Wikipedia

Wikipedia talk:New pages patrol/Reviewers/Archive 19

< Wikipedia talk:New pages patrol | Reviewers
Latest comment: 6 years ago by Insertcleverphrasehere in topic 286 15 0 articles
Archive 15←Archive 17Archive 18Archive 19Archive 20Archive 21→Archive 25

Backlog Drive Update: week three results!

Latest comment: 6 years ago14 comments6 people in discussion

Week three of the Backlog Drive is complete (see above section for more details). The backlog of unreviewed articles is now at 4150.

Results:

  • 169 reviewers participated in week 2, compared to 190 in week one and 176 in week 2.
  • 7248 reviews were completed (including redirects and articles), more than either of the other weeks. About 375 non-autopatrolled articles were created each day during this period.
  • The backlog of articles has reduced by another 1700 articles. (about 1900 in previous weeks)
  • The total number of unreviewed redirects has reduced by about 1200, from ~7k to just over 5.8k.
  • Great progress has been made at the back of the backlog, where the oldest articles are now from the 10th July 2017, at the beginning of the week the oldest articles were from the 29th May.
  • The most reviews this week were completed by Dschslava at 1338 (mainly redirects). Boleyn deserves an honourable mention with 939 reviews. Babymissfortune continues his streak of over 500 articles per week (tho only one who remains eligible for the top streak prize). (the totals for all participants this week can be viewed HERE)

— Insertcleverphrasehere (or here) 00:32, 22 January 2018 (UTC)

I think he is referring to the reviewing of user-pages of blocked users to make sure they do not violate user-page guidelines, this isn't part of the NPP backlog per se but it is something that is useful and tangential to the NPP overall project. Articles created by blocked users is a different area, but probably something we should hit as soon as possible. L3X1 reviews articles in some odd maintenance areas that the rest of us don't generally patrol, and his/her efforts are very much appreciated. While my statistics measurements above don't really take these efforts into account, and my data queries on Quarry don't count his/her efforts, I am perfectly happy to give awards for those counts anyway if L3X1 can provide me with some rough numbers as to # of pages reviewed per week and over the whole backlog time period (just send me a message at the end). — Insertcleverphrasehere (or here) 01:40, 22 January 2018 (UTC)
The number I see on the upper right of Special:NewPagesFeed is 19,856. The filters I use are Unreviewed, Userspace, and Blocked, and the advance-the-queue button is showing the faded grey of no more pages to review. L3X1 Become a New Page Patroller! (distænt write) 02:30, 22 January 2018 (UTC)
L3X1, I can confirm. What sort of numbers were we talking about before you started reviewing them? — Insertcleverphrasehere (or here) 02:41, 22 January 2018 (UTC)
Insertcleverphrasehere At least +746, which is the number I've reviewed in the past month. L3X1 Become a New Page Patroller! (distænt write) 02:48, 22 January 2018 (UTC)
L3X1 I don't know why I didn't think of this before, but I can run the data queries for userspace by changing two digits of code... So I'll just do that and add those totals to your review count. Anyone else that has been doing useful work in the userspace just raise your hand and I'll transmute those numbers as well. — Insertcleverphrasehere (or here) 13:04, 22 January 2018 (UTC)
Insertcleverphasehere, I do review userpages, mainly not blocked ones though. Boleyn (talk) 18:02, 22 January 2018 (UTC)
  • Yeah. L3X1 is talking about the userpages of blocked users. They have quite a handful content there. I got confused as I thought L3X1 was talking about articles created by now blocked users. That's a great thing you did there L3X1! At this pace, we should concentrate only on working from the backside of whatever filter we work on. In no time, the articles will get to 500 pages, and then we will have a lot of time on our hands. —usernamekiran(talk) 18:19, 22 January 2018 (UTC)

Do we have one of those graphs like we did for the other two weeks?Dschslava Δx parlez moi 23:16, 22 January 2018 (UTC)

​Not this week sorry, I've been incredibly busy with a GA review this weekend. I am planning something special with regards to this for the end of the backlog drive though (a .gif that scrolls through daily progress). — Insertcleverphrasehere (or here) 23:36, 22 January 2018 (UTC)

Random overnight thought

Latest comment: 6 years ago32 comments22 people in discussion

Should we add NOINDEX to articles tagged with {{undisclosed paid}}? I don't know whether it will accomplish anything, but hey. MER-C 20:25, 22 January 2018 (UTC)

Probably. — Insertcleverphrasehere (or here) 21:06, 22 January 2018 (UTC)
Such a proposal would probably encounter significant opposition. I do think it would remove much of the incentive for creating promotional content. The whole purpose of creating such articles is to increase the visibility of the subject and benefit from the cachet that a Wikipedia article provides. If we take that away, the number of promotional articles submitted is likely to drop significantly. The professional spammers would notice this immediately, and even the editors who were not aware of our COI policies and guidelines will find out soon enough. It's something that comes up frequently; I've encountered editors who think their article isn't published until it shows up on in Google search results. We would need to develop processes around substantiating the claim of undeclared paid editing. While the signs of UPE should be obvious to anyone who has been working in NPP for a while, hard evidence that a new editor was paid is very difficult to obtain. We would need to make it mandatory for any user who is suspected of being a UPE on the basis of behavioural evidence to respond to a request for disclosure of conflict of interest. Mduvekot (talk) 21:15, 22 January 2018 (UTC)
It will only work if the UPE tag can be put on articles created by obvious UPE SPAs. It's not even possible to get a CU on those unless they do something amazingly stupid like using similar edit descriptions on multiple accounts. Rentier (talk) 21:22, 22 January 2018 (UTC)
If you mean NOINDEX until reviewed, I do not see any problems with the suggestion.--Ymblanter (talk) 21:37, 22 January 2018 (UTC)
There's a hearty tiff going on at the moment at Template_talk:COI#Related tags, about this tag among others. A complaint appears to be that editors tend not to follow up with the required talk page actions (here, placing {{Connected contributor (paid)}} and providing some elucidating link). That may need some sorting out before this idea can acquire traction. --Elmidae (talk · contribs) 07:31, 23 January 2018 (UTC)
It think it's important to distinguish between "I think this may have been written in exchange for money" and "a major contributor who has been asked to clarify a apparent COI has not yet provided a response". There is process; If you tag an article, you also seek clarification from the contributor with the apparent COI. Indexing will have to wait until the issue concerning the COI has been resolved. It should be noted that failure to respond to undisclosed paid editing concerns typically results in an indefinite block. Mduvekot (talk) 14:17, 3 February 2018 (UTC)
And yet I see a discussion on the COI template talk page about how the clarification request often doesn't happen. So I am wary. Jo-Jo Eumerus (talk, contributions) 09:00, 5 February 2018 (UTC)

visible index

Latest comment: 6 years ago2 comments2 people in discussion

I recently removed this special:diff/824500331 after reviewing the page. A page gets automatically gets indexed after the 90 periods. But does it include that "visible" stuff? It can been seen in this revision at the bottom. We are supposed to remove it, right? (just a heads-up: she looks attractive in the thumbnail, but sort of scary in full sized image; approach with caution.) —usernamekiran(talk) 18:58, 7 February 2018 (UTC)

The INDEX magic word and template should really be removed from all articles (and really it shouldn't be used at all, in my opinion), since they're ignored in article space. Primefac (talk) 19:16, 7 February 2018 (UTC)

Week 4 and End of Backlog Drive results!

Latest comment: 6 years ago10 comments6 people in discussion

The fourth and final week of the Backlog Drive is complete (see above section for more details). The backlog of unreviewed articles is now at 3625.

Results of week 4:

  • 155 reviewers participated in week 4, continuing a general trend of reducing participation over the drive; dropping from 190 in week one, to 176 in week 2, and 169 in week 3.
  • 5462 reviews were completed (including redirects and articles).
  • The backlog of articles has reduced by another 525 articles. (much less than previous weeks)
  • The total number of unreviewed redirects has remained steady at around 5.8k.
  • More progress has been made at the back of the backlog, where the oldest articles are now from the 26th July 2017, at the beginning of the week the oldest articles were from the 10th July.
  • The most reviews this week were completed by Dschslava at 805 (mainly redirects).(the totals for all participants this week can be viewed HERE)

In short, a bit of tiredness perhaps in the final week of the drive, but still a successful week in reducing the backlog in both duration as well as in total number. — Insertcleverphrasehere (or here) 00:54, 29 January 2018 (UTC)

 
NPP backlog, showing change in the backlog from 10th-28th Jan. Green areas are where the backlog reduced, red areas are where it grew. Because I didn't start collecting data until the 10th Jan, this chart only represents the change in the latter 2/3rds of the backlog drive.

Total Backlog Drive results:

  • 304 reviewers participated in the drive, with 97 users reviewing 25 pages or more (minimum award criteria). Awards will be given out in the coming days.
  • An astounding 24277 reviews were completed (including redirects and articles)
  • The backlog of unreviewed articles reduced by 5950, with a further reduction in the number of unreviewed redirects by ~1700.
  • We have shaved almost 4 months off of the back of the backlog!
  • 6 Users completed over 1000 reviews in the month, with the two highest review counts being from Dschslava at 2579 (top reviewer), and Babymissfortune at 2329 (runner up).(the totals for all participants over all 4 weeks of the drive can be viewed HERE)
  • It is somewhat beyond my abilities to separate redirect and article reviews in my statistic gathering queries, so I apologist that I cannot single out the user who reviewed the top number of articles, though I suspect it is one of the top 6: Dschslava, Babymissfortune, JTtheOG, Boleyn, Seacactus 13, or Cwmhiraeth. If you could each self-report the rough percentages/ratios of articles/redirects, I can crown a 'Top Article Reviewer'. Thanks! (update: it is JTtheOG)

Well done! We have made fantastic progress during the backlog drive, reducing the backlog to less than half of what it was at the end of December, and we are now well on track to review the remaining articles before the end of ACTRIAL (14th March). We are nearly there! Big thanks to everyone who participated, and awards will be sent out in the next week. Cheers and keep up the good work! — Insertcleverphrasehere (or here) 00:54, 29 January 2018 (UTC)

Well done, everyone, especially Insertcleverphrasehere. The momentum's slowed, as expected, but if we can keep it up we'll still be hacking away at the backlog every week. I don't really review redirects, I perhaps did 100 at most. It was good that the updates mentioned the backlog of redirects, it encouraged me to do a few, and to do about 300 user pages too - the amount of self-promotion there is crazy! Good to see so many people getting involved in this drive. Boleyn (talk) 06:50, 29 January 2018 (UTC)
+1 Atsme📞📧 15:41, 31 January 2018 (UTC)
Nice work! You can see the flagging in the end Galobtter (pingó mió) 08:34, 29 January 2018 (UTC)
  • Update: Having reviewed and/or chatted with a number of the top reviewers during the drive, I believe the 'Top Article Reviewer' was JTtheOG, having reviewed roughly 1800 articles. Congrats. I will be slowly doling out the awards over the next week or so, as I am also checking for review quality by re-reviewing a small sample from each reviewer prior to handing out the Barnstars for all those who reviewed 100 or more articles. So if you haven't heard anything yet, I just haven't got down the list that far yet. — Insertcleverphrasehere (or here) 19:28, 31 January 2018 (UTC)
I have finally completed giving out the awards. Congrats again to everyone that participated. Note that the backlog has basically been static since the end of the backlog drive, so if we intend to meet our goal of reducing the backlog by March 14th, we are going to need to pick up the pace a bit (we need a reduction of about 100 articles per day). — Insertcleverphrasehere (or here) 04:04, 10 February 2018 (UTC)
In fourth week, 155 reviewers did the reviewing. So i think, now we have come to the "50 reviewers doing most of the work" phase. But still, even if we review just seven articles per day, we are talking about 350 articles per day. We should try to do 10. —usernamekiran(talk) 05:01, 10 February 2018 (UTC)

Moving to Draft

Latest comment: 6 years ago11 comments5 people in discussion

Ok, so John_Thompson_(businessman) is clearly not ready for mainspace, so other than his former job position there's no info. Is it better to speedy delete or send to draft? The creator was CU blocked so it seems draft is a wasted effort. Atsme📞📧 00:37, 3 February 2018 (UTC)

...Aaaaand it's gone. — Insertcleverphrasehere (or here) 04:48, 3 February 2018 (UTC)
Magical. :) For some reason, my speedy requests end up with tag being removed, not the article. Is there a way I can check how many speedy deletes I've requested vs how many were actually speedy deleted? Atsme📞📧 12:03, 3 February 2018 (UTC)
Yes, Atsme, by seeing how many blue links there still are in your CSD log. Kudpung กุดผึ้ง (talk) 12:07, 3 February 2018 (UTC)
Atsme I believe that CSD noms made with Page Curation do not show up in the Twinkle CSD log, and I know Twinkle allows you to choose to not notify the page author. L3X1 ◊distænt write◊ 15:21, 5 February 2018 (UTC)
'Nother question - is there a way to add an endif or other command to the curation tool so that notices are not automatically posted to a CU blocked user/sockpuppet? Atsme📞📧 14:40, 3 February 2018 (UTC)
Nope,not that I know of. Anyway, I don't think it's an issue that needs the attention of the 'overworked' devs ;) Kudpung กุดผึ้ง (talk) 06:05, 4 February 2018 (UTC)
Kudpung Requested updates to the page curation tools are a f***ing joke at this point. Honestly the silence is so deafening with regards to our requests at Wikipedia:Page_Curation/Suggested_improvements that we would probably be more successful just getting some talented wikipedian to rewrite the page curation tools as a script. Why the devs wrote it as a back-end program in the first place is beyond me. What a waste of everyone's time. — Insertcleverphrasehere (or here) 03:20, 5 February 2018 (UTC)
The silence is defening because I'm no longer pushing the rickety waggon along the windy cobbled street of the list I created at Wikipedia:Page_Curation/Suggested_improvements. As you know already, experience has shown that the more we complain, the more they deliberately put these things on a back burner. The development teams at the WMF only develop software they want to develop, and often it's not even what the community asked for. Remember also that they are first and foremost software engineers, they are neither businesspersons, nor do they have the required managerial skills for knowing what the stakeholders need or how to set priorities. I have spent many years on the NPP/NPR project, now it's up to someone else to take the initiative - I got the Page Curation system developed (although I was in no way responsible for its actual coding), then the NPR user group, and then ACTRIAL, and and now , finally, after m ore than a year and a half, the backlog is down from its ridiculous heights to something slightly more managable. I firmly believe however, that such a large and critically important software solution is the responsibility of the WMF to develop and maintain - there was never any talk when Wikipedia was created that the editors should also write the software, especially when people are being paid salaries for doing it. Now I'm off doing something else to protect the encyclopedia - as they say: a change is as good as a rest. Kudpung กุดผึ้ง (talk) 05:35, 5 February 2018 (UTC)
I recently complained about this (User_talk:DannyH_(WMF)#Page_Curation_updates_long_overdue.), with little success. It seems that DannyH (WMF) (who leads the community tech team) expects us to ask via the wishlist if we want fixes to the software they developed, and they have no intention of keeping it up to date otherwise. — Insertcleverphrasehere (or here) 05:22, 10 February 2018 (UTC)
They won't even fix bugs. Fixing bugs is not creating new tools or adding features to existing tools and should not be subject to voting. Mduvekot (talk) 14:45, 10 February 2018 (UTC)

Backlog as of 12th February, another visual snapshot.

Latest comment: 6 years ago1 comment1 person in discussion
 
NPP backlog, number of unreviewed articles by creation date. Red is older than 90 days, orange is between 30 and 90 days old, green is less than 30 days old.

Our goal of reducing the backlog by March 14th is, as of today, only one month away. As you can see, we have had a bit of a pileup at the front of the backlog over the last few weeks. During that period, the backlog has remained relatively flat, so we are actually making good progress toward our March 14th goal, even if it might not look like it from the raw number of articles in the backlog. There has been a lot of reduction at the back, and if we are going to have a backlog, a front heavy one is the best we can hope for. Thanks to those who have been working at the back, keep up the good work!

For a recap: articles older than 90 days are indexed by Google even if they have not been reviewed. Therefore, the red articles above are currently indexed and should be our immediate priority. By March 14th (the end of the initial phase of ACTRIAL) most of those articles marked in orange will also be in the indexed red zone, so they will have to go by March 14th as well. The green articles can be largely ignored for now, although a reduction there would certainly help us avoid hitting a cliff once that wall works its way back to the 90 index point at the end of April.

I hope this helps illustrate (literally) the challenge we face. It is important that we meet these goals, because once ACTRIAL's initial phase ends on the 14th of March, we will have at least one month of time where non-autoconfirmed editors will be able to create new articles (previously created 300 or so extra articles per day). Now that I have finished reviewing the backlog drive, I am heading back to the coal face as well. Cheers and happy reviewing! — Insertcleverphrasehere (or here) 03:09, 12 February 2018 (UTC)

review requested

Latest comment: 6 years ago2 comments1 person in discussion

Could someone please review Kandankarykkavu Devi Temple? I could have done it, but I dont know how to communicate with the author. —usernamekiran(talk) 11:36, 14 February 2018 (UTC)

done by Serial Number 54129. —usernamekiran(talk) 20:51, 14 February 2018 (UTC)

NPP Flowchart, marking pages as reviewed when deletion tagging.

Latest comment: 6 years ago6 comments4 people in discussion

Note: I just updated the NPP flowchart to connect all deletion nominations to 'mark article as reviewed', per this discussion a while back, and common usage. Note that you should keep track of nominated articles via PROD/CSD logs if possible to check that the CSD templates are not removed by the authors and decide if escalating a removed PROD is necessary. If you are using Twinkle for deletion tagging, you can activate the logs by creating userpages at Username/CSD log, and Username/PROD log. If you use the Page Curation tools for deletion tagging, you can use your deletion tag log. — Insertcleverphrasehere (or here) 11:32, 13 February 2018 (UTC)

How can you prevent an editor from just recreating an article after a version has been deleted? I recently came across an article (Zuzana Vačková I think) where several previous articles, each written under a slightly different title, had been successively created, and each deleted in turn. Another problem I have come across is articles turned down at AfC, which are soon afterwards published despite this. Cwmhiraeth (talk) 14:30, 14 February 2018 (UTC)
Keep deleting and salting the offending articles. Serve Grindcomber with a warning for creation of inappropriate pages. The next time they do it, ask an admin for action (don't waste time at ANI). Kudpung กุดผึ้ง (talk) 15:53, 14 February 2018 (UTC)
​Template:Salt is useful when tagging an article that has been repeatedly recreated. Nothing to about people that recreate at different titles except get an admin to block them and keep deleting the offending articles by checking the user's contribution history. Nothing actually stops a user from just creating an article that was turned down at AfC, and can in some cases be desirable (if the reviewing AfC editors were being overly strict). Basically just treat these articles like any other and judge each article on its own merits. — Insertcleverphrasehere (or here) 17:36, 14 February 2018 (UTC)
Thank you both. Did you see the cunning edit made by this user at 18:58 on 21 August 2017‎? Cwmhiraeth (talk) 18:28, 14 February 2018 (UTC)
Yes, salting is good; but it doesnt always work. I came across an actor who kept on creating article about himself. After the name was salted, he created it as "<name> (actor)", then "dancer", and recently he became a wrestler too. Sometimes, its just better not to salt an article so that it can be monitored easily. —usernamekiran(talk) 20:56, 14 February 2018 (UTC)

Lingotek

Latest comment: 6 years ago5 comments4 people in discussion

Lingotek is not a new article but for some unknown reason I had it on my watchlist. It underwent a large expansion yesterday and I thought it very promotional so I checked it for copyvios and found it was riddled with them. I have only ever nominated an article for speedy deletion with the page curation tool, which I could not use in this case. I wanted to nominate it for G11 and G12, but could not discover how to combine them, so in the end just went for G12. How do you nominate an article for speedy deletion for more than one reason? Cwmhiraeth (talk) 11:07, 16 February 2018 (UTC)

You can use either twinkle or {{Db-multiple|G11|G12}} ​ Galobtter (pingó mió) 11:12, 16 February 2018 (UTC)
Thanks. I was going to adjust the template but I see you have taken the matter into your own hands. Cwmhiraeth (talk) 11:22, 16 February 2018 (UTC)
A prime example of what I was on about here. This totally new 30K page, overwriting a previous article, was created by an IP [1] in violation of multiple policies but it wouldn't have come up for review as a new page: Noyster (talk), 12:27, 16 February 2018 (UTC)
The call to {{db-multiple}} should be {{db-multiple|G11|G12|url= |url2= |url3= }} with the URLs of the copied webpages specified. Cabayi (talk) 16:11, 16 February 2018 (UTC)

few articles with questionable notability

Latest comment: 6 years ago2 comments2 people in discussion

Hi.
Makhamakhi has created many articles, a lot of them have questionable notability. But this is my opinion. A second look is requested. Thanks, —usernamekiran(talk) 19:39, 16 February 2018 (UTC)

Sheesh, that's a lot of Indian cinema... rather mixed bag, some of that stuff seems well notable, some rather doubtful. As of yesterday, there won't be any more coming from that source at any rate. --Elmidae (talk · contribs) 10:23, 17 February 2018 (UTC)

Auto notice of deletion

Latest comment: 6 years ago3 comments3 people in discussion

Is there any way to disable the auto notice of a deletion going to blocked creators? It seems a bit like grave dancing when they have lost TP access on top of it all. Atsme📞📧 13:57, 23 February 2018 (UTC)

If you're talking about notifying the editor when you nominate a page for deletion, yes, there's a checkbox that can be un-checked. If you're talking about a watchlist notification, there is zero way for us to change that. Primefac (talk) 17:10, 23 February 2018 (UTC)
Such notices may be read by others besides the blocked editor. I often notice deletion notices on the talk pages of editors on my watchlist. If they are inactive for some reason then I will usually investigate so see whether the nomination is sound or not. For example, Uncle G only edits occasionally now and so when some articles of his were prodded recently, I checked them out. It seems especially helpful to do this in the case of blocked editors because they will not be able to act themselves. The important thing for the project is that the content gets good attention. Andrew D. (talk) 18:13, 23 February 2018 (UTC)

Backlog Raising to over 3700 again

Latest comment: 6 years ago11 comments5 people in discussion

The backlog has been raising recently, possibly due to a recent influx of additional articles per day (on the article creation stats charts, it seems like we have had a bit of an increase in article creation rate in February compared to January). It also seems like there have been more articles being created from redirects in recent days, though that might just be a lack of people reviewing the back of the backlog of late. :(

In any case, please consider reviewing a few extra articles per day. — Insertcleverphrasehere (or here) 09:00, 20 February 2018 (UTC)

The number of unreviewed redirects is also up over 8,600 as well (up from 5,800 on Jan 29th). — Insertcleverphrasehere (or here) 09:25, 20 February 2018 (UTC)

Honestly speaking, I've never reviewed redirects; and my review rate has gone down since like 3-4 months. After 2-3 days from now, I'll try to maintain an average of 10-15 pages per day. I think in the next newsletter instead of a vague "few articles per day", we should add some definitive number. Not a big one, but a small one like 10 pages per day. With emphasis on calculation, like "we have 200 reviewers, if we did only 10 pages per day, that would be 2000 pages per day." I mean, if we make it vague, most of editors wouldn't review. Same would go with some big/unrealistic number. But a small number might make them review pages. —usernamekiran(talk) 13:45, 20 February 2018 (UTC)
The backlog goes up and down, but the overall trend of the last 6 months is going down with periods where it rises a few hundred and platueaus before decreasing. I’m not particularly worried at this point. I also don’t really track the redirect backlog at all. I appreciate the work people do there, and if they enjoy it, great, but the main focus of NPP has always been mainspace articles. We’re doing pretty well on that count currently, so I’m less concerned with upticks on redurects or small increases in the backlog. TonyBallioni (talk) 14:43, 20 February 2018 (UTC)
​I wouldn't be worried either if we didn't have the end of the first phase of ACTRIAL coming up. As for redirects, the major concern is good articles being converted to redirects inappropriately, though most of those are caught by people with the page on their watchlist. — Insertcleverphrasehere (or here) 19:20, 21 February 2018 (UTC)
  • The backlog curve which is now back up to 3962 , clearly reflects the lack of interest since the drive ended. Let's be clear about this, while the drive helped enormously, 3962 is still an unacceptable number - just imagine if we had that many AfD in a backlog! Kudpung กุดผึ้ง (talk) 14:07, 24 February 2018 (UTC)
It's amazing, isn't it? I take a quick look at the feed to check the backlog number and the very first article on the top of the list is a blatant promo with all the hallmarks of PE (posted in one after 14 edits). That said, I wonder how many other reviewers would have caught it. Kudpung กุดผึ้ง (talk) 14:26, 24 February 2018 (UTC)
I saw it peak above 4000 today before dropping a bit again, but at least there has been a lot of work at the back. The current back of the backlog is at 1st November. We have 26 more days worth of articles to review at the back and we will remove all of the indexed backlog. There are currently ~450 indexed articles at the back of the backlog. If we can crush those in short order, hopefully we will be able to work on it one day at a time and make sure nothing gets indexed again. we also need to reduce the overall volume of the backlog, but it is most important to remove the indexed articles. I'll prepare another visual snapshot soon, but I am a bit concerned about the recent lack of front-line reviewing, February has a lot of days with over 100 unreviewed articles, which will make it difficult for us when this wall hits the back of the backlog in a couple months. — Insertcleverphrasehere (or here) 11:16, 25 February 2018 (UTC)
@Kudpung: The article you prodded seems to have been OneFactor. Now I have hardly been prodding anything because most biographies have at least one reference, and it is so easy for the creator to annul the PROD, so it seems pretty pointless. I would probably have put a CoI tag on this article, and put a notability tag on it or nominated it for AfD. Comments? Cwmhiraeth (talk) 14:25, 25 February 2018 (UTC)
No. Except to say that its undeclared paid editing and it's not a BLP. Just keep watching it if you're interested. Kudpung กุดผึ้ง (talk) 17:22, 25 February 2018 (UTC)

Then major difficulty of our analysis is that the chart we have (I usually use it as reference), has big gaps in it date-wise. So we can't exactly observe the flow. Basically, observation/analysis depends on weekly activities, and a little bit on monthly. I mean, there are two weekends, and 5 working days. We have to factor in the activity of page creators, and page reviewers. It is not only about reviewing. For example, say, we are reviewing 800 pages per day on weekdays, but 500 on weekends; and if the creators are creating 200 pages on weekdays, but 400 on weekends: then this scenario will show the backlog risen on Monday morning. What I am trying to say here is, without short term analysis made on long term, we cant make accurate predictions if the backlog is rising or falling. Maybe these short rises/falls are expected with some logic. Then there are factors like "drive" which cause anomalies. That also reminds me, I think now it is time to update {{New Page Reviewer granted}}. I will shortly initiate a discussion regarding that with sample(s). —usernamekiran(talk) 18:00, 25 February 2018 (UTC)

Discussion about a bot to create new stubs

Latest comment: 6 years ago1 comment1 person in discussion

A discussion has been posted at Wikipedia:Village_pump_(proposals)#A_Bot_for_Creating_Arthropod_Stubs,_Trial about a bot for creating stubs. All are invited to participate. TonyBallioni (talk) 03:53, 26 February 2018 (UTC)

New Page Reviewer granted (template)

Latest comment: 6 years ago10 comments4 people in discussion

Hello, I changed a few things in the notifying template that is posted on the talkpage of editor after the flag being assigned. Kindly take a look at it, also please suggest any changes if you have something in mind. —usernamekiran(talk) 18:18, 25 February 2018 (UTC)
User:Usernamekiran/New Page Reviewer granted

Update (per Lex's advice): Here is the change: special:diff/827623110. —usernamekiran(talk) 20:14, 25 February 2018 (UTC)
As a word of advice, if you don't want to edit the template directly, you may want to create a sandbox mirror ([2]) with the original content, then make another edit to the mirror, and then post the diff here. It's rude and frustrating otherwise, since other people wouldn't immediately know what's being changed without going back and forth. Alex Shih (talk) 19:00, 25 February 2018 (UTC)
Thanks :) —usernamekiran(talk) 20:14, 25 February 2018 (UTC)
Usernamekiran Looked over it briefly, my first comment is that it should be "while" not "with" as the first word in the last line. — Insertcleverphrasehere (or here) 21:22, 25 February 2018 (UTC)
@Insertcleverphrasehere: erm... yeah, no. If "while" is used, then the primary task (in the sentence) would be maintaining the quality. But we are trying to tell them to review as many pages as possible without losing the quality. But if you thought it was a grammatical mistake, then most of the people will do the same. :-/
—usernamekiran(talk) 21:30, 25 February 2018 (UTC)

IMO these changes are unnecessary: 1) It makes it sound as if 5000 is an acceptable level. It is not. 2) The word 'kindly', as used in native English speaking regions, infers a sense of demand, command or obligation This is not appropriate when calling for voluntary assistance. Kudpung กุดผึ้ง (talk) 22:36, 25 February 2018 (UTC)

Also this template shouldn't be treated as a newsletter; the content is quite bloated as it is I think. Alex Shih (talk) 04:20, 26 February 2018 (UTC)
I think this template is only sent once, when the useright is granted to the user. Agree that it shouldn't be used other than for that. — Insertcleverphrasehere (or here) 04:24, 26 February 2018 (UTC)
Yes. I am saying the same thing. (To editor Insertcleverphrasehere: not intended to be sent again, just for welcoming the users.)
What my point here is, for most of the editors this template is the first thing about NPP/R that they come around other than the tutorial. I think we can use this template to let them know about current events regarding NPP/R. Or maybe we should update the tutorial. I had a few ideas regarding that tutorial update, but I forgot.   Now that I think about it, I think we should make minor changes in the template per Kudpung, and Lex. I will create a draft for the two separate tutorials in my userspace, and then I will let you know guys about it here. Kindly give me 42-72 hours for that. But still, as long as we are discussing this, what changes should we make in the welcoming template? —usernamekiran(talk) 19:11, 26 February 2018 (UTC)

I made a boo-boo

Latest comment: 6 years ago4 comments3 people in discussion

I sent an incomplete article to draft not thinking about the creator being blocked. I rolled back the edits and now I can't find it - prolly gonna take an admin's help, huh? Woke Up Like This (film). Apologies. Atsme📞📧 01:08, 27 February 2018 (UTC)

Restored to mainspace. TonyBallioni (talk) 01:11, 27 February 2018 (UTC)
I thought making boo-boos was my area of speciality. —usernamekiran(talk) 02:08, 27 February 2018 (UTC)
Well, we must be why Tony keeps a box of band-aids handy.   Atsme📞📧 02:20, 27 February 2018 (UTC)

28th Feb Backlog visual snapshot

Latest comment: 6 years ago14 comments7 people in discussion
 
NPP backlog, number of unreviewed articles by creation date. Red is older than 90 days, orange is between 30 and 90 days old, green is less than 30 days old.

Hi guys and gals, another visual snapshot of the backlog for us to have a gander at. The backlog has remained stubbornly unchanged around 3700 during this time, give or take a few hundred, which is sill a very large and unsustainable number long term. It isn't all bad news, as we are very close to removing all google indexed articles from the backlog (only those marked in red above; or about 390 articles). Looks like there has been some serious work at the back of the backlog since the last snapshot on the 12th and a general transfer of unreviewed articles from the back of the backlog to the front. Our goal before the 12th of this March should be to review these articles in the red as a top priority, and make as much of a dent into the orange section as possible.

While reviewing from the back is very important, we have now reached the point where nearly all articles in the backlog are non-indexed, and we need to reduce the numbers of new articles coming in as well. Reviewing from the front or back require a different set of skills, so please do whichever suits you best, but as always, reviewing from the back is very much appreciated because few reviewers have the skills needed to review the articles that tend to fall to the back (often time consuming judgement decisions).

As always, thank you all for your hard work. — Insertcleverphrasehere (or here) 03:25, 28 February 2018 (UTC)

I was just looking over what's in the queue, from oldest first - I didn't get an actual number, but there must be hundreds of stubs created by blocked users/socks. Many have the same stub format, and consist of films, tv episodes or sports such as 2018 Copa Truck season which is a schedule for 2018 events. I looked at the edit history, and approx. 3 reviewers and 1 admin edited that article, so why is it in the feed? It has no refs, it's unsuitable for article space, so do we just skip over it so it's not indexed and watch the backlog grow? That is pretty much what I'm seeing in the first 100 or so articles in the queue. Some are sourced in a language other than English which may explain why reviewers skip over them. I still can't, to this day, figure out why en.WP allows sources in a different language. We now need language specialists on our NPP team. I try to help over at AfD but I can imagine how loud the screams if we start shooting a bunch of these articles over to AfD (where most belong). Maybe I'm being overly critical, but articles like Swing Kids (2018 film), Woke Up Like This (film) and TSYD Challenge Cup are classic examples of what we're dealing with and what I see may be causing a big part of the backlog. Surely there's a more expedient way to deal with these kinds of articles? Atsme📞📧 03:36, 1 March 2018 (UTC)
​Lack of confidence among reviewers is an issue, because it creates a situation where multiple people review a page without taking action. I personally make it a habit to always review a page fully when I have looked at it, except in very rare circumstances. I do this to reduce repeat reviewing. If I am confused what to do first, I fall back on the NPP flowchart that I developed a while back (yes even though I wrote it, I still find it useful to find an answer where I get confused). Note that in some cases an article can be marked as reviewed, but then get unreviewed again due to redirect/unredirect which can make it seem like it was never reviewed at all.
As for your specific examples:
  • Swing Kids (2018 film) is borderline, but OK. The sources (using google translate) may not be the best quality (I have no idea what news sources are reliable in South Korea), but they don't send giant red flags up to me. I think it would survive AfD, and is certain to receive more coverage in future. -->Marked as reviewed
  • Woke Up Like This (film) has some dodgy sourcing, probably not enough to deomnstrate notability alone, but a quick search indicates quite a few reviews and other commentaries on the film in the news. Would likely survive AfD if anyone did the search like they are supposed to (WP:BEFORE). -->Marked as reviewed.
  • TSYD Challenge Cup has no sources, and nothing jumps out at me in searches. Content of this topic generally applies to the TSYD Cup article and could be merged there, but there really isn't anything here to merge. I would PROD it if the title wasn't a plausible redirect candidate. -->converted to redirect to TSYD Cup and marked for review.
Does that help? — Insertcleverphrasehere (or here) 04:37, 1 March 2018 (UTC)
​As for non-English sources, they certainly make our job harder, but non-English sources definitely DO contribute to notability. They have to if we want to avoid a massive systemic bias issue. — Insertcleverphrasehere (or here) 04:42, 1 March 2018 (UTC)
Systemic bias? If we are an English encyclopedia, why would it be considered systemic bias to expect verifiability in English? Why isn't the onus of verifiability on the creator of the article - as in, quite simply, provide English sources for the English encyclopedia or create the article in the respective language WP? I am not biased to anything or anyone...(except my grandchildren and NPOV)...so if the article & sources are in English, wonderful!! If they're in another language, we have other language projects, right? What am I missing here? Our readers are English speaking readers so how do sources in another language they don't speak benefit our readers? Atsme📞📧 01:54, 3 March 2018 (UTC)
​En.Wikipedia is not an encyclopedia of English-language related topics, it is an encyclopedia of all notable topics, regardless of language origin. Other language projects follow the same logic. The German Wikipedia does not only cover topics covered in German-language sources, its purpose is also to cover all notable topics, but in the German language. — Insertcleverphrasehere (or here) 05:22, 3 March 2018 (UTC)
Honestly, I'm a little shocked by this display of parochialism. There are tens of thousands of well-developed articles, espcially on geography and history, that would remain stubs or be completely absent if sourcing was limited to the English language; and for which no coverage in other WP projects would be available for English-speakers either because of complete inaccessibility of not only source texts, but actual article text. What you are proposing is, in essence, locking readers of enWP into an artificial information bubble determined by coverage language. That is a more extreme systemic bias than anything enWP is already labouring under. --Elmidae (talk · contribs) 07:38, 3 March 2018 (UTC)
This is an encyclopedia written in the english language - that's all. Sources aren't too much for the readers anyhow and offline/paywalled sources are generally as useless to the reader as something written in swahili Galobtter (pingó mió) 08:47, 3 March 2018 (UTC)

Instead of looking at it that way, how about thinking about our English speaking readers whose purpose we're supposed to serve, and maybe come up with a plan that auto-translates at least the titles of the sources so those of us who are reviewing them, including our readers, will know if they're legitimate of promotional, and if it's worth taking the extra step to attempt some other form of translation to actually read the source? What good does it do to have the article if we can't verify it? Step outside the box - look for resolutions to issues, not ways to criticize the editor who shows concern. Jiminy Cricket. Atsme📞📧 14:37, 3 March 2018 (UTC)

​Install Google Chrome, it has a built in google translate feature that cuts several steps out of reading non-english sources. — Insertcleverphrasehere (or here) 15:03, 3 March 2018 (UTC)
I am firmly opposed to excluding non-English sources, translating titles or otherwise changing the content of articles for the sake of making them easier to review. As for practical solutions, I would suggest requesting help from our multilingual reviewers. Say you need a native Dutch speaker, a PETSCAN query on "Has all of these templates:" with the templates {{User wikipedia/New page reviewer}} and {{User nl}} will find me and I'd be happy to help evaluate the Dutch sources. If no such user exists, we can write to the project for the relevant country to see if there's someone there who can help. Mduvekot (talk) 16:00, 3 March 2018 (UTC)
Meh...it's easier to just skip over them and review articles written/sourced in English. I was just trying to help reduce the backlog which is what this discussion is all about. We're all volunteers and we each have so much time to devote to editing, fixing, creating, researching sources and do all the other things that need to be done on WP. I'm happy to know there are editors who have such time and are willing to invest it in foreign language translations. I think it may prove more beneficial time-wise in reducing the backlog if those editors who are linguisticians and/or skilled in a particular language to go through the queue, and review those articles so that others can continue reviewing without stopping and wasting valuable time to go find such editors. I'm of the mind that we utilize the individual talents of our editors where they can be most productive. I'm most productive with English...possibly some Spanish. Oh, and there's also a backlog growing in AfC which could use some help in those same areas. I suppose that would be easer than trying to get the WMF to do anything "automated" in an effort to speed up the process and lighten the work load on it's shrinking slave volunteer labor force.   Atsme📞📧 16:52, 3 March 2018 (UTC)
According to the WMF, ACTRIAL makes NPPers not come in to work. Also according to the WMF, we don't need to communicate on TPs anymore, we can do on the history pages in our edit summaries. L3X1 ◊distænt write◊ 17:56, 3 March 2018 (UTC)

Well, getting back on topic a bit... thank you for continuing to produce these graphs Insertcleverphrasehere. I find them a very useful way of visualising where we are with the backlog. Clearly there's been a bit of fatigue after the drive. On the bright side, we are at least keeping on top of it (i.e. the backlog isn't growing), and the articles in that green hump will probably be easier to review than the stubborn ones at the back. – Joe (talk) 19:51, 3 March 2018 (UTC)

Latest comment: 6 years ago6 comments2 people in discussion

Can someone give a pointer about what to do with this list consisting entirely of redlinks? What's giving me pause is the apparently preceding discussion that resulted in dumping the lot into mainspace, for reasons that are opaque to me. It seems people consider that there is some base for considering this a valid article, but I'm not seeing it...? --Elmidae (talk · contribs) 14:56, 5 March 2018 (UTC)

The redlinked articles were draftified and apparently have now been G13d (for the second time). A list of redlinks is not suitable for Wikipedia. Primefac (talk) 15:23, 5 March 2018 (UTC)
Righto. So... G-8 ("dependent on a non-existent or deleted page")? Or PROD? --Elmidae (talk · contribs) 15:31, 5 March 2018 (UTC)
WP:A3/blank, since there was nothing other than "this is a list" which duplicates the title. Primefac (talk) 15:32, 5 March 2018 (UTC)
Much obliged! Any idea why the folks in the discussion felt that this needed to be moved to mainspace? I'm thinking I'm missing some underlying reasoning there. --Elmidae (talk · contribs) 15:34, 5 March 2018 (UTC)
The list itself was never moved, so the bluelinks became redlinks and I guess no one ever bothered to update it. Primefac (talk) 15:37, 5 March 2018 (UTC)

286 15 0 articles

Latest comment: 6 years ago8 comments4 people in discussion

We have 286 15 0 articles that will be still be indexed on March 14th (at the back of the backlog and aged older than December 14th). Not a tough ask, but I think it is essential that we at least reach this goal of having nothing indexed by the time of the one month 'open' phase of ACTRIAL. From the middle of December to the middle of January there are actually relatively few articles in the backlog (an average of less than 5 per day), so over the month of the next phase of ACTRIAL we should be able to keep up easily and keep from hitting the noindex point if we can review the 286 articles that are aged older than December 14th. Anyone who can review a few spare articles at the back, please do so, and I will put my best foot forward as well. Cheers. — Insertcleverphrasehere (or here) 23:24, 6 March 2018 (UTC)

Thanks Elmidae, for helping out at the back today. Cheers. — Insertcleverphrasehere (or here) 05:01, 7 March 2018 (UTC)
And also thanks to Graeme Bartlett for reviewing some of the back today. — Insertcleverphrasehere (or here) 08:46, 8 March 2018 (UTC)
BING!  :) --Elmidae (talk · contribs) 07:59, 9 March 2018 (UTC)
Cheers mate! Thanks for finishing it off. Also want to thank Cwmhiraeth for helping out a bit earlier in the day with this last push. — Insertcleverphrasehere (or here) 08:47, 9 March 2018 (UTC)
Congratulations everyone. Well done. I'd like to take to take the opportunity to point out the the back of the queue contains articles that really needed patrolling. I just stumbled upon Oksana Tanasiv. Just because such an article "survived" for 90 day does not mean that it should have been indexed, or even allowed to continue to exist as a blatant violation of What Wikipedia is not. Mduvekot (talk) 22:55, 9 March 2018 (UTC)
Yeah, I found 7 CSD candidates, 8 AfD's, 7 PRODs, a few that were converted to redirects, and a number of copyright violations in need of revision deletion. I probably reviewed about 2/3rds of the above number, so that's a ratio of at least 10-15% in need of serious action. Redirects are often converted to new articles, and these are dropped at the edge of the back of the backlog, and are often indexed immediately, so it is important to review these promptly. I have been doing quite a lot of these redirect conversion reviews over the last couple months, and recommend to anyone doing them to check the page history before doing anything else. — Insertcleverphrasehere (or here) 00:37, 10 March 2018 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:New_pages_patrol/Reviewers/Archive_19&oldid=833376652"