Wikipedia talk:New pages patrol/Reviewers/Archive 5

Archive 1

←

Archive 3

Archive 4

Archive 5

Archive 6

Archive 7

→

Archive 10

More death by numbers

I took a (manual) look at the page curation log for the fortnight from 30 March to 12 April. In that time, the backlog increased by probably close to another thousand. Numbers will be roughly rough (doesn't count things in the patrol or reviewer log but not the page curation log, doesn't count autopatrols, counts each thing once, may have considered a few articles with unicode titles as duplicative). 5235 articles were curated by 102 reviewers, and distribution was"

• 1 with more than a thousand articles reviewed
• 9 with 100-999 articles reviewed
• 36 with 10-99 articles reviewed
• 46 with 2-9 articles reviewed
• 10 with 1 article reviewed

And by day:

• 30 Mar - 348
• 31 Mar - 279
• 1 Apr - 577
• 2 Apr - 238
• 3 Apr - 297
• 4 Apr - 433
• 5 Apr - 263
• 6 Apr - 340
• 7 Apr - 356
• 8 Apr - 263
• 9 Apr - 379
• 10 Apr - 488
• 11 Apr - 399
• 12 Apr - 575

I still believe NPP is attacking much of the problem at the wrong end -- it's like trying to sweep the ocean off the beach. I don't know if full ACTRIAL would be needed, but at least changing from defaulting from creating in mainspace to using the article creation wizard (option in Preferences) should help improve quality and reduce the friction in performing curation -- I know there are some who enjoy adding cats, wikiprojects and tags, but... ^~Hydronium^~Hydroxide^~(Talk)~ 12:09, 13 April 2017 (UTC)

Absolutely.--Ymblanter (talk) 18:25, 13 April 2017 (UTC)

I think most people who have this talk page on their watchlist would probably support full ACTRIAL, but implementing that is a minefield and needs to be done at the right time and in a manner that would have community support. I think the default to article creation wizard would be a step in the right direction. TonyBallioni (talk) 18:34, 13 April 2017 (UTC)

I've noticed that Hydronium Hydroxide's number of reviews per week and mine (+/- 2975 on average in April so far) differ significantly from the numbers reported by Special:NewPagesFeed, where as I post this, its at 1810. Anyway, this is not sustainable. Something needs to happen, First, I think, we need a better way of creating our stats. Mduvekot (talk) 20:17, 13 April 2017 (UTC)

Kudpung asked Cryptic to work on the stats for through Feb a few months ago and they are here. Yes, a bit dated now, but I think it was using a query rather than manual count. TonyBallioni (talk) 20:28, 13 April 2017 (UTC)

• The Wikipedia (and WMF) tendency to base everything on stats has never been really entirely constructive and that's why I have always based my arguments mainly on empirical findings drawn on many years of close monitoring of the system. It goes without saying that something is clearly wrong with NPP but it's difficult to reach any conclusions without people getting defensive.

Unfortunately, for NPP it does seem as if only some compelling stats will provide the required knee-jerk for some determined engagement at New Page Review. Cryptic ventured to help with some stats and so have others, but the problems, as several users have recognised, is in finding sources to be able to provide some reliable stats. One of the reasons for creating the New Page Reviewer group - along with hoping that it would raise the standards of patrolling, was to encourage a move to Page Curation, not, as the detractors claimed (source: see the RfCs), to deprive users the use of Twinkle. Another reason was to be able to centralise the source of stats for patrolling, which given the way that stats are maintained by Twinkle scripts, has unfortunately proven to be almost impossible.

Unpatrolled pages on 15 April 2017

The latest picture of unpatrolled new pages shows an alarming increase which is still waiting for explanation and for which still not even a theory has been proposed. This kind of backlog however, will certainly force the issue of implementing ACTRIAL one way or another. Ironically however, some stats are probably going to be needed to reinforce an argument which although IMO, is a no brainer, editors new to Wikipedia since the massive ACTRIAL debate will need some convincing unless they are prepared to read up on the history of it. Let's not forget that even the worst unpatrolled and totally unacceptable new pages will be automatically released for indexing by Google after 90 days. Kudpung กุดผึ้ง (talk) 08:52, 15 April 2017 (UTC)

Hi Kudpung. I suspect it's a combination of straws that broke the camel, but speculation as to the reason or reasons for the jump is a bit of a stab in the dark. NPP activity are flows, and it would be useful to have 2-3 years of relatively granular data that can suggest answers to such questions as:

• Is output steady but new page creations have jumped, or the reverse?
• Is there a correlation between the newsletters/reminders/prompts you've sent out, and reviewer activity?
• Is there a decline in active reviewer numbers?
• Is there a decline in top reviewer activity?
• How much impact do occasional reviewers have?
• ...?

Data needed would be:

1. Weekly aggregates to see the flow pattern -- how many articles were:

• created in en-wiki?
• deleted from en-wiki?
• added to the NPP queue? (whether through direct creation or via move from draft)
• marked as patrolled in total?
• marked as patrolled from the NPP queue?
• marked as patrolled due to autopatrol?

2. Breakdown of number of articles reviewed by each anonymised reviewer. (CSV of numbers by week --2017-01-02:1271,820,521,...1,1,1-- would suffice. Names aren't necessary, since since any individual reviewer's pattern may be lumpy, and reviewers will be gained and lost during the period.)

^~Hydronium^~Hydroxide^~(Talk)~ 00:41, 16 April 2017 (UTC)

I'm not involved in this anymore, Hydronium Hydroxide, in fact I'm less than enthralled by the community reaction to me having shaken myself free of six years of voluntary shackles and organising an election to establish some semi-formal kind of process management to continue where I left off. I have also suddenly found that I have 20 hours a week more to spend with my grandchildren. I made the post above really just as a reminder that since I left, by some strange and rather awkward coincidence, things have got dramatically worse. But I'm certainly not laying any blame at anyone's feet. Kudpung กุดผึ้ง (talk) 15:38, 16 April 2017 (UTC)

• You'll never hear me say I'm not one to occasionally jerk off to a spreadsheet, but we don't need to fire up SPSS to understand what the problem is here: there are more people creating new articles and doing so faster than there are people reviewing them. Either we get more warm bodies contributing, or we don't. We really don't need a peer reviewed study on the issue. We need a practical solution to improve participation. TimothyJosephWood 16:09, 16 April 2017 (UTC)

• I agree that we need to recruit. People are reluctant to put themselves forward for rights on Wikipedia, partly because that often ends up with their voluntary contributions being ripped apart before a decision is made. I'm not sure how I was given the rights, I guess as a long-term, active contributor, I'd met the criteria. If it hadn't automatically been given to me, followed up by regular reports, I would never have got involved. I think we need to automatically give the rights to more long-term editors. Yes, they may find it a different skill they need to develop, but no one can develop it without practice. If we keep the numbers so small, there really is no point in complaining that the editors with the rights aren't doing enough - they're volunteers who may be working on other areas for a while or busy in real life - or put off by the doom and gloom surrounding the project. Boleyn (talk) 17:13, 16 April 2017 (UTC)

• • Another problem is that it takes time to do it right and you still can get whined at in a "damned if you do, damned if you don't" way -- even if you take (for example) ten minutes per article, you can still miss something, if you are too hasty and approve something that someone else dislikes, they don't hesitate to tell you that you are incompetent, and yet if you prod tag something pretty worthless, the creator is usually at your userpage with a wall of text complaining about how evil you are. It's under the "low level annoying" level for me, but it's discouraging. 15,000 articles times 10 minutes = 150,000 minutes or 2500 hours or over 104 user days to clear the backlog. Sure 200 users each in a 12-hour marathon might be able to do it... but? If one says you need 30 min per article to do it right, well, we shall never catch up ... but if one says you only need 30 seconds, I beg to differ... Montanabw^(talk) 02:48, 22 April 2017 (UTC)

Getting a longer backlog again

(Discussion moved from WT:NPP)

Now we have 15500 and it is steadily growing for the whole month. I think we never had it below 12000.--Ymblanter (talk) 10:32, 29 January 2017 (UTC)

@Ymblanter: is it that there's not been enough qualified candidates applying for the tool? O Fortuna!^{...Imperatrix mundi.} 10:43, 29 January 2017 (UTC)

I think we have a steady stream of candidates, but apparently they work all together slowlier than the backlog grows. I am sure Kudpung has some insight.--Ymblanter (talk) 10:44, 29 January 2017 (UTC)

(FYI DGG, WereSpielChequers, The Blade of the Northern Lights, Jbhunley, Ymblanter, Fortuna Imperatrix Mundi, Ottawahitech, Fuhghettaboutit,Robert McClenon,BU Rob13, I dream of horses, Noyster, VQuakr, Σ, Esquivalience, Kvng, Cryptic, MrX).
The backlog has been steadily rising since its unexpected sudden growth in mid 2016. It was that extraordinary increase that led me to pioneer (following several supportive talks over the years) the creation of a user group for qualified patrollers, thinking that it would not only increase the interest in doing the task, but also introduce some quality into it. Although consensus was achieved at every stage of the process, it was not without some loud, and sometimes unpleasant opposition not only from users who will vote on any RfC for anything without understanding, but also from other admins, who apparently perceived it as some kind of Twinkle vs. Page Curation power struggle. So against all odds and resistance from the WMF to provide and/or maintain the tools we need, nevertheless over 300 requests for the right have been accorded, meaning that if each reviewer had made about 50 patrolls, the backlog would easily by now have been down to a day's intake of around 800 non-autopatrolled articles - of which around 80% are all likely to be deleted by some deletion process or another.

Despite newsletters urging for participation however, it appears that most of the requests for the user right were simply from hat collectors - or so it would certainly seem, because many of them hadn't edited Wikipedia for months or even years, while others have had the right removed again almost as quickly as they obtained it. I am beginning to believe that New Page Reviewing is struck with the same apathy as all maintenance areas: a total lack of interest except from very young and/or inexperienced users. See also Signpost, Jan 2017 (administrators).

Mid July to 01 Oct 2016

We were able to get the long bugged 'NO_INDEX' repaired for new pages under the condition that they become automatically indexed by Google after 90 days which at 15,000+ is happening to many articles that are not reviewed in time, what ever pitiful or disallowed state they my be in. The only solution now, as I see it, is to implement plan B, WP:ACTRIAL, for which an overwhelming consensus still exists - in fact apparently even stronger than before; or to extend the New Page Reviewer right to vet and 'promote' all wannabe patrollers (a move that was vehemently opposed by the community who still insist that the task of patrolling be largely left open to newbies, young, and/or inexperienced users under the premise: Wikipedia is the encyclopedia in which anyone can tinker with the back office.

The WMF has dismissed our concerns again ostensibly in favour of using the gross number of new pages apparently as a marketing point for obtaining donations. They even discretely removed the talks from the programme at Wikimania 2016 that had been reserved to facilitate a discussion on this critical issue of new-page control. Following our concerns that were completely ignored by Tretikov and her predecessor Gardner, Ryan Kaldari (WMF) and MusikAnimal (WMF) very kindly carried out some minor updating of the Page Curation tools under some pressure from us, only to be told by new CEO Maher that it is not to be considered a priority issue, so the WMF 2011 plans to develop a proper landing page for new, new-article creators have once again been shunted this week into a siding (see Phabricator) after the community being told that these critical issue must take their place in the wish list queue (line) for cosmetic and convenience gadgets.

There is a move, apparently strongly supported by Opabinia regalis to have NPP done largely by WP:ORES, but having looked into this, and sitting through a 1-hour promotional video I was emailed this week by a WMF employee, I still fail to see how this particular task can be aided much by Artificial Intellgence where the actual deletion and/or retention of new articles without biting the newcomers, has to be done by qualified humans. Being an Internet based project, here is a typical rot set in that all problems in the world can be solved by computer software and automation. A project was started at Wikipedia:The future of NPP and AfC, but which also appears to have lapsed into lethargy.

However, due to some bad faith mischaracterisation of my goals to see the quality of Wikipedia content correctly monitored, I am now reluctant to pursue these issues and I am currently channeling my spare time into other MediaWiki projects unconnected with en.Wikipedia. Kudpung กุดผึ้ง (talk) 15:07, 29 January 2017 (UTC)

It's interesting what you say about hat-collecting- that was the spcific reason I didn't apply for it at the time. Also, are you basically saying that- for 'advertising' purposes- it suits the WMF to have new pages- but not necessarilly good ones; quantity over quality? O Fortuna!^{...Imperatrix mundi.} 15:32, 29 January 2017 (UTC)

@Fortuna Imperatrix Mundi: think a native speaker who can spell properly would have devined that much, yes. Kudpung กุดผึ้ง (talk) 15:37, 29 January 2017 (UTC)

My aspersion cast at the WMF may sound 'naughty' but it's the only logical explanation I can come up with, especially as the growth (which is nevertheless still growth) has been in steady decline since the watershed year of 2007. My opinion on that is that all the traditional encyclopedic articles have now mostly all been written. What we now have is the expansion of the Wikipedia as an almanac, an advertising platform, a soapbox for politicians, a graffiti wall for childish vandals, and a work of biographies of barely notable people. However, only those who regularly patrol new pages will be able to draw those conclusions - it's not something that bots or AI can deduce. Kudpung กุดผึ้ง (talk) 15:53, 29 January 2017 (UTC)

No, the naughtiness referred to was in the WMF's actions, not your analyis. But I'm too busy learning to spell to exercise precision in edit-summaries  :) O Fortuna!^{...Imperatrix mundi.}

My English grammar and orthography, like yours, are impeccable. Like you, I am a notoriously bad typist (comes from 60 years of 60wpm touch-typing) - typos are perfectly excusable. Kudpung กุดผึ้ง (talk) 16:20, 29 January 2017 (UTC)

BTW, we have to AGF on requests for special rights - we have never declined a request purely on suspicion of hat collecting, however obvious it may be. Kudpung กุดผึ้ง (talk) 16:36, 29 January 2017 (UTC)

Very kind, Kudpung. That's a pretty scary graph there; is there any specific reason for the unreviewd pages literally doubling (if I've read it correctly) in three months last year? O Fortuna!^{...Imperatrix mundi.} 17:20, 29 January 2017 (UTC)

You read it correctly, Fortuna. I made it. I think perhaps one or two highly experienced users just got fed up of doing all the work. There was one patroller who was patrolling hundreds of pages per sitting, but they were topic banned for poor quality reviewing. I don't know it that was in the same time window though.Kudpung กุดผึ้ง (talk) 17:28, 29 January 2017 (UTC)

• The big problem for the NPP is that most new articles are of low quality and so the work of reviewing them is boring. Me, I prefer to work on interesting topics and there are still plenty of those – far more than I have time for. In recent weeks, I have started articles about:

1. the first movie director to use special effects
2. one of the oldest bridges in England
3. a hero of the Holocaust
4. the story that started the Golden Age of science fiction
5. the first female professor of Greek in Britain
6. an iron age fort
7. a biological heritage site
8. a natural wonder
9. a fundamental concept of education and training
10. an outstanding palaeontologist whose collection fills a museum

But, because my work is autopatrolled, the NPP aren't so involved. Their focus is rather negative and that's not much fun. Andrew D. (talk) 16:53, 29 January 2017 (UTC)

Andrew Davidson, I don't see how it helps to tell us what you are currently working on. Do you have a solution for the backlog perhaps? Kudpung กุดผึ้ง (talk) 17:26, 29 January 2017 (UTC)

My point was to address Kudpung's comment that "all the traditional encyclopedic articles have now mostly all been written. ... only those who regularly patrol new pages will be able to draw those conclusions". An effect of auto-patrolled status is to distort what the NPP sees. What might help in putting the backlog into context would be some overall statistics for 2016. About 300,000 articles were created in 2016. It might help to know what proportion of these were auto-patrolled, deleted, tagged, improved, &c. We could then see what was happening overall to the new content and focus our efforts accordingly. Is there a breakdown like that? Andrew D. (talk) 17:59, 29 January 2017 (UTC)

Andrew we're not discussing autopatrolled status here. Look at the graph I made - it's clear enough. If you want other stats, make them. But please don't use strawman arguments or lay red herrings - they don't move us forward. Kudpung กุดผึ้ง (talk) 19:00, 29 January 2017 (UTC)

I agree wholeheartedly with Kudpung's comment: "What we now have is the expansion of the Wikipedia as an almanac, an advertising platform, a soapbox for politicians, a graffiti wall for childish vandals, and a work of biographies of barely notable people." - I'm now going through a bunch of articles about volleyball players who are, by the notability criteria for sportspeople, presumed to be notable due to having played in a national team. They almost certainly fail the GNG, though, and each one of those articles is simply referenced by a link to a team roster. No big deal: they're well enough formatted, and I can review all of them pretty quickly on the assumption that any attempt to mass-redirect them to their respective teams would fail. I still wish that we had much stricter notability guidelines, because an encyclopedia that has an article about every olympic medallist ever and every player in every national team ever, and every song ever to make the top 40 charts is painful to keep clean. --Slashme (talk) 20:06, 29 January 2017 (UTC)

To be honest, I agree with Andrew Davidson that there are many topics still not covered which are absolutely notable by any reasonable standards, and we are unlikely to run out of them soon. However this is indeed not a reason for our backlog.--Ymblanter (talk) 09:26, 30 January 2017 (UTC)

@Slashme: FYI, all these volleyball player biographies are coming from a mass-recreation of articles by a single user; see the section I started below this one. – Joe (talk) 14:25, 30 January 2017 (UTC)

• Anyone who has regularly patrolled new pages since 2007, will be left in no doubt whatsoever that the English language Wikipedia has now slowly but surely mutated into what is largely an almanac, a graffiti wall for children , a platform for politicians and social soapboxers, and a source of income for spammers. The vast majority of serious encyclopedic content has been written. Davidson's contributions, while of course meritorious, are hardly representative of the high pressure flow of junk that is 90% of the daily junk from new users, also largely from a region that is not even the traditional English speaking world. I do not recall Davidson as ever having been helpful at NPP. In fact, anyone who suggests that the profile of the average new article is depressing, is correct in their assumption. The only realistic way of addressing the root cause of the backlog, now over 16,000 is to roll out WP:ACTRIAL. 22:20, 30 January 2017 (UTC)Kudpung กุดผึ้ง (talk)

• Kudpung says the only way to address the backlog is with WP:ACTRIAL. I suggest that leaving unreviewed article with NO_INDEX indefinitely is an alternative. Has that been looked at? There are probably other ideas. We're not going to come up with the manpower to do the reviews. Let's do some brainstorming before going back to hit our heads on the WP:ACTRIAL wall. ~Kvng (talk) 14:58, 22 April 2017 (UTC)

OK so why does this occur?

Why do ridiculously old articles suddenly reappear in the NPP feed? Today I just checked off Marine Le Pen, which has existed since 2004. What gives? Montanabw^(talk) 02:36, 22 April 2017 (UTC)

This edit [1]. Redirects that are converted into articles show up in the feed. TonyBallioni (talk) 02:37, 22 April 2017 (UTC)

What TonyBalloni said. Any time a redirect is 1)nominated for deletion, 2)converted into an article, 3) has substantial text added to it, or 4) an article is converted to a redirect via vandalism and then reverted back to an article - it shows up on the New Pages feed. For redirects converted to articles this is a good thing, as the article may or may not be appropriate and should be evaluated like any other new page, the only difference being that if the article is inappropriate, instead of deleting it you should just revert back to the redirect (unless it's a copyvio. Then you should ask for the offending revisions to be deleted.). For the rest, just marking it patrolled to get it off the feed and moving on is usually the best course of action. One thing to look for with those is cut and paste moves. Often there will be an article at title X1, and a redirect at X2. Someone will think X2 is a better name, and cut and paste since they can't move over the redirect. At that point you need to request a history merge if the new title really is better. If the old title is better, just revert it and warn the person. ~ ONUnicorn^{(Talk|Contribs)}problem solving 17:22, 24 April 2017 (UTC)

Request for Clarification About Moving to Draft

I am requesting a clarification about moving a new page from article space to main space. Bessbrook Landmine Attack was created, concerning an incident in The Troubles (e.g., political violence in Northern Ireland). It consisted only of an Aftermath section, containing no details as to the incident itself. On New Page Patrol, I moved it to Draft:Bessbrook Landmine Attack in draft space as being obviously incomplete, as an alternative to a PROD or AFD for not describing the actual attack. (Some editors create a very incomplete article in article space as a placeholder and then wait to finish it, possibly because they don't know how to use user space or draft space. I am not at this time asking about the merits of creating incomplete articles, only about moving them to draft space.) The author then, rather than expanding it in draft space, created the reasonably complete article in article space. So far, so good. The draft then duplicated the last paragraph of the article, and served no purpose, and needed deleting. I didn't see a speedy deletion criterion. A10 is only applicable to article space. None of the G criteria were applicable, so I tagged it for MFD, and it is now at Wikipedia:Miscellany for deletion/Draft:Bessbrook Landmine Attack. User:SmokeyJoe said at MFD that I had "improperly" done a "backdoor pseudo-deletion" by moving it to draft space, and that a history merge was needed. I am not arguing that a history merge is or is not needed, although in my opinion a view of the two histories shows that the history of the article is sufficient. I am asking whether moving an obviously incomplete page to draft space is considered an "improper backdoor pseudo-deletion". I had thought that the option of moving a page to draft space had been discussed and was thought to be a valid exercise of reviewer judgment, no more drastic than PROD or AFD. Moving to draft space is even shown in the flowchart. If the author of a page that is moved to draft space disagrees, they can move it back to article space, just as the author of a page that is PROD'd can remove the PROD tag. In either case, the reviewer should consider any arguments given by the author, and may go to the community by AFD. (We know that restoration of a PROD tag is not permitted and that moving a page a second time is move-warring and is disruptive.) So: Is moving a page that appears to be incomplete an "improper back-door pseudo-delete", or is it a valid alternative to requesting deletion? Robert McClenon (talk) 19:54, 22 April 2017 (UTC)

Thanks for bringing this up, I was actually just thinking of posting to ask for more guidance on when it's appropriate to move to draftspace, and, until the Page Curation version of this option is up and sends an auto-notice, what guidance should be provided to the entry's creator when making this move, as I'm guessing many new editors will not understand what's happened or how they're supposed to proceed. This may bear on your example Robert, because I'm not so surprised the editor just started a new mainspace page rather than pursued the draft, if they weren't already working with draftspace.

But for my two cents, this instance was a perfectly proper, even commendable move to draftspace, as much less bitey alternative to dealing with a page you could have proposed for deletion. I think it's just that once it was replaced by a better version in mainspace, the abandoned draft just happened to fall between the cracks of A10 on the one hand and on the other, AfC reviewing's options for rejecting a draft as a dup. The latter didn't come up just because the draft hadn't been submitted for review, since the editor made a new mainspace page instead. So MfD was the only option, but at all not because of an effort to delete the subject, just because the draft was no longer necessary. Innisfree987 (talk) 21:10, 22 April 2017 (UTC)

In instances like these, where it's a copy/paste(ish) recreation, and there's no substantial history at the Draft page, is that I just redirect back to the main article. Primefac (talk) 21:34, 22 April 2017 (UTC)

Okay. The same as if the draft had been accepted. Thank you. I will continue occasionally to move pages from article space to draft space, after consideration, but without thinking of it as a "backdoor pseudo-delete". Robert McClenon (talk) 00:52, 23 April 2017 (UTC)

Oh yes, good call! Thanks. Innisfree987 (talk) 21:45, 22 April 2017 (UTC)

Saves everyone the hassle of an MFD ;) Primefac (talk) 21:47, 22 April 2017 (UTC)

Music to my ears! Innisfree987 (talk) 02:16, 23 April 2017 (UTC)

• At WT:Drafts, with wide participation, it was widely and clearly agreed, albeit non unanimously, that unilaterally moving someone else's creation to draftspace is not ok. The options are: Get it deleted, get agreement from the author to move it, or use AfD or RM to move it. --SmokeyJoe (talk) 00:59, 23 April 2017 (UTC)

By chance do you have diffs so we can read the discussion? I looked through the archive but the only related discussion I turned up closed with no consensus on the matter. I definitely could be overlooking something though (I didn't read the entire contents of the archives, just checked by subsection title) and I'd be interested to know in more detail where the discussion stands. Innisfree987 (talk) 02:15, 23 April 2017 (UTC)

If there was consensus, and I don't see consensus, that moving a page from article space to draft space is not acceptable, then I will note, first, that that was earlier than the times that I have asked the same question here, and, second, if so, we have conflicting consensuses that need to be resolved at Village pump (policy). I am waiting for the record of that discussion. Robert McClenon (talk) 03:46, 23 April 2017 (UTC)

• Documented consensus is here: Wikipedia:Drafts#Incubation. One relevant diff

Relevant discussions here: Wikipedia_talk:Drafts/Archive_5#RFC:_Clarification_over_main-space_to_draft-space_moves

& here: Wikipedia_talk:Drafts/Archive_6#Clarification_over_main-space_to_draft-space_moves. --SmokeyJoe (talk) 04:29, 23 April 2017 (UTC)

The discussion in Archive 5 was closed as No Consensus, with a note that there were too many conditional answers, both "yes, but" and "no, unless", and is being cited both as evidence in favor of the right of reviewers to draftify articles and against the right of reviewers to draftify articles. The Archive 6 "clarification" is no clarification at all; it is a statement of the original ambiguous closure and a unilateral comment about the closure. So we still, in my opinion, have no consensus. Robert McClenon (talk) 23:56, 24 April 2017 (UTC)

So this still leaves the question of whether a reviewer may move a page from article space to draft space. The possible answers are "Yes, subject to reviewer judgment", "Yes, subject to specific guidelines (to be provided)", or "No". I don't see how a "No, unless" or "No, but" is an answer. In the meantime, should I continue to use judgment and move obviously incomplete pages to draft space, or should I PROD them? Robert McClenon (talk) 23:56, 24 April 2017 (UTC)

The discussions were not really focused on new pages by driveby editors. The point was that draftification by any editor of any page is not generally ok. --SmokeyJoe (talk) 00:39, 25 April 2017 (UTC)

If anyone, including any reviewer, could draftify any new creation unilaterally, that gives them great power to impose their own preferences on what sort of new articles are accepted from new contributors. The power to enforce meta:Immediatism, for example. Give the newcomer a choice between draftifying and a deletion process, sure, but unilateral draftification is unilateral pseudo-deletion. If the author is not present, a drive-by contributor for example, there is no point draftifying, either it is a suitable topic or not. --SmokeyJoe (talk) 04:29, 23 April 2017 (UTC)

Duh. I may be in a minority of reviewers, but I don't think that preserving the rights of drive-by contributors is worth doing. Drive-by contributors are not real contributors; the fact that they are better than drive-by vandals doesn't say much. An author can always object to draftification by moving the page back into article space, in which case the reviewer can begin accept the article or nominate the article for deletion. It isn't clear to me how unilateral draftification is worse than PROD. PROD is either unilateral deletion (not pseudo-deletion) after one week or a contested action. Move to draft is only unilateral pseudo-deletion if no one objects. I don't see the reasoning to objecting to draftifying. Robert McClenon (talk) 23:04, 24 April 2017 (UTC)

Unilateral draftification, without limits, could be too easily abused. My reaction in the recent MfD may have been overstated. I suggest agreeing to and documenting cases that are suitable for immediate draftification. Ideas:

• The draftifier is a NPP reviewer;
• The page is a recent creation;
• All major authors are new editors;
• The topic is not a new topic (e.g. Current affairs)
• The page does not meet WP:STUB
• No one has objected to draftification.

possibly, draftification could proceed via a PROD process. (I think that might generate more work than worth).

In general, I strongly support notifying the page creator. I also think a note should be added to the talk page, not to the page itself AFC style.

--SmokeyJoe (talk) 00:39, 25 April 2017 (UTC)

Well, User:SmokeyJoe, I will take your comments as a backhanded apology for your backhanded insult in the MFD. It appears that you have some general objections to arbitrary or random draftification of pages, and no one has been saying that arbitrary or random draftification is okay, and that, based on a principle with many exceptions, you criticized my draftification in your comment in the MFD, and realize that what I did was all right after all. Okay. Maybe you don't disagree after all, except to say that it shouldn't be done arbitrarily or randomly, and no one said it should. That is also what the closure of the RFC concluded. Is that now what you are agreeing? Robert McClenon (talk) 00:47, 26 April 2017 (UTC)

Yes. And I apologise. Although I have seen you around a lot, I don't really know you, and on looking I see that you are an excellent and responsible reviewer. I would like to be helping you with your work, not hindering it. I am very happy to trust your judgement, but it would be better, in my opinion, to define the bounds of that sort of judgement, to constrain others whose judgement is less demonstrated. There were some tentative cases of dubious draftifications, which combined with wavering proposals for more routine deletion of everything old in draftspace, is a real, non-theoretical concern. I would like to see some documentation on when pages can be moved to draft. The documentation of acceptable moves from mainspace to draft should be written and and cross-referenced at Wikipedia:Drafts#Incubation & Wikipedia:New_pages_patrol#Drafts, or vice versa. Do you agree with my several dot points above as constraints on draftifications? Is much missing? Below, TonyBallioni is offering examples of things that should be draftified on discovery. In the WT:Draft discussions, I was not the most extreme anti-draftifier, but I think the concerns raised there need to be noted. --SmokeyJoe (talk) 03:46, 26 April 2017 (UTC)

I agree that guidelines for draftification are in order, and that it should be an option for New Page reviewers for new pages, rather than being available arbitrarily. I will comment more with 24 to 36 hours. If the author objects to draftification, they or someone else should move the page back to article space, and AFD may be an appropriate next step. As I see it, draftification should be for articles that are obviously incomplete, either because sections are missing, or the edit summary says that they are in progress, or for some other reason, if there is reason to think that the completed article may be worth keeping. Articles that are obviously crud and are not likely to be made into non-crud should be CSD'd if there is a reason, or PROD'd. Those are my comments for now. Robert McClenon (talk) 05:27, 26 April 2017 (UTC)

I would also suggest extremely rough translations or articles in with portions written in a language other than English to the point where inclusion in mainspace would lead to a PROD. Basically anything that is a notable topic but pages needing translation would be unlikely to save. TonyBallioni (talk) 00:43, 25 April 2017 (UTC)

19XX in Malaya articles

Ok so this new user Muhd Syamie (talk · contribs) has been just rapidly creating new articles basically of the form of 19XX in Malaya. The problem is that most of these are blank articles and just contain empty section headers. I redirected one to the main Malaya article but he seems to revert back on that too. Chrissymad asked him to stop but he won't. What should be done next? Yashovardhan (talk) 17:02, 26 April 2017 (UTC)

They seem to have temporarily stopped...I hope it's because they're reading their talk page... CHRISSYMAD ❯❯❯¯\_(ツ)_/¯ 17:08, 26 April 2017 (UTC)

While I usually AGF, that's a pretty huge list of empty articles. I'm going to A3 them unless there's content. Primefac (talk) 17:09, 26 April 2017 (UTC)

Ya, hopefully. I will probably monitor this user for some time if he starts again (or uses a sock). Yashovardhan (talk) 17:14, 26 April 2017 (UTC)

(edit conflict) @Yashovardhan Dhanania and Chrissymad:, I've deleted all but 3 pages. Those three actually have more than nothing (one only has one sentence) but I was more concerned with the truly empty ones. Feel free to CSD or AFD as appropriate. Primefac (talk) 17:15, 26 April 2017 (UTC)

Chrissymad, if you want to look at socks, might be worth taking a run at Haziq Zikri - they created a similar run of page creations, in the same method, but who knows if Syamie is just copying Zikri or if there's actually a connection. The couple Zikri articles I looked at seemed decent. Primefac (talk) 17:18, 26 April 2017 (UTC)

(edit conflict)@Primefac: Ya, I just had a look. I would prod them and see if the user objects. Will save 3 afd's if he doesn't. Yashovardhan (talk) 17:22, 26 April 2017 (UTC)

Primefac My mind already went there around noon :P CHRISSYMAD ❯❯❯¯\_(ツ)_/¯ 17:23, 26 April 2017 (UTC)

Proded them. Its gonna be a wait and watch. Yashovardhan (talk) 17:31, 26 April 2017 (UTC)

@Primefac: @Chrissymad: Ok, so he removed 2 prod tags and blanket one page. The blanket one goes to CSD but should we take the 2 left to afd or merge them to the history article? Yashovardhan (talk) 05:14, 27 April 2017 (UTC)

@Primefac: If you're still following, you might want to see this. Yashovardhan (talk) 18:10, 28 April 2017 (UTC)

An article that was redirected came back

See Quacker (Tom and Jerry) which was a stand alone article that was merged and redirected. Today an IP reverted that edit, so I did a rollback. The article is still listed on our project page but there is no way for me to check it off as reviewed. Is there a way I can fix that? ^{Atsme📞📧} 14:55, 30 April 2017 (UTC)

I just checked and It's marked as reviewed by you. Yashovardhan (talk) 15:40, 30 April 2017 (UTC)

Backlog 20,425

Still growing. Fast.

• In the 5 months fron July - Dec it more than tripled
• In November when NPR was introduced it stabilised and slowly began to drop
• What suddenly happened in mid Feb and caused a 50% increase in just 9 weeks?
• Certainly not an equivalent growth in new articles.
• What's it going to look like by 13 July this year?
• What, if anything, is being done about it?
• Is ACTRIAL the ultimate solution?

Kudpung กุดผึ้ง (talk) 19:03, 29 April 2017 (UTC)

I'm not sure what the solution is, but it seems that a higher rate of new articles are being created by new editors than even six months ago. The ultimate solution is to change our charter so that at least a minimal amount of experience is required before new users can create new articles.- MrX 19:24, 29 April 2017 (UTC)

I've attempted to delete several but the respective creators are quick to remove the tag. Far too few that should be deleted actually are, including those that are totally unsourced. The inclusionists expect other editors to provide RS for those articles instead of deleting them. It's pretty frustrating. ^{Atsme📞📧} 14:59, 30 April 2017 (UTC)

I've often argued for making new users go through AFC to create an article rather than just jumping in here. I guess there was even a proposal that reached consensus but was not accepted by the WMF. I'm not really sure about that though and my words should be taken with a big bag of salt (correct me if I was wrong here :). I'm still against allowing new users to create pages at least until they're auto confirmed. Yashovardhan (talk) 15:49, 30 April 2017 (UTC)

Kudpung, perhaps we should ping the WMF and advise them they better get busy? ^{Atsme📞📧} 17:26, 30 April 2017 (UTC)

Um... wasn't it discussed above that we don't actually need WMF's approval for an en-specific ACTRIAL? Primefac (talk) 17:30, 30 April 2017 (UTC)

ACTRAIL is discussed about every time there is a backlog update, and it's been pointed out here and in other locations that there are multiple technical ways to enact it on en-WP without WMF devs. There are several questions about the implementation though that are non-technical: what do non-AC article creators do (if it is AfC, do we have a functional AfC system in place to deal with the shift if backlog)? Does the community consensus still exist for ACTRAIL? If it is implemented technically, would WMF step in with an office action to prevent it? These all need to be thought about before we move forward on anything. TonyBallioni (talk) 17:38, 30 April 2017 (UTC)

What about the idea of switching the default in preferences from creating in mainspace to using the article creation wizard (which I assume would send everything through AfC)? Seems likely to be less controversial than ACTRIAL with the community and with WMF. Admittedly this will massively increase workload in the already-flawed AfC process, and secondarily at MfD, but on balance I still think it's better for the encyclopedia and new creators to work through issues outside of mainspace--much less visibility for unacceptable pages and for flawed but improvable pages, declining drafts is much less bitey than the deletion processes. Innisfree987 (talk) 00:05, 1 May 2017 (UTC)

WP:ACTRIAL, Innisfree987 and TonyBallioni, was precisely that, but the background needs to be studied very carefully - it was much more than just stopping non-confirmed editors from posting new articles in mainspace and sending them to the Wizard. To soften the blow for new users, for ACTRIAL a whole system of templates with built-in logic was devised (which did not need any creative coding by the WMF) but which also required some minor updates to the Article Wizard. To soften the blow for the inclusionists who wrongly interpret ...the encyclopedia anyone can edit, to mean 'the encyclopedia anyone can spam and post unadulterated trash into' , ACTRIAL as its name implies, was to be a 6-month trial just to see if it would work. It would certainly greatly reduce the creation of unwanted pages and would therefore not impact on the workload at AFC - in fact it might even help reduce that too.

ACTRIAL as we now know (and this was our error in asking the WMF at Bugzilla to pull the handles) can be implemented by any one of several local script/filter that various editors have proposed and there is every reason to believe that that particular consensus is just as valid today as it was then. It just needs someone (or more likely a small team) bold enough to insert the scripts, the templates, the Wikipedia GUI notices, and upgrade the Wizard. What Scottywong and I developed between us 6 years ago in preparation for ACTRIAL going live, was not rocket science.

Without a technical implementation (and I don't mean waiting years for ORES to be perfected), this is a critical issue for which there are no philosophical solutions. The WMF staff has had an almost 90% turnover since then, and it would be a very short sighted move to reject ACTRIAL today, particularly in the face of the huge increase in organised spam, professors organising Wikipedia education drives to pursue their own agendas, and the huge increase in the number of inappropriate new articles from non-native-English regions. At the risk of sounding radical, I doubt that the community would give two hoots about an office action, which IMO would not be technically enforceable without blocking and banning anyone who tries: Imagine a world where all the new page patrollers went on strike - please remain seated with your computers on until Wikipedia comes to a complete stop. And that seems to be pretty much what they are doing already, because between them, the 390 patrollers could easily have gotten the backlog down to zero, but those few who have been patrolling regularly now see their work in the face of an exponentially growing 20,000+ backlog as a useless, lost cause - well, at least I certainly do, in fact it's as much as I can do to simply cope with all the articles that have already been tagged for deletion without actually reviewing and passing any others as OK for keeping. It would be interesting to see what DGG has to say on these issues. Kudpung กุดผึ้ง (talk) 03:13, 1 May 2017 (UTC)

• What happened September 2016 - Feb 2017? I daresay it was User:Kudpung's flurry of activities that drew attention, and that the end corresponds to the completion of Wikipedia_talk:New_pages_patrol/Coordination#Coordinator_Voting_begins_now_until_23:59_UTC_Monday_06_March. Comparison with rearranging of deckchairs comes to mine. WP:ACTRIAL remains the only good answer I have seen, though I stop at agreeing with the word "ultimate". All processes require assumptions, testing, review, and improvements on the assumptions. It is a cycle, there is no ultimate. Both new page creation and AfC needs to require at least autoconfirmation. Newcomers with new topics should be advised to improve existing content before creating new orphan pages. An example of improvement to existing pages would be the addition of meaningful links to their new topic. If no mainspace page can be improved by creation of information linking to their new topic, then the chances of their new topic being a suitable topic for inclusion is very very slim. --SmokeyJoe (talk) 03:13, 1 May 2017 (UTC)

Apart from comparing the shape of the graph with the dates of various events such as the roll out of a new user right in November (which indeed was followed by a slight decline in the backlog), or Kudpung shaking himself free in February 2017 of his self-inflicted attempts (partly with the help of The Blade of the Northern Lights, WereSpielChequers, and Scottywong) over the years to get NPP improved, there are still no hypotheses being posited on the how and why these things came to shape the graph.

By tomorrow, the backlog will have reached almost 21,000. At this rate by the end of the month it will be nearly 30,000 and by the end of the year, nearly 100,000 most of which will have passed the 90-day NO_INDEX deadline and regardless of their unsuitability (spam, attack pages, vandalism, COPYVIO) be indexed by Google. Kudpung กุดผึ้ง (talk) 05:30, 1 May 2017 (UTC)

My hypothesis was that something stirred up the activity of the reviewers. Something dramatic happened, it's hard to precisely read the x-axis, around October 2016. I recall seeing you doing/commenting on things around then. It was probably this, advertised at WP:CENT on October 6. That activity stirred up other activities, such as this. The election for your replacement too. In February, is was all winding up, the discussions closed in early March. People then went back to their long-term behaviours. I guess this could be researched, but assuming largely true, my conclusion is a somewhat depressing: "The graph indicates that article creation exceeds the rate of review; advertising the need for reviewing can have a strong short term effect with no lasting effects". However, the last three data points are being relied upon very heavily. This is somewhat depressing, and it calls for something more serious to be done. Non-autoconfirmed newbies writing new orphan articles seems pretty obviously a flaw to me. Maybe there is no need to push this point. We can just wait until it becomes known that new articles are going live to google before being reviewed, and this will cause an exponential increase in spam dumping, and then we will more readily agree to delete all new orphans and turn off page creation by new editors. --SmokeyJoe (talk) 06:06, 1 May 2017 (UTC)

We would need some stats to prove that new page creations have increased so dramatically since last July. I don't think they have. What I do believe is that the WMF won't help because they rely on the gross number of page creations for boasting about growth - they put their hands over heir ears when we try to tell them that over 50% of these creatons get deleted sooner or later. And that's why they refused to allow ACTRIAL in 2011 and why they won't prioritise the development of a proper landing page for new users and why I assume (he won't tell me exactly why) Jorm was told back then to quietly shelve the Article Creation Flow project. Kudpung กุดผึ้ง (talk) 08:41, 1 May 2017 (UTC)

Surely they don't need to brag. Surely they understand the end of the phase of exponential growth? --SmokeyJoe (talk) 08:48, 1 May 2017 (UTC)

This says that new articles are indexed after 30 days, not 90. If this is correct (Cenarium added the information) it means that much of the backlog is already on google: Noyster (talk), 12:26, 1 May 2017 (UTC)

AFAIK, Noyster, the page at Wikipedia:Controlling search engine indexing may not reflect the actual situation or be up to date. At the time of the roll out of the New Page Reviewer Right, knowing that there was a huge backlog already, the time limit was set at 90 days. The IIRC, actual manipulation was carried out by Caldari - please check with him. Kudpung กุดผึ้ง (talk) 02:00, 3 May 2017 (UTC)

• The limiting factor is reviewers. The fundamental mind-set and skill for NPP and AfC review are very similar--the key differences are that NPP reviewers ar expected to tag problems as well as judging the basic acceptability with respect to copyvio, WP::V and WP:N, and that failure to pass NPP requires choosing and initiating the further deletion processes. There is relatively little to be gained shifting incoming articles to AfC if we do not have an adequate number of reviewers. ACTRIAL leading to AfC will just add more steps to the problem.

There is no way to deal with this that does not involve getting additional capable active reviewers and the only place to find them is by getting mroe of the experienced people involved.

Nor will ACTRIAL help dealing with undeclared paid editors--already they know to make a few edits first. The only thing that will help in the short run is more stringent decisions at deletion processes. The best way to achieve this is for people who care about the problem to participate more at AfD. It doesn't require elaborate analysis of referencing or arguing point by point with those who do not realize the harmful effects or mastering the details of argumentation based on the interpretation of the often ambiguous guidelines for notability . There are simply policy-based arguments--among those I have used are:

Borderline notability plus borderline promotionalism is justification for deletion.

Regardless of possible notability, promotional articles must not be allowed t stay in WP; the policy of NOT ADVOCACY can override questions of notability

Gross promotionalism should be deleted at G11--lesser promotionalism is cause for deletion at AfD if not immediately fixed during the discussion. The mere statement that it can possibly be fixed is as inadequate as saying copyvio can possibly be fixed. Userifying or draftification is acceptable for unclear notability, but promotionalism should be removed altogether.

The real reason for ACTRIAL is the greater opportunity for non-confrontional guidance of good faith beginners--sand this require real individualized focussed help at AfC--not just putting on a confusing template.

In short: there is no way to remove junk without doing work. There are no devices that will have any effect without significant skilled effort. DGG ( talk ) 09:17, 1 May 2017 (UTC) .

• ACTRIAL mustn't be allowed to send newcomers to AfC, it must apply equally to AfC. --SmokeyJoe (talk) 11:15, 1 May 2017 (UTC)

Can we organize ACTRIAL ourselves in the following way: We ensure that BLP violations, vandalism, and spam articles are deleted promptly (which I believe happens anyway) and all other unreviewed articles are moved to draft automatically after 29 days to avoid indexation?--Ymblanter (talk) 16:44, 1 May 2017 (UTC)

Moving unreviewed articles to draft automatically after 29 days would be far more discouraging for new editors than being forced to create in Draft and having their articles promoted to mainspace. It seems to mechanise a BITEy behaviour, and deprive the editor of the warm-and-fuzzy approbation of having their article approved. Just my 2¢. Cabayi (talk) 07:17, 2 May 2017 (UTC)

Agree with that. 'Promote not demote' will retain new editors. — O Fortuna ^{semper crescis, aut decrescis} 14:29, 2 May 2017 (UTC)

If you want to ensure that BLP violations, vandalism, and spam articles are deleted promptly (which I believe happens anyway), Ymblanter, you will either have to dramatically increase the number of suitably qualified and/or experienced New Page Reviewers from its current 390 to around ten times that (chance would be a fine thing), or encourage the ones we have to use the tools they asked for. Only articles that are tagged for deletion are dealt with promptly, and that is by admins. The purpose of ACTRIAL is to make potential creators of inappropriate content think twice before wasting their time. This was supposed to be coupled with the creation of a proper landing page for new users whose intention was to create a page. AS DGG stated above: In short: there is no way to remove junk without doing work. There are no devices that will have any effect without significant skilled effort.. Kudpung กุดผึ้ง (talk) 02:14, 3 May 2017 (UTC)

• 'Promote not demote' will retain new editors. New promotion spam editors amongst them. Increasingly.

Template:Welcome begins...

Hello, Example, and welcome to Wikipedia! Thank you for your contributions. I hope you like the place and decide to stay. Here are a few links to pages you might find helpful:

• Introduction and Getting started
• Contributing to Wikipedia
• The five pillars of Wikipedia
• How to edit a page and How to develop articles
• How to create your first article
• Simplified Manual of Style

Notice how "How to create your first article" follows four points of six links of logically preceding things? Where did the notion come from that people would come straight to Wikipedia and write an article before even becoming orientated? Note for example Wikipedia_talk:WikiProject_Articles_for_creation#Stop_posting_discussion_on_AFC_draft_pages.2C_use_the_discussion_page_instead where I have learned that regular reviewers and experienced admins alike think its normal to facilitate article writing by newcomers who haven't even worked out talk pages. --SmokeyJoe (talk) 03:47, 3 May 2017 (UTC)