Wikipedia talk:New pages patrol/Reviewers/Archive 14

From WikiProjectMed
Jump to navigation Jump to search
Archive 10 ← Archive 12 Archive 13 Archive 14 Archive 15 Archive 16 → Archive 20

WARNING: WordPress scam/spam/malware

Please beware of any new pages about WordPress blog software.
New article What Can Wordpress Plugins Do For You is being posted by multiple sock accounts and contains multiple spam and/or malware links. It's expected to continue as several sleeper accounts have been discovered.
New pages of this kind may have slightly different titles. Please tag any such pages immediately with G3 (vandalism) and report them either immediately to an admin who is online or to AIV where they will get fairly prompt action. Do not under any circumstances click on the embedded links - to do so may damage your computer. (TonyBallioni please include this warning in your next NPP newsletter). Kudpung กุดผึ้ง (talk) 00:06, 31 August 2017 (UTC)

I suppose there isn't a way to search all existing drafts for "Wordpress"? TimothyJosephWood 00:07, 31 August 2017 (UTC)
I think this is the search you want: [1] EEng 00:29, 31 August 2017 (UTC)
That is a truly terrifying search for reasons beyond the scam TonyBallioni (talk) 01:18, 31 August 2017 (UTC)
are we talking about articles like these: Ασφαλεια αυτοκινητου τριμηνη Allianzgyhozenzapu65210.? I came across 3 articles like these. All of them were tagged for speedy within few minutes of creation. All the creators account were named after girls. And I dont think searching for drafts is useful Tim, as I believe they do it in one go.
If there are few more articles found like these, I think a newsletter to NPPR guys should be issued immediately, what do you think Tony? —usernamekiran(talk) 08:29, 31 August 2017 (UTC)
usernamekiran, could you please check that link again please. It doesn't go anywhere. If the page was deleted I should be able to see it. Thanks. Kudpung กุดผึ้ง (talk) 09:13, 31 August 2017 (UTC)
Checked against UNK's CSD log, link was missing final full stop. Cabayi (talk) 09:19, 31 August 2017 (UTC)
Thanks Cabayi. @Kudpung: You should be able to see it now. —usernamekiran(talk) 11:30, 31 August 2017 (UTC)
Yes, that is classic spam, although probably not connected. Always list the creator of such rubbish at AIV for blocking or better still let an admin know, because as in this case, I'll have it checked for socks and sleepers. This sort of thing will probbly disappear when ACTRIAL is rolled out.Kudpung กุดผึ้ง (talk) 11:48, 31 August 2017 (UTC)
Regarding always listing at AIV, I have not personally done this, I just use the NPP toolbar, and four entries appear in my watchlist, whereupon my eyes glaze over, and I have no idea if AIV has been notified. (checked before posting. AIV not notified) Should I notify as well? -Roxy the dog. bark 12:59, 31 August 2017 (UTC)
The toolbar does not list vandalism authors at AIV. You have to do it manually. But that's a good idea - obvious blockable creators of new pages should be able to be listed quickly for blocking. Kudpung กุดผึ้ง (talk) 13:30, 31 August 2017 (UTC)

Have to say that admins are on the ball, things that I have highlighted get dealt with quickly, and mistakes I have made noted for me as well. -Roxy the dog. bark 13:40, 31 August 2017 (UTC)

  • As a response to Kiran's question, I'm not familiar enough with the conversation to answer the likely questions I would get on my user talk if I were to send it out myself separately from the newsletter. If Kudpung feels it is appropriate to notify everyone now, as an admin and the coordinator emeritus of our project, he certainly can do so. TonyBallioni (talk) 15:00, 31 August 2017 (UTC)
It's nothing that can't wait for the next newsletter. The phenonenon is not new. Anyway, these things rely on being able to dump their crap spontaneously in the encyclopedia so perhaps making them wait four days when ACTRIAL goes live might dissuade them. Kudpung กุดผึ้ง (talk) 19:31, 31 August 2017 (UTC)

What about NPP review of (nearly blank) pages or pages with empty skeletons?

I've been patrolling the older end of the User: namespace (10 to 30 days old, usually). One thing I run across frequently are pages that are essentially blank (may have a sandbox template or something similar) or that have just the Article Wizard skeleton of an article. So far, I've been passing these by without marking them as patrolled.

Some of these pages are eligible for speedy after a year of inactivity by their creator (e.g. {{db-blankdraft}}), but that's never true of the pages I'm looking at.

Since they're not breaking any rules with their current lack of content, should I mark them as patrolled? Or does that put undue responsibility on Recent Changes Patrol to catch things that are later added to these pages? It strikes me that leaving them hanging around in the NP feed might just waste more time for other NPP reviewers. — jmcgnh(talk) (contribs) 08:56, 3 September 2017 (UTC)

Hello jmcgnh -- I see the same pattern. First, determine whether you are dealing with a newbie or a WP:SPA. That's occasionally difficult. The experienced editor lodges a trivial two-sentence article. You put a PROD on it and the response is PROD removal and, "Can't you see this is a work in progress? References will follow." My response is this: "Every version of an article in the mainspace should stand on its own, even if it is short."
Case study: user Avery333 published this short article a month ago Bell Resources. It's dodgy. The text claims his company was formed July 2017, yet the wayback machine has stuff from before. Then he publishes an autobiography (my instinct says so) Mark Langer Avery. It's brief with scant references, in short, not notable. And, of course, it depends on Bell Resources. I PRODded it. He reverts the prod, adding one reference. Eventually, this will all come crashing down, but having survived NPP Review and a PROD, he'll be emboldened and won't won't go down without a fight.
Do the right thing and nip the bad ones in the bud. If they are brief, you haven't wasted an author's time. If you let the article grow to 10K characters, then rejection is a disservice. "Justice delayed is justice denied." Rhadow (talk) 09:26, 3 September 2017 (UTC)
Thanks Rhadow for the response. That sounds okay for mainspace review, but the articles in userspace I'm talking about have zero content. Users are allowed to have test pages, partially built content, and a lot of other things on their userspace pages without breaking any rules and, most of the time, I just mark that as patrolled. I catch plenty of would-be autobiographies (for which there's a talk-page template), a few baldly promotional user pages (many of which can be speedied with a combination of U5, G11), and a fair bit of junk that I still don't have a standard response for.
So I'm still asking, what do other reviewers think is the right thing to do for blank or empty skeleton pages in userspace? — jmcgnh(talk) (contribs) 09:38, 3 September 2017 (UTC)
  • Userspace pages are not listed in the New Pages Feed and unless they have particularly toxic content, we generally leave them alone. Due to the huge and sudden unexplained backlog since mid 2016, the effort is to patrol new pages with special/additional focus on the hallmarks of paid editing and other advertorial masquesrading as articles. If you PROD an article, check your PROD or patrol log a few days later to see if the PROD has been removed. If it has, without sufficiently addressing the issue(s), and you are fairly sure it should still be deleted, the next step is to send it to AfD.
That said, with ACTRIAL starting in just 4 days, we are likely to see a significant drop in the creation of dubious pages. We hope so anyway. Kudpung กุดผึ้ง (talk) 09:47, 3 September 2017 (UTC)
If they index their user page this gives them an instant route to Google. We have no policy against it, and we may see an increase in these cases with ACTRIAL. We'll soon know (EF log): Noyster (talk), 13:28, 3 September 2017 (UTC)
They can't, because user pages still need to be patrolled by a Reviewer before they can be indexed. And fortunately because they don't appear in a feed, nobody bothers to patrol them. Kudpung กุดผึ้ง (talk) 14:48, 3 September 2017 (UTC)
A couple of observations:
Mduvekot Certainly userspace is "noindexed by default", but this no-indexing is easily overridden and Visual Editor offers this option on a plate. And Kudpung, when users do this their userpage does get on to Google search results: Google, if you will, "Gayboy Jürgen" or "Camille G. Weston", and these userpages do not appear to have been patrolled. If there are not more such cases it is because I have removed hundreds of them over the past few months, but would prefer not to have to bother: Noyster (talk), 21:19, 5 September 2017 (UTC)
Yes, Drewmutt put me onto patrolling userspace new pages. I don't think many people are reviewing these. I do some patrolling in mainspace as well, but many of the pages there fall in to a grey area where I'm not yet confident I know whether "tag and release" is appropriate, but don't seem to be CSD candidates, either.
My understanding is that userspace pages are not indexed unless the user requests it explicitly. None of the pages I've run across so far have included that request and I might well look at the page more carefully if it did.
Occasionally, a user sandbox is a draft submitted at AfC. If they seem ready, I move them to Draft: but most of the time, they're not ready - like the one linked above - so I leave them as sandboxes and generally leave an AfC comment about what needs to be improved.
If Kudpung would prefer that I not bother with userspace and stick to mainspace, I'll defer to their judgment. But if I'm to continue to patrol userspace, I'd like a reading on whether essentially blank pages and empty skeletons should be marked as patrolled. — jmcgnh(talk) (contribs) 20:27, 3 September 2017 (UTC)
There is one area where it's worth checking user sub pages: Users who are operating scams by creating pages about living people then demanding money to publish them. It's not very common, but it does happen. Kudpung กุดผึ้ง (talk) 00:51, 4 September 2017 (UTC)
Although it is interesting to note that in such cases, creating the article in userspace has de facto added that text to commons already. Instead of paying, those people could simply register an account and move it to articlespace themselves and there is nothing the scammer could do about it. :) — InsertCleverPhraseHere (or here) 01:16, 4 September 2017 (UTC)

I didn't expect this to be such a hard question!

With respect to userspace pages that are essentially blank or empty article wizard skeletons, is there some reason not to mark them as "reviewed" with the NPP tool? If I can't get a straight answer to this question, I'll start doing this so they stop cluttering up the queue of unreviewed pages in userspace. — jmcgnh(talk) (contribs) 04:57, 4 September 2017 (UTC)

Well, if they're blank they don't have any harmful content, so what's he problem with simply clicking the 'patrol' link? Kudpung กุดผึ้ง (talk) 08:24, 4 September 2017 (UTC)
Thanks. Will do. — jmcgnh(talk) (contribs) 08:39, 4 September 2017 (UTC)
If I understand rightly if we do mark them as reviewed and they are then moved into the main article space by any user including the creator then they will not pop up in the new pages feed and could be referenced without being reviewed. Domdeparis (talk) 17:21, 5 September 2017 (UTC)
Domdeparis, you understood wrongly. Unless the Foundation has got it wrong again with their coding of the Mediawiki software, all pages that are created in, or moved to mainspace are listed in the feed. If you are concerned, consider running an experiment with a user sub page you create yourself for the purpose and see what happens. Kudpung กุดผึ้ง (talk) 23:46, 5 September 2017 (UTC)
I just tested this with the help of Drewmutt. A page he marked as reviewed as a user subpage of mine was marked as unreviewed when it was moved to article space. This works as intended. TonyBallioni (talk) 00:59, 6 September 2017 (UTC)
Thank you, TonyBallioni and thank you, Drewmutt. I hope this clears up some issues. — jmcgnh(talk) (contribs) 02:31, 6 September 2017 (UTC)
Ok perfect so if we come across a sandbox that has been marked as reviewed then no need to unreview it but no point in reviewing it either in the first place as the reviewing disappears as soon as it is moved to the main space and it will come up in the feed. Thanks for the info. Domdeparis (talk) 06:04, 6 September 2017 (UTC)

New feature/user right for new page reviewers

In last few days, I came across a few re-created articles that I requested to be salted. That made me thought this. New page reviewers require to see the creator(s) of a deleted article (or recreated one). Not the content like sys-ops can, but only the creator of the article, currently anybody can see which sys-op deleted it but not the creator. I think this feature would be handy for new page reviewers. As the reviewers are already chosen carefully, I dont think it would be misused in any manner. —usernamekiran(talk) 18:36, 6 September 2017 (UTC)

I am not sure if there is another way, but currently I have to this only on the basis of my own knowledge (oh there was an article about this subject), or by skimming through the history of user's talkpage; which is time consuming, so I dont do it always. —usernamekiran(talk) 18:39, 6 September 2017 (UTC)
usernamekiran Xtools might help. You can see a list of articles you created (including deleted articles) here.
There was talk of adding a feature (a small icon or something) to to the feed to denote a new page with a title that had already been previously deleted. It would be more complicated for such a script to show the name of the creator.I would think when clicking on the history the reason for the deletion would be enough. Check out the 55 requests here and if it's not there, please add it. Kudpung กุดผึ้ง (talk) 19:18, 6 September 2017 (UTC)
There is a way to check this using "what links here", but it only works if the creating user didn't clear their talk page. — InsertCleverPhraseHere (or here) 19:48, 6 September 2017 (UTC)
@Mduvekot: Yes, the xtools shows creations of a particular user, but not the creators of particular page. So it isnt much useful if we have multiple creatots/puppets onboard. :-/
@Insertcleverphrasehere: Yes, it only works if the creating user didn't clear their talk page. :-|
@Kudpung: Thanks. I will add it there soon. :)
—usernamekiran(talk) 01:46, 8 September 2017 (UTC)

NPP Queue and the default treatment of aged articles

Today, there are two articles in the unreviewed queue from January 2016. One of them, Avatar 2, appears well formed with 54 references. No reviewer has chosen to click on this one. As a result, it is in limbo, not indexed, available to the experienced WP reader, but unavailable to the typical Googler.

The NPP queue serves as a valuable quarantine. It traps obviously worthless articles. It gives an immature article time to improve. Personally, I think an article should remain in a user space until it's ready for prime time. The result of this viewpoint is that an article should withstand review at its first appearance in the queue. It may be short. I'm not sure it should be allowed to depend on references yet to be found. I know that makes article creation an individual process and not a community process. If an NPP gets a PROD or AfP, the community can and does get involved to improve the article. The worst of the new articles are dumped immediately with A* or G* deletions anyway.

When the reviewer community turns its back on an article, what to do then? Assume good faith and automatically mark reviewed anything in the queue for a week, month or year? Assume that if no reviewer is unwilling to click, that the article has no little intrinsic value, and automatically send it to the dustbin?

The least attractive option is what we do now -- let the article sit in purgatory. Review delayed is review denied. We could start by making the default view of the NPP queue oldest first. I know it's fun to look a the new feed, but forcing a peek at the old ones first would help clear last year's items from the list of 14,429 unreviewed articles. Rhadow (talk) 14:09, 8 September 2017 (UTC)

Rhadow, articles get indexed if they're unpatrolled after 90 days, so most of what you're saying is incorrect. Primefac (talk) 14:14, 8 September 2017 (UTC) (see next line)
Hello Primefac -- Try this experiment yourself: Into Google type Avatar 2 wiki. I did and did not get the article in my results. Rhadow (talk) 14:21, 8 September 2017 (UTC)
Rhadow, please see my comment immediately below this one, as well as the comment by TonyBallioni. Primefac (talk) 14:23, 8 September 2017 (UTC)
Actually, looking at it now, it was significantly expanded earlier today, which puts it back in the "noindexed" pile. Thus, it's essentially a zero-day unreviewed article that just happens to show up at the back of the queue. This happens at AFC sometimes when someone changes a "submit" timestamp. Primefac (talk) 14:16, 8 September 2017 (UTC)
Gah, shoot me for replying before digging. It turns out the article is a copy/paste pagemove of the draft version. I'll do the appropriate histmerging if necessary. Primefac (talk) 14:19, 8 September 2017 (UTC)
  • To expand on what Primefac said, it was transformed from a a redirect into an article today. Anytime a redirect is turned into an article, it registers as a new article and is put into the new pages feed by design. TonyBallioni (talk) 14:19, 8 September 2017 (UTC)
    • It's also a severe copyright violation. For those reading, I'll handle this mess. Primefac (talk) 14:23, 8 September 2017 (UTC)

Hello Primefac and TonyBallioni -- So it appears I stumbled on a special case. Sorry. I retried my experiment of unreviewed items from February 2017. I found them on Google. So I gather the practice is de facto review at ninety days. The unreviewed item remains on the NPP queue, but it is an indexed article in the mainspace. I can simply disregard any item in the list from February 2017 to June 10, 2017, right? Rhadow (talk) 14:34, 8 September 2017 (UTC)

No. An article hits Google after 90 days, but still needs to be reviewed in terms of triage: i.e. is this a copyvio, is this an advertisement (often, yes), does this need tagging for sources, etc. If you want to help out to make sure that less articles get to the 90 day mark, I'd recommend using the NPP Browser to more effectively search for articles that you feel comfortable viewing without having to scroll through the feed. TonyBallioni (talk) 15:46, 8 September 2017 (UTC)

Triage? That I understand. The ones that are going to die anyway, you give them morphine. The ones that need attention, you give them care. What can I do about the ones that are OK? Make them sit in the waiting room again? I don't have the authority to discharge them. Rhadow (talk) 16:19, 8 September 2017 (UTC)

Currently, nothing, since you don't have the reviewer user right. After you have a bit more experience on Wikipedia and feel comfortable with our policies and guidelines, you can apply for it and would have the ability to mark it as patrolled. TonyBallioni (talk) 16:22, 8 September 2017 (UTC)

Does Reviewer Right include getting autopatrolled?

If not, it should. If we trust people to review the articles, they should be trusted to create articles that don't need more review. Cheers, Sadads (talk) 01:33, 7 September 2017 (UTC)

Absolutely not. As TonyBallioni says, they are different skillsets. There are still too many reviewers that still haven't got the hang of things despite the tutorials. We have also been proven recently to have been even over generous with according autopatrolled rights. There are also other serious issues that would make such a suggestion very ill advised. Kudpung กุดผึ้ง (talk) 04:49, 7 September 2017 (UTC)
  • Strong oppose--The skillsets are quite different.And this has been discussed in vain for a lot many times.Regards:)Winged Blades Godric 10:18, 7 September 2017 (UTC)
  • ...Umm... My brain can't brain these comments. If someone can't be trusted to make articles that don't need reviewing, how can they be trusted to review? But having said that, not remotely worth a ten page debate either way... so I guess I'll see myself out. TimothyJosephWood 13:19, 7 September 2017 (UTC)
I know, you'd think--and I've thought--and yet when I went to look into some examples... The only real possibility for implementing this would be to do it in the converse of Sadads's proposal, to cut all reviewers who wouldn't pass autopatrol. And if we think we're short-handed now... (For my sins I'll go review 10 articles right now.) Innisfree987 (talk) 03:49, 10 September 2017 (UTC)
  • My understanding is that it's better to have two separate oversights, and better to have two sets of eyes when things are uncertain. Good reviewer doesn't always translate to good content creator. Alex ShihTalk 13:25, 7 September 2017 (UTC)
  • I agree with Lex, reviewing somebody else's creation, and creating own and then reviewing it are very different things. Then we have some reviewers who are not good with all the policies. They might create an article which is out of their area of expertise, hence needing a review. Talking about myself, I am recently getting a lot tempted to publish two of my unfinished draft only because I am too lazy, and bored to work on them lol. (I think she is very attractive Tim, thanks for the introduction. )
    —usernamekiran(talk) 02:04, 8 September 2017 (UTC)
Uhh... You're welcome? TimothyJosephWood 02:07, 8 September 2017 (UTC)

Not sure what to do with this one

If someone could take another look at Bespoke Post, I would appreciate it.

It appears to have been deleted after an AfD, declined at AfC, accepted at AfC, and then deleted under G11. The new recreation seems to be the declined AfC submission. My instinct is to draftify it, because it isn't quite unambiguous, (after-all, someone thought it was enough to pass AfC) but a second opinion would be appreciated. menaechmi (talk) 16:52, 11 September 2017 (UTC)

@Menaechmi: Sent it to G4, again. The last AfD was clear about the advertorial content. I'm pretty sure this is getting SALT'ed, too. An SPI might be in order. Chris Troutman (talk) 16:58, 11 September 2017 (UTC)
...This is on my watchlist, and for the life of me I can't figure out why. TJWtalk 17:03, 11 September 2017 (UTC)
confused face icon Just curious...does the same editor create the article? Atsme📞📧 22:57, 11 September 2017 (UTC)
The final two creations were by the same editor. Two different editors created the first two versions in '14/'16. Primefac (talk) 12:25, 12 September 2017 (UTC)

A curiosity

Have a look at History of the United States by state & its source code. I've left some comments on the author's talk page but I'm not sure where to take it from here. It's certainly an interesting experiment at the very least. Cabayi (talk) 23:17, 10 September 2017 (UTC)

I think PROD is the best way to go, though it will invariably end up at AFD. I'm not sure this is a proper use of WP:LST, mostly due to the reasons you left on the creator's talk page. Way too big, way too messy. Even as an article in and of itself, I don't really see keeping it around, since it's more of a meta history. Primefac (talk) 00:34, 11 September 2017 (UTC)
Is all the information on this page copied from other pages? If not, I'd say we should keep it around at least as a draft so that material can be merged from it into other articles (i.e. draftify it and then leave messages on other relevant pages suggesting that there is material that can possibly be merged). If any material is merged though, this should remain as a redirect for attribution purposes per WP:MERGE (possibly targeted at History of the United States). I'd suggest AfD rather than a PROD to get more eyes on this, this is an odd one for sure. — InsertCleverPhraseHere (or here) 01:06, 11 September 2017 (UTC)
Insertcleverphrasehere, the material isn't copied, it's transcluded. Thus, there technically isn't any "content" on the page itself; just a bunch of transcluded sections from other articles. Primefac (talk) 01:13, 11 September 2017 (UTC)
Ignore what I said about merging then. Thanks for the reply. — InsertCleverPhraseHere (or here) 01:24, 11 September 2017 (UTC)
It's a neat technical trick but leaves the attribution lacking. If the material were copied or split from another article there would be a {{copied}} template pointing there for attribution. If it were a template, then it's got the kind of attribution yoiu'd expect for a template - none. This method leaves the article looking, on the face of it (and as demonstrated in the comments here), as if it were written by one author - and no attribution to the authors of the material. I don't see 50+ {{copied}} templates being viable. Cabayi (talk) 13:00, 12 September 2017 (UTC)
Cabayi, the attribution is on the page that the content is translated from. We don't have {{copied}} templates on every page where a template itself is transcluded, but we have attribution and contribution histories for those as well. As long as there is attribution (somewhere) I don't really see the issue. Primefac (talk) 13:22, 12 September 2017 (UTC)

Autoconfirmed article creation trial

On September 14, 2017, the Wikimedia Foundation (WMF) started a six-month trial to require users obtain confirmed or autoconfirmed status in order to directly create articles in mainspace. If you have reviewed pages since then, you may have noticed that no articles are being created by very new users. This is because the trial has already started. The WMF will collect data on this trial for the six-month period. Once the trial is complete, the WMF will consult with the community to determine if this restriction on article creation should continue.

This does not mean that new users cannot write articles. New users that are not confirmed or autoconfirmed may still create draft articles, and volunteers participating in Articles for Creation may approve or decline these drafts. However, users that are not confirmed or autoconfirmed cannot create articles without going through this process until they become confirmed or autoconfirmed.

What should I expect?

You should expect fewer number of articles passing through the New Pages Feed daily. Since it is expected that confirmed and autoconfirmed users will be more familiar with Wikipedia content policies, most articles will likely no longer have obvious problems. However, articles may still fail notability guidelines, so you should be careful to check whether each article covers a notable subject. Remember that even confirmed and autoconfirmed users can still be relatively new to Wikipedia, so you should be patient with these new users.

Why wasn't there a request for comment or other discussion on this topic?

The proposal for this trial was made in 2011. A few hundred editors participated. The proposal gained consensus, but the WMF at the time rejected the proposal. However, after renewed discussion, the WMF has agreed to perform this trial as a research experiment. After a few months of consultation with the community, the trial was launched with the WMF taking responsibility for collecting statistics.

How can I know more about this trial?

The full details of the trial are available at Wikipedia:Autoconfirmed article creation trial. Esquivalience (talk) 02:15, 15 September 2017 (UTC)

Just four points

  1. The community will discuss what happens after the trial. The volunteers will welcome any input from the Foundation but with the exception of a few purely legal issues the WMF is not empowered to make any ruling by fiat over the individual Wikipedias.
  2. Due to a 2012 issue involving a senior WMF staff and his junior assistant over a community NPP sub-project, this Wikipedia is conducting its own parallel set of stats and analysis. We highly appreciate the Foundation's efforts to provide professional auditing of the trial, but these safeguards are necessary.
  3. Advocacy: Although paid-for pages do not constitute the largest portion of unwanted new pages, one of the main goals of ACTRIAL is to see if it can dissuade paid editing. It's expected that these people who exploit our volunteer work for their own and their clients' gain and scamming the subjects of BLPs will eventually find workarounds. Reviewers are asked to be especially vgilant for all the hallmarks of commissioned work and report anything as quickly as possible to COIN and SPI and make liberal use of the Template:Uw-paid1 warnings.
  4. As the winners of the February coordinator election never took up their posts, I would like to personally thank TonyBallioni who has (unwittingly) become the de facto coord. Please give him all your support.

Kudpung กุดผึ้ง (talk) 05:00, 15 September 2017 (UTC)

ACTRIAL and the New pages feed

Tracked in Phabricator
Task T175225

Would it be possible to somehow tweak the filters for "Were created by new editors" to be more open. Would love it to show new pages by relatively new editors, even if they are now autoconfirmed. It used to show people with over a hundred edits. Now sure what the filter is set to now, but maybe have this include anyone not "extendedconfirmed" at this point.--☾Loriendrew☽ ☏(ring-ring) 19:37, 20 September 2017 (UTC)

See T175225. This is currently being worked on by the WMF. TonyBallioni (talk) 19:41, 20 September 2017 (UTC)
Can we please make sure that the terms used for describing the combination of age and number of edits that we conflate to "experience" is clear and unambiguous? What intuitively distinguishes 'Learners' from 'Newcomers'? It seems that the WMF thinks that we, "those not in the know", (sic) can't tell the difference between autoconfirmed and extended confirmed and they have decided to invent new descriptions that nobody knows the precise meaning of. Mduvekot (talk) 20:36, 20 September 2017 (UTC)
Mduvekot, it is done that way because they are integrating these filters with other filters that are being rolled out across multiple WMF wikis. The definition of autoconfirmed and extended confirmed change from wiki to wiki so they are trying to standardize the filters. TonyBallioni (talk) 02:34, 21 September 2017 (UTC)
@TonyBallioni: I know, and it's too late anyway. I should have let this one go. I know you're trying to do the best you can (it's much appreciated), and I suppose I could work with it even if the filters were called X and Y. I'm admittedly naive about the way the page curation tool is being developed, but given the response to T169244 and T169120 my skepticism about the effectiveness of the development of the filters shouldn't come as a surpise. I wish I had a better understanding of how features are designed by the WMF. I've read https://www.mediawiki.org/wiki/WMF_product_development_process and came away more confused than I was before. I have worked on very large, really complex (agile) software projects, so you'd think I ought to be able to understand it. Mduvekot (talk) 12:20, 21 September 2017 (UTC)

New editor creations backlog drive

I'm not a fan of backlog drives, as I've stated before: but I think with ACTRIAL we have a real chance to clear the backlog of unreviewed pages created by new users. There are currently over 700 dating back to June. Since there is a 90 day NOINDEX period, we're coming up on the end of it. I think it'd be good to undertake as a project trying to clear the review backlog of these pages as soon as possible. Also, my standard request that everyone check for copyright issues and not to go too fast. Thanks to all. Hopefully ACTRIAL will help us move forward towards a world where everyone has access to free quality knowledge. :) TonyBallioni (talk) 19:24, 15 September 2017 (UTC)

I'll second that. A quick look at that backlog seems to reveal that they are actually quite easy to patrol, but note TonyBallioni's caveats. Kudpung กุดผึ้ง (talk) 06:41, 16 September 2017 (UTC)
I'm up for that! It would be so satisfying to see that number go down. It was a bit like trying to empty a bathtub with a dripping tap using a teaspoon. Domdeparis (talk) 07:50, 16 September 2017 (UTC)
June has been cleared, attacking July!Domdeparis (talk) 12:21, 17 September 2017 (UTC)
Good work mate! I'd help out, but kinda busy with work at the moment... — InsertCleverPhraseHere (or here) 09:15, 18 September 2017 (UTC)
July is finished, I'm going to take a bit of a break now, good luck with August! Domdeparis (talk) 17:14, 20 September 2017 (UTC)
@TonyBallioni: Is there a link to the ACTRIAL related back log? If there is I can't seem to find it. Thanks in advance. — Preceding unsigned comment added by Steve Quinn (talk • contribs)

@Steve Quinn: just set the filters at Special:NewPagesFeed to be new editors. TonyBallioni (talk) 04:39, 22 September 2017 (UTC)

@TonyBallioni:. Sorry for pinging you again. Just want to say - I was doing this for awhile, but wasn't sure this was correct. Now I know it is. Thanks again. ----Steve Quinn (talk) 04:45, 22 September 2017 (UTC)

New New page patrollers

Kudpung's point about PERM causes me to wonder how we spot potential new New page patrollers. Encountering a few quality creations at Special:NewPages is a good prompt to examine their other contributions & their Article creation count, and suggest someone should apply for Autopatrolled. Likewise, quality requests at User:AnomieBOT/TPERTable for Template Editor. What's the equivalent for finding new patrollers and growing the corps? Cabayi (talk) 18:19, 23 September 2017 (UTC)

I don't know why people would ask for the right and not use it, perhaps people just really wanted the option - I was given the right without applying, and like many people, would never have considered applying, because 1, I didn't know it existed and 2, I've seen people put themselves forward for things like becoming an admin, and the process can be brutal so I would never put myself through something like that. I had done a lot of work on CAT:NN and created lots of articles; I'm sure there are lots of editors like that who would use the right wisely if they knew of it and were approached. Boleyn (talk) 19:42, 23 September 2017 (UTC)

Article review flowchart for the tutorial page

Hi guys (and gals). I'm not the first person to notice that the majority of new page patrolling is done by a very small group of people. The main issue I see is that many of the articles that end up stuck in the backlog are the articles that are "time-consuming judgement calls" that only experienced reviewers bother tackling. In order to help alleviate this issue, I've been working on a flowchart that is intended to facilitate any user with the basics of new page patrolling to review any new article to a degree of reasonable accuracy.

This workflow is based on the 'tutorial' NPP page, as well as the CSD criteria page, with a healthy dose of my own personal 'judgment' when it comes to patrolling new pages. I am not a good proofreader, so please point out any errors I have made so that I can correct them. This flowchart is very much a work-in-progress, so please point out any process that I've set up that you disagree with, or any oversights that I might have missed.

If you guys could try out the flowchart by using it a few times when reviewing some pages, that would help the most with ironing out the bugs.

Thanks. — InsertCleverPhraseHere (or here) 11:02, 6 September 2017 (UTC)

Insertcleverphrasehere, great idea. The points that hit me on a first reading...
  • G10 deletions - blank the article. It comes as standard with Twinkle but may need a reminder for those reviewers doing it manually.
  • Not English - Tag the author's talk page with a {{subst:contrib-XX1}} where XX is the language code. The templates available are listed at Template:Not English/doc#See also.
  • Remove Infringing text - and request WP:REVDEL using {{Copyvio-revdel}}
  • In the BLP section, where No leads to BLPPROD - Yes should include an instruction to tag the talk page with {{WikiProject Biography|living=yes}}
  • Draftify - Tag the talk page with the relevant projects - after all the objective is to have as many eyes on the page as possible to ensure its improvement rather than an underhanded deletion via G13.
For all deletion requests:
  • Check the page logs via the history - has it previously been deleted? Does the author's talk page show old deletion requests using a variation of the title?
  • Is this the 4th or later creation of the article under any title? If so tag the page with a request for {{salt}}.
  • Check the What Links Here - are the previous CSD/PROD/AfD notices on the talk pages of different users? Look into sockpuppetry.
  • Does the page contain an image? Is it hosted on commons? Follow the image to Commons & check the image's contributor.Is it the same editor? If not, look into sockpuppetry. If the article was an autobio, request deletion of the image on Commons - "self promotion" or "out of scope" are acceptable deletion reasons on Commons. Bear in mind the image may have a legitimate use on other wikis.
It would be great if the flowchart could be provided in a format which allowed the reviewer to Copy & paste the suggested tags rather than the tags being locked into an image. Just my 2¢. Cabayi (talk) 12:34, 6 September 2017 (UTC)
(edit conflict)Thanks for your comments Cabayi. I have included a number of your suggestions on the latest revision of the flowchart. A few I'm seeing were definitely needed (g10 blanking for example). Others such as looking into sockpuppetry and notifying users of Notenglish are great things to do, but not explicitly within the requirements of patrolling a page (and so probably don't belong on the flowchart but rather elsewhere on the tutorial page). In the final version, I will see if I can upload it as an SVG file so that text can be copied directly off of the image. I am also going to look into getting clickable links added to the graphic. — InsertCleverPhraseHere (or here) 13:09, 6 September 2017 (UTC)
  • It's just a very first draft. For a flow chart, it contains a lot of information which tends to make it look crowded. Perhaps present it in A4 Landscape view and check that it can be read at least on a 13" laptop or a large tablet without shifting the page around the screen. Adding a bit more depth of colour to the boxes and using traditional flow chart shapes to them still needs to be done along with making the shapes clickable links. All in all, an excellent idea that will serve both as a learning aid for new reviewers and as a quick reference for the more experienced ones. Kudpung กุดผึ้ง (talk) 12:50, 6 September 2017 (UTC)
As much as I wanted to use traditional flow chart shapes, and I have access to them, I found that they were unwieldy because of the amount of extra space circles and diamonds take up when you need to squeeze a bunch of text into them. While it is a flow chart, it is also an instruction manual, so a reasonable amount of detail is important I think, even if it does make it look a bit crowded. The 13" laptop shouldn't be a problem, as I am editing on one right now, not sure about the tablet though as I don't have one on hand. I decided to do the flowchart vertically so that the editor can keep it open in another tab and slowly scroll from top to bottom as they review the article. — InsertCleverPhraseHere (or here) 13:21, 6 September 2017 (UTC)
Insertcleverphrasehere I love it. I have been working on a flowchart of my own, and this one is much more developed than mine. One thing I noticed when reviewing Oseloka H. Obaze was that recreations of deleted articles, or copy/paste moves of drafts are not easily caught. (you already caught an earlier version, Henry Obaze Oseloka as a copyvio ). It's something that a reviewer should probably notice when they get to the "Is the article title correct" step, and find that the middle initial was added to work around a previously deleted title. Of course CSD:G4 does not apply, but I thought I'd mention that I saw no explicit step to check for recreations of deleted articles (per deletion discussion) in the current flowchart. Mduvekot (talk) 14:32, 6 September 2017 (UTC)
Minor issue: I noticed that after Remove infringing text and tag with revdel, you go back to Does the article have sufficient context to identify the subject. I think you can loop back to copyvio or continue on to Does the article have 2 or more references. Mduvekot (talk) 16:36, 6 September 2017 (UTC)
I'll change it to have it point down at "does the article have 2 or more references". I've considered "check the deletion log" as an additional step. However, as you have pointed out often recreations are at a different title, and aren't easy to find. Additionally, the main problem with checking the deletion log is that it doesn't give the non-admin any additional information aside from whether to {{salt}} the article. A user can't (or shouldn't) tag with G4 unless the current page is near-identical to the one that was deleted, and they can't know that because they can't see the deleted article. The exception is when they actually saw the other article, but in that case they know about it already, so there isn't much point in checking the deletion log. Even checking the deletion log so see if {{salt}} should be added is an unnecessary step, as this should get caught by the deleting admin anyway. I don't think it is especially useful or necessary to recommend checking the deletion log in this chart (our tutorial doesn't mention it either), though I am open to being convinced. — InsertCleverPhraseHere (or here) 19:16, 6 September 2017 (UTC)
  • I'd recommend simplifying this a lot. Wikipedia has a ton of policies that are in tension with one another, and a lot of the work of NPP is sorting out those tensions and triaging articles the best way possible. When I review I ask myself these questions:
    • Is this a copyvio?
    • Is this likely a commissioned work and does it show signs of socking?
    • Is this advertising? If so, is G11, PROD, or AfD better?
    • Does this qualify for any of the other CSD criteria?
    • Would this likely survive AfD? If so, are there obvious issues that need tagging?
    • If it wouldn't survive AfD, is PROD a better option because it'd be non-controversial?
    • Are there any other obvious issues that need tagging?
I split those out, but you could likely condense them to 5 or less. I'm sure I go through the entire flowchart in my mind, but I don't think of it that way. My concern here is that such a lengthy process will scare people off, when really it is just a series of basic questions if you are somewhat familiar with en.wiki policy. TonyBallioni (talk) 14:58, 6 September 2017 (UTC)
I hear you Tony, I do the same thing. However, the reason why you only need a list is because you've done so much reviewing before and are already so familiar with policies that apply to NPP. For example, your question 'would this survive AfD?' is based in a fundamental understanding of that process born through long experience. For someone who isn't, there just isn't a good way for that user to figure out what step is next when they get lost. As a direct result we end up with either A) incomplete reviews, or B) reviewers ignoring an article because they aren't sure. If you try out the flowchart you'll find that it is actually a lot simpler than it looks when actually using it, because typically you'll only be following through a third or so of the cells at most. This tool isn't meant for experienced reviewers (except perhaps as a reminder list) as experienced reviewers often only need to take one look at a page and they know what its issues are. This page is meant for those that don't know the reviewing process very well and I'm not sure how simplifying it could be done without the loss of vital information that is necessary for those very users to have. — InsertCleverPhraseHere (or here) 19:16, 6 September 2017 (UTC)
  • I feel like I'm flogging a dead horse but unilaterally draftifying an article has no basis in the deletion policy, and the ongoing RfC at VPR shows no community consensus for it. If a reviewer can't decide whether the article is notable (i.e. they think it's "borderline"), or don't have the information needed to make that call, the more conventional approach would be to leave it for another reviewer, or start an AfD and let the community decide. It's perverse that AfD, the project's oldest and de facto "default" deletion process, only appears on this chart as a backup for contested PRODs, which is explicitly reserved for uncontroversial deletions. – Joe (talk) 19:26, 6 September 2017 (UTC)
    • Joe, my reading of that RfC is the exact opposite of yours: the community has pretty clearly rejected the idea that unilateral draftification is not allowed by policy. Barring that, the community has also rejected exempting moves from WP:BOLD. Barring consensus otherwise, draftifying articles is simply a bold move. TonyBallioni (talk) 19:33, 6 September 2017 (UTC)
In any case, this flowchart only makes the case for draftification in a very narrow context, i.e. when the article has failed a notability check both on the sources included and in a google search and doesn't meet any automatic criteria and contains useful prose. AfD was added below a contested prod specifically to address two issues: 1) until you try a PROD, you won't know whether it is 'uncontrovertial', and 2) AfD is constantly clogged with articles that have a couple lines and that nobody cares about enough even to comment on, most were never prodded and many would likely have been uncontested. Considering that this flowchart already explicitly includes steps to check for additional sources and for automatic notability before PROD/AfD are invoked, I don't see much reason not to try for a PROD before entering a week-long process at AfD. — InsertCleverPhraseHere (or here) 19:37, 6 September 2017 (UTC)
@TonyBallioni: Well we'll have to wait for the close, but there are lots of support !votes in there, and some of the opposes also express their opposition to unilateral draftication, but object to some other aspect of SoWhy's proposal.
By a quick nose count, its 31-6 opposed to the suggested proposal (not counting guideline and the like). Yes, consensus is much more than a nose count, and I suspect the close will be nuanced and support the drafting of a guideline (which I don't think anyone is opposed to), but with roughly 83% opposition, there certainly isn't consensus for implementing the proposal at that RfC by any means (and I think SoWhy has admitted this elsewhere). TonyBallioni (talk) 19:59, 6 September 2017 (UTC)
I don't think that is a fair and just description of Draftify, Joe. It's got nothing to do with deletion or even soft deletion. New age Patrol is a triage, but that is probably a term mainly only understood by people with a military medic background or experience in large scale emergency services. Until I got sickened by the general lethargy of the community to anything about it other than complain about the research and suggestions made by others, I was largely responsible for getting the Draft namespace created in 2013 and advocating for a Draftify feature. Any truly competent reviewer will know how to use it with discretion, and anyone who doesn't, shouldn't be patrolling pages. I just wish people would stop thinking of NPP as one big article genocide - 50% of it is supposed to put new article creators on the right track so they can get help. Draftifying their articles is one way of doing it. The other alternative is simple deletion. And that doesn't encourage anyone. Kudpung กุดผึ้ง (talk) 07:57, 7 September 2017 (UTC)
@Insertcleverphrasehere: And yet that context is still wider than the one for AfD, which is supposed to be our primary deletion process.

until you try a PROD, you won't know whether it is 'uncontrovertial'

I think this is a pretty catastrophic misunderstanding of WP:PROD, which states that it "must only be used if no opposition to the deletion is expected". If the reviewer thinks that the article is potentially notable, then by definition it is potentially a controversial deletion. This is a collaborative project, why are we so keen to insist that even tricky judgements be foisted on a single reviewer? – Joe (talk) 19:52, 6 September 2017 (UTC)
Are you suggesting that the 'no' from "contains useful prose" be directed at AfD instead of at PROD? I can understand the reasoning behind this. I'll implement it and see how it looks. — InsertCleverPhraseHere (or here) 19:57, 6 September 2017 (UTC)
(edit conflict) I was suggesting both answers to useful prose? go there, but that would be something. Anyway, like I said, I know I'm flogging a dead horse. This new pseudo-deletion via draftspace is already in the instructions, so I suppose it doesn't make much difference if it's in the accompanying chart or not. I just thought I'd try to make a case to not further sideline consensus-based deletion in favour of something with no oversight at all. Otherwise great work: I think the chart is very clear and will be useful for new reviewers. – Joe (talk) 20:05, 6 September 2017 (UTC)

(edit conflict)I tried this out some today, I really like the direction it heads, but I found a few points of friction that might be worth addressing:

  • Disambiguation pages aren't covered. (Obvi not a huge issue, but it warrants being mentioned after my demo run)
  • Merges/translations/spinoffs without attribution are a huge problem, so just a quick check for proper attribution would be a nice touch.
  • Subject-specific notability guidelines should cover all major awards - so that step is redundant. (I also faintly object to the use of "automatic" over "subject-specific" when the only automatic notability criteria is the GNG)

menaechmi (talk) 20:01, 6 September 2017 (UTC)

Thanks for your input. Changed "automatic" to "subject-specific" and removed the award cell. I have also noticed that the flowchart doesn't work for dab pages or for some standalone lists. It is more intended for 'standalone topic articles' (there must be a better phrase for this), but I'll look into how to integrate dab pages and lists into the flowchart (or else just put a note in its eventual caption saying that it is intended for reviewing standalone topic articles, not lists or dab pages). Could you clarify what you mean by proper attribution? Not sure how to integrate this, but in WP:MERGE I notice it says to always leave a redirect when material is merged, specifically for this reason. Perhaps I should change the flowchart so that when merging material it always points at the 'redirect' cell instead of checking for 'plausible search term'. Would this solve this issue? — InsertCleverPhraseHere (or here) 20:34, 6 September 2017 (UTC)
Wow thanks! That was quicker than I ever expected. I figured it might just be a scope issue, and I think it's fine if the scope is just articles I just wanted to be sure to mention it because I wasn't sure. An example might be Army ranks and insignia of Chile, which was split off from Chilean Army, but nothing ever said so. So I made a dummy edit with the summary "(Content in a previous edit was split from Chilean Army, see that page's history for attribution)" as far as I am aware, all that the CC BY-SA-3.0 requires is a link for attribution (Wikipedia:Copyrights#Reusers' rights and obligations). There is also a template {{copied}}. I'm not sure how you would incorporate it either, but I've seen people get under fire after merging content without attribution. I think as long as they go to the WP:Merge page it'll be fine. Thanks again, menaechmi (talk) 23:14, 6 September 2017 (UTC)
I'll redirect the arrow from the merge cell to the redirect cell , and add a note to the merge cell saying to "leave a link to the merged article in edit summary". I should be able to squeeze that in. — InsertCleverPhraseHere (or here) 23:22, 6 September 2017 (UTC)
  • Changes Note to @TonyBallioni: and @Kudpung:. I made some simplifications to the lower part of the chart, I changed the 'categories', 'tags', 'title correct?', and 'wikiprojects' cells to be if-then checks, rather than yes-no with extra cells. All in all this cuts 4 cells out of the flowchart and very much simplifies the lower part of the flowchart (and makes the flowchart shorter in overall length). See the changelog at the file page for more info. I'll keep looking out for more ways to simplify without losing information. — InsertCleverPhraseHere (or here) 04:28, 7 September 2017 (UTC)
I believe that the flowchart is ready to be added to the tutorial page. While it is sill subject to change from suggested edits, I think it is polished enough that it can be added as is (new input may come after it is added over there). I tried working out how to do this as an SVG file, but decided against it for two reasons. The output file from the program I am using gives an error message Upload failed: This SVG file contains an illegal namespace "http://www.w3.org/1999/xhtml". (uploadscriptednamespace, http://www.w3.org/1999/xhtml). But also, the SVG fails to render properly with some web browsers (works in google chrome but not internet explorer). In future, converting to a clickable graphic is a good idea, but it is not immediately essential, and I think we should wait a while to see if other changes are suggested/needed. — InsertCleverPhraseHere (or here) 23:01, 8 September 2017 (UTC)
Question Why is this flowchart so much more complicated than AFC's Workflow Chart which describes basically the same process? Roger (Dodger67) (talk) 17:07, 22 September 2017 (UTC)
Because AfC doesn't have as many issues to deal with from submissions. In particular, many CSD criteria don't apply to AfC submissions. But most of all it has to do with the fact that NPP has to be reactive rather than proactive. In the case of NPP, we have to act in the case of an improper submission, and the action has to be valid, with AfC the opposite is true, action is only taken when the submission is a good one, or else to give feedback (analogous to tagging for NPP). Improper submissions do not need to be rigorously evaluated for CSD criteria etc, because they are clearly not ready for main space. With AfC, except in rare cases such as copyvio or attack pages, you can generally just look at notability (only one section of this flow chart), and the responsibility is on the submitter to demonstrate notability. For NPP the responsibility is on the reviewer to demonstrate a lack of notability, which is more difficult and easier to get wrong. — InsertCleverPhraseHere (or here) 23:42, 22 September 2017 (UTC)
And that, Insertcleverphrasehere, is the most clear, concise, and accurate comparison that has ever been made. Thank you for it. Kudpung กุดผึ้ง (talk) 00:47, 23 September 2017 (UTC)
Indeed, an excellent explanation, Insertcleverphrasehere thanks. Roger (Dodger67) (talk) 05:13, 23 September 2017 (UTC)
Great idea to produce this, just a few comments:
  • The previous version on the tutorial page was much simpler and therefore less intimidating. Could a higher level overview flowchart be retained alongside this detailed process one?
  • All the Amber and Red boxes ought to link onwards to the Green "Reviewed" one to match how curation tools work
  • I'm not convinced by the Draftify flow for two reasons. 1. It is possible to get there with an unreferenced article but not have marked it as unreferenced. 2. I think BLPs where the sources are non-English would end up falling into this 'trap', which I don't think is intended. Lineslarge (talk) 14:02, 24 September 2017 (UTC)
Retrieved from "https://mdwiki.org/wiki/Wikipedia_talk:New_pages_patrol/Reviewers/Archive_14"