on the creator's page and another on the article's talk page, but no speedy deletion template has appeared on the article page. What should I do next? Cwmhiraeth (talk) 12:30, 25 May 2018 (UTC)
@Cwmhiraeth: requested on your behalf using TW (normally I used TW for CSD, page curation have a certain 'bot' for not processing this well, @Primefac: do action, thanks a lot --Quek157 (talk) 12:56, 25 May 2018 (UTC)
To help equip AfC reviewers to handle that traffic, the Community Tech team are going to be dedicating some bandwidth during May and June to improving that process. We began a good discussion about which improvements would be most impactful toward helping AfC get high quality drafts into the main namespace quickly, and we've settled on a plan. I'm posting here because we think that the planned work will also benefit the NPR process, and I want to make sure anyone interested will be able to weigh in.
Please check out the project page for the details. In brief, the idea is that AfC does not currently have an efficient way to prioritize drafts for review, and the New Pages Feed is an existing interface that could help. The plan is:
New Pages Feed will be extended to include AfC drafts as a new list of pages for review, in addition to the existing "Article" and "User" pages that are currently available.
The feed will be enhanced to allow prioritization by quality and copyvio scores for all namespaces in the interface.
These new capabilities will be available to both the AfC and NPP reviewers.
Though we have limited bandwidth over the next several weeks to work on this project, we definitely want to get this as right as possible in concert with the AfC and NPR communities. I'm going to be posting regular updates on that project page, including mockups of different ways that the changes could be implemented so that we can get thoughts and reactions from reviewers. It's definitely going to be critical to get feedback from this group on how much impact (hopefully positive) the changes will have on the usage of the New Pages Feed.
So the feed will include things in article space and draft space? Will the drafts be submitted, or are we supposed to figure out if the user is done with them? And AfC requires people to explicitly join and be an active reviewer. What does this mean for new page reviewers going through the feed? It sounds like you're trying to combine the two into one process. Natureium (talk) 18:29, 10 May 2018 (UTC)
@Natureium: thanks for the questions. This effort is definitely not about combining them into one process, though I know there has been conversation in the two communities about whether or not that would be a good idea. This is about leveraging the existing New Pages Feed interface to also be helpful for the separate AfC process. To that end, the plan is to allow the New Pages Feed to distinguish between the Article, User, and Draft namespaces. And then within the Draft namespace, the reviewer will be able to choose the subset of drafts that are submitted and awaiting review (because drafts can also be submitted and declined, waiting for the author to improve them before resubmitting). So to be clear, we don't plan to make it possible to list both Articles and Drafts together in the feed at the same time.
AfC does require people to explicitly join and be approved to be able to use the AFCH script. We don't plan to change anything about that workflow. Once selecting a draft for review from the New Pages Feed, an AfC reviewer would be on that draft's page and use the AFCH script as usual.
NPR reviewers will be able to do their work the same way they always have within the Article namespace, but will also be able to sort and filter by quality and copyvio scores in the New Pages Feed if they wish. We think this will help reviewers either find the most problematic articles first for deletion, or prioritize the best articles first for approval. Does that sound like it will be useful?
Does this answer your questions? What are your thoughts on this plan?
Yes very helpful. There are actually three kinds of Drafts that should be sortable 1. Submitted to AfC awaiting review. 2. Declined by AfC (might be submitted again) 3. Not submitted to AfC (may not even be tagged). Group 3 is worth looking at for attack, Spam and other problematic pages as well as useful pages that could be promoted since some people seem to think Drafts magically move to mainspace on their own. All types of Drafts could benefit from patrolling. Legacypac (talk) 16:09, 11 May 2018 (UTC)
@Legacypac: thanks for the clarity. I've passed this along to the software engineers who are currently planning out the work, and we'll be discussing whether it will be feasible to enable filtering to all three of those kinds of drafts. -- MMiller (WMF) (talk) 16:22, 11 May 2018 (UTC)
I'm not a software engineer but it seems easier to include all of Draftspace with a filter for AfC Submitted tags than to exclude a block of untagged pages. All Drafts are temporary.
1. Draft created
2. Draft MAY be submitted AfC
2.a Draft may be accepted and becomes a redirect to mainspace
2 b Draft is declined and may be resubmitted or not
3. A patroller notices the Draft and seeks deletion (afc or not) OR
4. The creator or someone else moves it to mainspace and title becomes a redirect or it is simply redirected without a move (topic exists)
5. Draft reaches 6 months unedited and gets considered for WP:G13
(Sorry hard to draw a flowchart with words)
The entire universe of Drafts (other than redirects) is therefore finite and somewhat stable in total numbers. It is (currently poorly) managed as a backlog of unreviewed pages. Adding Drafts to Page Curation Tool will greatly assist in helping to manage all unreviewed pages. Patrolers may choose to improve pages they encounter even if they are not ready to move them to mainspace yet. Arguably marking "Reviewed" would usually main moving it to mainspace. Otherwise sooner or later the page will need to be tagged for deletion or redirected and does not need to be "reviewed". Legacypac (talk) 16:54, 11 May 2018 (UTC)
MMiller (WMF), as a NPP, the biggest issue, in my opinin, were always the way the process rolls. Rentier made a great job creating the NPP browse, I often use it to select topics I am familiar with. I've also seen a tendency of tagging without any attempt to improve the article although many of them, doing a quick Google search, have reliable references. Reviewer should be encouraged to try to improve an article prior to filling them with tags and moving to the next one. I feel this would be a constructive behaviour and beneficial for the entire Wikipedia community. Looking forward for the new features. Robertgombos (talk) 22:48, 24 May 2018 (UTC)
@Robertgombos: thanks for the comments. The thoughts you're bringing up about reviewers being encouraged to improve articles were definitely important themes in the conversation leading up to the new features. It sounded like efforts in that area could be most effective coming from the reviewing community, which is why the WMF project as planned focuses more on technical changes to the New Pages Feed. Please do check out the latest wireframes for the project and leave any comments! -- MMiller (WMF) (talk) 22:38, 25 May 2018 (UTC)
@Robertgombos:, MMiller (WMF) the focus at NPP and AfC has never been on the improvement of articles by reviewers. If they do that the backlogs will never decrease. NPP is a triage (look that word up). AfC provides more basic help but namely telling creators in greater detail where to get more help and advice and directing them to the various (and confusingly mukltiple) help pages.
The browser by Rentier is excellent and is of course the kind of thing the WMF should be doing for Wikipedia rather than us constantly relying on patches by volunteers. It should be the 'Third' option in the filter selector at the New Pages Feed. Kudpung กุดผึ้ง (talk) 01:19, 26 May 2018 (UTC)
@Kudpung:, yes, I know. Looking forward for the NPP improvements to be implemented. Since ACTRIAL, I am absolutely sure the SD rate decreased which is just great. So, it seems like everything is heading to a good direction. Robertgombos (talk) 08:25, 26 May 2018 (UTC)
the Russian articles
Latest comment: 6 years ago8 comments6 people in discussion
I’ve actually tagged one of them. Since there’s so many of them coming in, my feeling is that the editors should be encouraged to find (if possible) at least one English language source per article, for the sake of good practice. This is based on a loose interpretation of wp:NONENG. It’s not mandatory that sources be in English, but since there’s so many of them coming in I think it would add value to the collection of articles were we to suggest looking for English sources, while encouraging the great efforts being made at the same time. I don't suggest tagging articles which can be verified through the sources, I probably went overboard with that more citations tag, but leaving suggestions when completing the review might be a good idea.
Just a note. only content non English this applies should be A2. sources can be in non English. if we can't read, use Google translate, still can't pass to another who can and I'm sure someone can (above there is someone who can! =) ) .talkpage is perfectly fine bit the tag is unnecessary in my opinion sources are enough for such a stub (clarified --Quek157 (talk) 11:36, 25 May 2018 (UTC))
I can read Russian sources and I am strongly opposed to the notion that every article must have at least one English-language source. This contradicts to both letter and spirit of our policies.--Ymblanter (talk) 13:03, 25 May 2018 (UTC)
Ping me for Chinese if needed. Just to add, in order not to WP:BITE (very marginally) I replaced the refimprove to needing inline citations, that will be gentler. Hope you don't mind as I don't wish any to get discouraged. --Quek157 (talk) 13:23, 25 May 2018 (UTC)
I can also help with Chinese. I stand corrected. Hope I didn’t discourage anyone, I’m just looking at a very large amount of material coming in and trying to think of a way to increase the value of it. Cheers for the guidance and I’ll take note of the objection for future reference. Edaham (talk) 14:30, 25 May 2018 (UTC)
TonyBallioni (talk) 23:17, 21 May 2018 (UTC)
Done. publication probably withn the next 48 hours. I'll also take a look at the NPR newsletter draft and discuss it with Tony, maybe one should be sent out soon. Kudpung กุดผึ้ง (talk) 01:32, 24 May 2018 (UTC)