Wikipedia talk:Autopatrolled/Archive 1

This is an archive of past discussions about Wikipedia:Autopatrolled. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Archive 2

Image

Is anyone willing to create an image for this? Kevin Rutherford (talk) 23:53, 26 June 2009 (UTC)

I'd suggest the Wikipedia puzzle ball with a check mark over it or something similar. Malinaccier (talk) 00:39, 27 June 2009 (UTC)

Removing autoreviewer

Have we any procedure for this, or is it at Admin's discretion when inappropriate pages are being created (with post on AN for review of action)? KillerChihuahua^?!?Advice 18:26, 13 September 2009 (UTC)

I was just about to remove autoreviewer from a user when I saw your note. In this case, I felt uncomfortable with the user having this right just yet because of some copyright problems in their history. I discussed my concern with the admin who had granted autoreviewer and they agreed. I think that is all the procedure we need -- a discussion and agreement with the granting admin is sufficient to remove the right. In case of blatant problems with creating inappropriate pages, I think revoking autoreviewer rights is at the discretion of the individual admin in the same way as rollbacker rights. — CactusWriter | ^needles 14:28, 14 September 2009 (UTC)

Inappropriate autoreviewers

MBisanz (talk · contribs) has given the autoreviewer right to many users who have created just a handful of articles, and to at least two users who have never created a single article.

I asked MBisanz about one of these, pointing out the correct criteria. The response was: "Well he has created some talk pages, I guess he just looked trustworthy."

I again mentioned the correct criteria, but subsequently, MBisanz granted the right to at least one other user who had never created an article. (The thank you note from the new autoreviewer makes it clear that this user does not even understand the right.)

Links:

JimCubb (current rights · rights management · rights log (local) · rights log (global/meta) · block log) - articles created
Islanders27 (current rights · rights management · rights log (local) · rights log (global/meta) · block log) - articles created
MBisanz's user rights log

Agolib (talk) 23:18, 4 November 2009 (UTC)

I apologize for the length of time it took my to comment on this but I have been somewhat busy in real life and doing what I do here.

I have not created an article and it is very unlikely that I will ever create an article. In addition to that disqualifying condition, the only edits that I make to articles are minor other than sort values.

As I understand the "documentation" for this, my being listed as an Autoreviewer means that when I add or fix a sort value on an article the article is marked in some way that it has been patrolled. That is not the case. I never make substantial edits to articles. I may make minor copy-edits, spaces after commas where none existed, but the last major change I made to an article was over a year ago when I rebuilt a sentence that had been partially rewritten.

I should not be an Autoreviewer and I apologize for the damage that may have been done by my erroneous designation as one. JimCubb (talk) 01:22, 28 December 2009 (UTC)

Thanks for your constructive and conscientious attitude. First, please be aware that you have done absolutely nothing wrong. No "damage" has been done. Only pages created by users with the autoreviewer right will be automatically marked as patrolled. As you have not created any articles, this has never happened. It does no harm that you have the right (as long you don't go "rogue" and start creating inappropriate pages). My note was not in any way meant as a criticism of you or the other user whom I mentioned. I was merely pointing out that the administrator granted the right to people who did not qualify for it, did not need it, and, as it turns out, did not even realize what it meant. Agolib (talk) 23:42, 2 January 2010 (UTC)

I did not take it as a criticism of me at all. However, since the "next" article I create will also be the first article I create and will be filled with errors, the permission should be removed. JimCubb (talk) 00:38, 3 January 2010 (UTC)

Removed per your request. Ϣere SpielChequers 08:13, 3 January 2010 (UTC)

Requirement question

Do the pages created have to be articles or can they be talk pages or files? Even if talk pages and files count I still don't think I'm qualified yet. I also don't know how to check how many pages total that I've created. Sign My Guestbook! User:Sumsum2010 20:34, 10 April 2010 (UTC)

Hi Sumsum. This is about articles, your total is, 18 of this date. Ϣere SpielChequers 10:01, 16 May 2010 (UTC)

Thanks! Might be a while=) Sign My Guestbook! User:Sumsum2010 18:48, 16 May 2010 (UTC)

Copyvio problems = no autoreview permission?

I recently nominated User:Cohee for the autoreview privilege. The following discussion ensued, with WereSpielChequers correctly pointing out that the discussion was better suited for this page. Here it is:

Extended content

60+ articles created, most recent on 11 May 2010. --Cryptic C62 · Talk 19:00, 13 May 2010 (UTC)

Not done User seems to have issues with uploading copyright violations. -FASTILY ^(TALK) 22:48, 13 May 2010 (UTC)

That has nothing to do with article creation. --Cryptic C62 · Talk 02:11, 14 May 2010 (UTC)

Yes it does. If an editor is consistently violating the copyright held in images, it's just not a good idea to remove their article submissions from reviewing eyes. ~ Amory (u • t • c) 12:28, 14 May 2010 (UTC)

Don't bots usually pick up on image copyright problems nowadays? If that's the case, then it doesn't matter if his articles go through NPP or not because the bots just run independently of NPP. --Cryptic C62 · Talk 16:09, 14 May 2010 (UTC)

How do we know if the articles are likely to be policy-compliant if we can't trust the user to comply with WP:C? Tim Song (talk) 18:29, 14 May 2010 (UTC)

Actually, Cryptic C62, it's people like me who painstakingly review each and image that is uploaded. -FASTILY ^(TALK) 22:21, 14 May 2010 (UTC)

If a person is manually reviewing images as they are uploaded, that's essentially no different than a bot doing it in the context of the autoreview privilege: In either case, the image upload will be checked by someone regardless of whether or not the relevant article goes through NPP. As for Tim, good article writers don't necessarily know anything about the myriad of rules that involve image copyrights. I'm a perfect example! I consider myself to be an experienced article creator, yet I don't have the slightest clue how to go about finding/uploading useable images other than those that I've made myself.

On a side note, I must point out that I really don't care if this particular editor receives the permission or not. I'd just like to get all of the arguments on the table to determine a consensus-based precedent since it's usually the same handful of editors who hang out here. I'm not convinced that there exists a relation between image uploading and article creation skills, though I am willing to abide by the consensus if one is evident. --Cryptic C62 · Talk 23:35, 15 May 2010 (UTC)

Autoreviewer bypasses new page patrol, but not the copyvio bots which do much of our detection of copyvio. So there is a practical argument for giving Autoreviewer status to people who've mastered other aspects of article creation but might still breach copyright; But I and I think consensus take the opposite view, that if we know someone is liable to breach one of the rules about article creation we want to be cautious about other articles they create. There is a wider issue here over how far back we go for warnings about copyvio, but I would prefer that more general discussion take place on Wikipedia talk:Autoreviewer, rather than in the context of an individual editor. Ϣere SpielChequers 10:09, 16 May 2010 (UTC)

Thoughts? --Cryptic C62 · Talk 14:29, 17 May 2010 (UTC)

I would have to agree with WereSpielChequers. If the user can't be trusted with one policy why trust them with another? Sign My Guestbook!·Sumsum2010·Talk 03:45, 8 June 2010 (UTC)

Proposal to rename Autoreviewer to Autopatroller

I proposed this here. Cenarium (talk) 16:58, 11 June 2010 (UTC)

Article creation only?

WP:RFP/A criteria appears to be for new article creation only. That's not stated in this article anywhere, though suggested by "A suggested requirement is 75 valid articles." Reading through WP:NPP, it appears that the majority of the concern is on new article (mainspace) creation. I'll go ahead and revise this article accordingly if no one corrects me. --Ronz (talk) 05:31, 22 December 2010 (UTC)

Done. --Ronz (talk) 19:51, 23 December 2010 (UTC)

Proposed changes

Whether it makes sense to require such a large number of previously created articles is being discussed at WP:VPP#Autopatrolled - reduce number of qualifying articles. You are invited to join the discussion. WhatamIdoing (talk) 18:57, 5 January 2011 (UTC)

Please note that the VPP discussion has agreed that 75 is far too high. Recommended levels of previously created articles ranged from 10 (yes, ten) to 25 articles. I have changed it to a suggested level of 50, and I expect it to decline further over time. WhatamIdoing (talk) 03:15, 11 January 2011 (UTC)

Bot to find article creators

This was also discussed in March 2010 at Wikipedia_talk:Database_reports/Archive_3#Candidates_for_Autoreview_privileges - but that seems to have been a false start.

I've appointed a few autoreviewers recently after seeing their unpatrolled articles at Special:Newpages. I suspect there are a lot of article creators out there who have never been considered for autoreviewer and are unaware of the process. What do people think of the idea of getting a bot to produce a prospect list of authors who are:

neither admins nor autoreviewers
have created over 75 articles other than redirects
have not been blocked in the last 12 months
are currently sufficiently active to have created an article in the last month?

I suspect a high proportion of such editors would qualify for autoreviewer, but if they never submit more than one article in the same day they could be editing for years before anyone considers them for autoreviewer. - I'd happily review some of these prospects if we had such a list. Ϣere SpielChequers 15:03, 6 December 2009 (UTC)

I think this sounds like a great idea (though I've no experience at all with autoreviewer granting). The only possible snag I can foresee is people stumbling across their name on its output and getting annoyed - "Your bot said I was good enough for autoreviewer; why didn't I get it?". I doubt it'd be a major problem though.

I'd also be very happy to write the bot if there's a consensus for it / no-one objects. Olaf Davis (talk) 22:28, 12 January 2010 (UTC)

Thanks, its the coding/running of this that I need, I'd be happy to assess the resulting list and appoint a bunch of autoreviewers. However like you I'm not sure what the norms are with producing such lists of usernames. wp:EDITS gives users the option to opt out, and I know of two editors who choose not to be autoreviewers - so a stop list might be in order. But I'm a little uncertain about the privacy aspects of doing this in bulk, this might well be the focus of the bot approval process. Ϣere SpielChequers 00:46, 14 January 2010 (UTC)

I actually proposed something very similar a little while ago. The result was this list, which I refer to as the magical list of magic. That's where I've been pulling my names from. It's not entirely perfect, so I've been working with User:MZMcBride to try to make it work perfectly. The relevant discussion is here. --Cryptic C62 · Talk 14:25, 17 May 2010 (UTC)

Interesting but the first one I looked at hadn't edited since June 2009. So at the least we need a fresh report or to add the proposed "created an article in the last month" criteria into the query. Ϣere SpielChequers 16:50, 25 September 2010 (UTC)

┌───────────────────────────┘
I decided to be bold and started in on this task. Several of us have gone through the old list of 1000 and granted permissions here. We are now working on a new list of 2000 and will probably be finished in a week or so. - Hydroxonium (H₃O⁺) 14:17, 30 January 2011 (UTC).

Autopatrolled traits

Following discussions on my talk page with another editor, it was suggested that I propose this here -- kind of a guide to the traits that should raise red flags when granting autopatrolled permission. I think the important point is to focus less on the number of articles and more on the quality of articles. For example, someone with 10 FAs but nothing else should get the permission, but anyone with 100 articles but problems listed below shouldn't. Ultimately what we're here to do is allow things through the gate that don't have any problems that NPPers are there to pick up. Thoughts? --Mkativerata (talk) 04:38, 30 January 2011 (UTC)


What NPPers look for (in order of seriousness)	Traits that should exclude an editor from the permission	Where an admin can look to see if an editor has those traits
Copyright violations	Any recent history of creating articles with copyright violations	Spot checking recent article creations (this tool can help), talk page copyright notices, and block log.
Unreferenced or poorly sourced BLPs, or attack pages	Any recent history of creating unreferenced or poorly sourced BLPs, or attack pages	Spot checking recent article creations, talk page notices, and block log.
Unreferenced or poorly sourced articles	Any recent history of creating unreferenced or poorly sourced articles	Spot checking recent article creations. Talk page notices.
Articles that fail inclusion guidelines	Any recent articles that have been deleted via the deletion process	Spot checking deleted contributions and user talk page deletion notices
Articles that promote an entity or product or that overly rely on sources no independent of the product or entity	Any recent history of creating such articles	Spot checking recent article creations
Articles that are substantially below acceptable wikification standards	Any recent history of creating such articles	Spot checking recent article creations to see if they have been tagged by NPPers

I would like to see the above list expanded and added to a new "reviewing" section WP:AUTOREV. New Admins have experience in different areas and what seems like common knowledge to one Admin might not be so common to another. So freely sharing this information can only be helpful, IMO. - Hydroxonium (H₃O⁺) 13:58, 30 January 2011 (UTC)

Here is a link to the Copyright Violation Detector tool at the toolserver. This could be added to the table above. - Hydroxonium (H₃O⁺) 14:49, 30 January 2011 (UTC)

Checking the block log is also something one should be expected to do; anyone who has been blocked for BLP/copyright/etc. issue, even on an already existing article, can be reasonably expected to have similar issues when it comes to creating new articles. NW (Talk) 17:46, 30 January 2011 (UTC)

I've added both of those in, thanks. --Mkativerata (talk) 18:46, 30 January 2011 (UTC)

Automating submissions for autopatrol right

I started a thread about automating submissions for autopatrol right at Wikipedia:Village pump (proposals)#Automating submissions for Autopatrol right. I'm leaving a note here for anybody that might be interested in commenting. - Hydroxonium (talk) 13:39, 12 March 2011 (UTC)

Recent change in number of articles created

Hi there everyone! Can anyone tell me why the minimum number of articles to get the permission granted has been recently increased from 40 to 50? Thanks.--Jetstreamer (talk) 18:51, 17 January 2012 (UTC)

I have 3 reasons for it:

1ST; More experince is always better and it isnt a problem cuz if a editor can make 40 notable articles, then they can surely make 10 more to get.

2ND; And also 50 sounds better then 40 as its half of century.

3RD; As wikipedia has covered almost all major notable topics, it is important that a little more effort should be made to increase editorial standards.

I hope that it was little helpful answer! Yasht 101 06:36, 9 March 2012 (UTC)

Miss use

Is there somewhere to report if you are concerned about articles someone is creating who has this right.Edinburgh Wanderer 21:09, 5 March 2012 (UTC)

Use wp:ani.--Deathlaser : Chat 17:43, 4 May 2012 (UTC)

Reviewer bar

No longer see the reviewer bar to mark new pages as reviewed. Tried on three different computers using Chrome, Safari, and Firefox. Any ideas? Mkdw ^talk 08:56, 31 March 2013 (UTC)

Request for Comment: Including AFC acceptance

NO CONSENSUS

No consensus to implement this change. (non-admin closure) — {{U|Technical 13}} ^{(e • t • c)} 01:48, 4 May 2015 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Hello I would like to open a discussion on the limits to being granted the Autopratolled user right. I would like to see if there would be enough support to change the requirements to not only include new articles created by the user in question, but also AFC submissions which have accepted by the user as the same. I understand the purpose of the article limit is to ensure the applicant has a stable understanding of inclusion policies, which many AFC reviewers will also have. - McMatter ^(talk)/_(contrib) 19:10, 15 February 2015 (UTC)

Do you mean counting accepted AfC submissions as articles created or requiring both accepted AfC submissions and articles created? Pathore (talk) 21:34, 16 February 2015 (UTC)

I mean to lower the bar a little, by counting Accepting AFC submission and moving them to the Main article space as equivalent to creating an article yourself for the purpose of this user right. As both must demonstrate understanding of Wikipedia policies. With the bar lower, this could possibly allow for this user right to be used for other purposes such as allowing autopatrollers to override salted articles. - McMatter ^(talk)/_(contrib) 21:47, 16 February 2015 (UTC)

Why? Moving AfC doesn't increase the backlog; there's no reason to count that for autopatrolled. See also WP:HATSHOP. --L235 (t / c / ping in reply) 22:06, 16 February 2015 (UTC)

I don't see what your reference to WP:HATSHOP has to do with anything here, unless you are accusing me of this. Even if this came to be, would I not want nor qualify for the user right. I am not a new user, and have simply asked this question here to try and help improve Wikipedia and instead of providing helpful feedback on the question at hand you talk about the AFC backlog which I didn't even refer to and make passive accusations on my motives.- McMatter ^(talk)/_(contrib) 04:29, 21 February 2015 (UTC)

Wait—what?! Just clicking "Accept" on someone else's AfC submission counting the same as actually writing an article yourself? And then allowing anyone who has accepted enough AfCs (even though providing this kind of perverse incentive will reduce the quality of accepted AfCs) to create pages with names that have been salted? That's a very bad idea. Pathore (talk) 22:54, 16 February 2015 (UTC)

There is currently a RFC going on to require AFC reviewers to be accepted by Admins prior to being able to use the AFC script. Obviously there would have to be some checks and balances just like any other user right, an admin has to approve it. The guidelines for any user right are just that a guideline and user rights can be revoked without warning if the user using them is being abusive or irresponsible with them. The salted article was one idea I thought this user right could also be helpful towards, amongst some other potential good uses as the user with this right should have a very good understanding of how Wikipedia articles should be. I do understand your concerns and thank you for your input.- McMatter ^(talk)/_(contrib) 04:29, 21 February 2015 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Number of articles

Hi, everyone. I noticed that in order to be given this right, it is recommended that you have created 50 sourced articles. This does not seem like a good standard, because if you have let's say 25 GAs/FAs, you are obviously better qualified/trusted to receive the user right, than someone with 50 one-source stubs. I'm just posting to see if anyone else feels that way. TRLIJC19 (talk) 05:57, 11 July 2012 (UTC)

also the question of mass non-article creations has been raised at the last entry there where the user feels the need to reduce the burden on the patrollers but is not creating articles. I personally would see no harm in given the right in that instance. The question is would it benefit? Agathoclea (talk) 08:59, 26 July 2012 (UTC)

Somebody who has created 50 one sourced stubs on clearly notable subjects may have grasped our notability criteria better than someone who has created 25 GAs on barely notable subjects and had 15 articles deleted as non-notable. FA and GA are about the quality of an article. Autopatroller is for people who give a valid start to lots of articles, the two may not overlap well. I suspect that fifty is a ridiculously high bar and that most article creators would be ready for the flag long before. However every time I've looked we have had a huge backlog of experienced article creators who we could appoint as Autopatrollers, so the 50 threshold at least controls the number of editors who self nominate for the flag. Ϣere SpielChequers 10:57, 31 March 2013 (UTC)

I Agree with TRLIJC19.. I personally think criteria should not be just 50 articles but the live article edits also should also be considered. Say if 4000 to 5000 edits are put to criteria, editors would also tend to develop articles they created rather than just creating and leaving it. I think genreral tendency to create and leave can be avoided upto certain extent. KAS^(talk) 12:32, 14 November 2013 (UTC)

I don't see any compelling reason to change the current recommendation. As one the most regular admins at WP:PERM for the past couple of years, where a certain amount of admin discretion is allowed, I would certainly accord Autopatrolled to an editor who has created, say, 25 medium sized articles, perfectly formatted and sourced, and who has no issues raised on their talk page, while I would often decline some applicants who have even created over 100 articles if they are mainly stubs, redirects, and dab pages, and/or some are still tagged for attention. Note that I take the time to review every single creation, or at least until I come to the firm conclusion that I will decline. Kudpung กุดผึ้ง (talk) 00:28, 15 November 2013 (UTC)

I know admin's would review correctly, But I was also talking about the things that could be avoided like create and leave tendency upto a certain extent, if the edit counts are also put into consideration.. KAS^(talk) 08:56, 18 February 2014 (UTC)

you talk of create and leave almost as if it was a bad thing and not the opposite of owning articles. Autopatroller is a flag to make new page patrol more efficient. Whether an article is going to remain a valid stub or rapidly grow to FA is irrelevant to the new page patrol process. What matters is whether it needs review by new page patrollers or not. Ϣere SpielChequers 12:19, 26 July 2015 (UTC)

Autopatrolled – reduce the number of qualifying articles... again?...

I am toying with the idea of proposing this. The original level was "75 articles" authored to be auto-granted "Autopatrolled". (I wonder how many editors have even written 75 new articles?!!) Then, in early 2011, there was the following discussion: Wikipedia:Village pump (policy)/Archive 83#Autopatrolled - reduce number of qualifying articles – while not a formal RfC (in fact, someone in this discussion suggested that an RfC was "overkill"), there was consensus to reduce the number, though no real consensus on what the level should be reduced to. WhatamIdoing took the initiative to reduce the suggested number of articles to auto-qualify for "Autopatrolled" in the accompanying documentation to its current number of "50" new authored articles (I wonder how many editors have written 50 new articles too?!), with the caveat that WhatamIdoing "expected the number to decline over time" (at least, I think that's what was said – I can't find it now!!).

Well, I'm now wondering if it's time to reduce the number of qualifying articles again. Partly because I think the level is too high, and partly because I don't think we want the page patrollers wasting time on editors who are presumably experienced editors. My personal feeling is that the level in the documented guideline should be reduced to 25–30 new articles authored to qualify for Autopatrolled status. (Though I'll add that I think "Autopatrolled" rights should still be granted at the granting Admins discretion, regardless of the actual number of articles authored, to preventing "gaming" of the system...)

My questions are:

Is there any interest in revisiting the question of how many articles need to be written to qualify for "Autopatrolled" status? And,
If there is interest in revisiting this, should it be done at WP:VPP again? And should it come in the form of a formal RfC or not?

I welcome any thoughts on this. --IJBall (contribs • talk) 07:44, 26 July 2015 (UTC)

Wikipedia:Database_reports/Editors_eligible_for_Autopatrol_privilege hasn't run for over a year now, but showed nearly 600 candidates last June. I'm pretty sure it would be a longer report now. That doesn't answer the question of how many editors will have created over 50 articles as it only shows editors who were active when it was run. But it does show we have lots of editors who meet that part of the criteria, though many of them clearly fail for other reasons. I would suggest doing a bit of research before proposing a change, if you can get a list of active eligible editors with 40-49 articles contributed then it would be worth going through and appointing the obvious ones. Such a trawl would give an idea as to whether it was worth lowering the threshold. Alternatively look at some random editors who have the auto patrolled right and check whether they were ready for it by their fortieth or fiftieth article. My suspicion is that those who merit autopatroller usually reach that standard long before their fiftieth article, but it should be possible to test that. Ϣere SpielChequers 12:04, 26 July 2015 (UTC)

OK, it looks like the data you'd like is probably appraisable(is this even a word?!) from Wikipedia:List of Wikipedians by article count – we'll want to look at just "non-redirects" data column for this, 'natch! I'm coming to the end of my second summer job this week, so I may have time to parse the data in a few days... --IJBall (contribs • talk) 15:48, 26 July 2015 (UTC)

Update: In checking Special:Log/rights, I came across one instance in which an editor was granted "Autopatrolled" rights in the last 24 or so hours will just under 30 (non-redirect) article creations. (In the last two or so weeks, another editor was granted "Autopatrolled" with just approx. 35 non-redirect articles created.) So it looks like the "50 created articles" level is only loosely followed right now as it is... --IJBall (contribs • talk) 16:56, 26 July 2015 (UTC)

Yes, it's a guideline and there are three different routes by which the right is dished out. Some admins, probably at newpage patrol will come across newish editors who have clearly grasped notability and referencing and just appoint them as autopatrollers - I suspect that accounts for your two recent examples. Many people get nominated or are nominated for this user right, and occasionally when we have a list I and others trawl through the list of prolific article creators and appoint suitable ones as autopatrollers. Of the three the fifty threshold is most relevant at requests for permissions where it gives newbies time to get the basics right before they ask for the userright. But fifty may be too high a threshold so a bit of research would be helpful. Ϣere SpielChequers 19:01, 26 July 2015 (UTC)

If memory serves (I haven't bothered to click the link to the archive that IJBall linked above) there was no solid consensus for any particular number, and basically no support for any number above 25. I changed it to 50 (a) so that something would actually happen, and (b) because nobody could have objected to 50 when the talk is all about 10 to 25 being sufficient. Lowering it to 50 does not seem to have hurt the project. I would be perfectly happy to see it changed to 25 or 30 — and if we notice next year that 25 is causing no harm, then I'd be willing to see it dropped to 20, or maybe even lower. WhatamIdoing (talk) 22:29, 26 July 2015 (UTC)

OK, in light of that, then I think the next step is to gather some of the data that WSC asked for, and then follow up by starting a topic at WP:VPP, barring any objections here... --IJBall (contribs • talk) 00:58, 27 July 2015 (UTC)

I'm happy to see the above sentiments. As a regular at permissions, I have always thought that the 50 article threshold was excessive. I understand the justification is that the right's meant for 'highly active' article creators, not just trustworthy ones in general. But frankly this principle excludes editors who have established trustworthiness well before reaching the 50 article threshold and I don't see any way such a high benchmark protects the encyclopedia. I generally treat the guideline pretty flexibly and will seriously consider requests of editors with 25-30 articles and based on my observations other admins at Permissions do the same thing. It would be interesting to see if lowering the benchmark would be worth it and if it would I would squarely support it. Swarm ♠ 00:16, 4 August 2015 (UTC)

WP:VPP update

Update: I've posted the proposal to lower the requirement to 25 valid articles created to be granted Autopatrolled rights at WP:VPP, here:

Wikipedia:Village pump (policy)/Archive 122#Threshold for Autopatrolled rights. --IJBall (contribs • talk) 23:19, 11 August 2015 (UTC) (Updated link --IJBall (contribs • talk) 19:55, 27 August 2015 (UTC) )

Implemented --IJBall (contribs • talk) 19:55, 27 August 2015 (UTC)

Reverted pending a properly published RfC with representative comments and closed by an univolved editor. --Kudpung กุดผึ้ง (talk) 02:38, 9 September 2015 (UTC)
The change has been reimplemented after review on WP:AN. — Martin (MSGJ · talk) 09:41, 9 September 2015 (UTC)

Editors eligible for Autopatrol privilege report has been updated

Hey guys, this report has been updated (after being broken for a year or so): Wikipedia:Database reports/Editors eligible for Autopatrol privilege. Ryan Kaldari (WMF) (talk) 22:52, 2 November 2015 (UTC)

Is that report actually useful to people? Ryan Kaldari (WMF) (talk) 22:55, 2 November 2015 (UTC)

@Ryan Kaldari (WMF): It would be more useful if this would only include users who have recently edited. Most of the reported users haven't edited in a while. Cenarium (talk) 02:07, 15 January 2016 (UTC)

Peter F Morim Peter morin (talk) 07:55, 3 April 2017 (UTC)

Wording

Under 'Obtaining the right' it says "The autopatrolled right is not needed by those who patrol pages (though in practice the qualifications are similar)." I thought the NPP right doesn't allow you to patrol your own articles? Does this wording need correcting? Kb.au (talk) 06:14, 20 January 2018 (UTC)

Reviewing and removing autopatrolled

A few weeks ago, a bug was introduced that caused autopatrolled user's edits to appear in the new page feed for several days. There was some discussion at WT:NPR about the fact not all of these articles got the rubber stamp you'd expect, but required clean-up, tagging, and in some cases even deletion. This is something I've also noticed when reviewing new articles through WikiProject alerts, which include autopatrolled articles. A common denominator seems to be that the editors were given autopatrolled a long time ago, perhaps when it was given out more haphazardly, or perhaps they simply haven't kept up to date on content guidelines for new articles.

At that time my understanding was that administrators could revoke the right at their discretion, but my attempt to put that into practice immediately ruffled feathers and was undone. That has discouraged me from tackling several similar cases since then. But I still think we have a significant, unaddressed problem in the lack of oversight of autopatrolled creations. It is not hard to find autopatrolled editors whose articles really ought to be seen by a second pair of eyes. The question is, how should patrolling admins go about reviewing and revoking the right from them? If unilaterally conducting a review it is going to cause a fuss, should there be somewhere to seek a second opinion on an editor's creations? How do we do that without creating an unpleasant WP:ANI-style atmosphere of accusation?

It's a touchy subject because often we're talking about experienced and valued content creators who take the removal of the right as an admonishment. In my view it isn't like that: having your contributions reviewed by other editors is the normal state of affairs across the project, and autopatrolled is an exception that only exists to keep highly prolific article creators from disrupting NPP. (I wish I could remove autopatrolled from my account to make the point that not having it isn't a big deal). I'd like to find a way of making losing the right be as unremarkable as possible, for example making it temporary by default, and encouraging editors not to automatically request it be reinstated when it lapses, unless they really need it. – Joe (talk) 13:19, 28 July 2018 (UTC)

(Non-administrator comment)But isn't it an admonishment? It is the community (or at least one admin) saying that the community (or one admin) no longer feels you have the trust to make articles without at least one other pair of eyes on it. This coupled with loss aversion suggests to me there won't be any way to make this routine. Which isn't to say it isn't still the right thing to do... Best, Barkeep49 (talk) 14:12, 28 July 2018 (UTC)

The only editors who are affected by autopatrolled are new page reviewers. A sensible reform would be to let only new page reviewers request it for someone who is creating so many and only high quality articles that reviewing them unnecessarily creates or adds to a backlog, provided the nominee agrees of course. Vexations (talk) 14:33, 28 July 2018 (UTC)

That's not a bad idea. I've left many messages on user talk pages encouraging them to check out Autopatrolled rights because all their pages were high quality - they just didn't know it was a thing. SEMMENDINGER (talk) 15:25, 28 July 2018 (UTC)

You can already request that other editors be granted autopatrolled. There is even a bot that maintains a list of eligible editors. – Joe (talk) 21:17, 28 July 2018 (UTC)

I'm so offended that I'm not on the bot-generated list. Although I'm actually not sure why I'm not. Natureium (talk) 11:42, 30 July 2018 (UTC)

It's not an admonishment. It doesn't even benefit the editor any, it's just a matter of pride. (I marked as reviewed an article during the autopatrolled outage and got a message on my talk page from the creator that seemed to be mildly offended that their page was reviewed.) I have also found articles that were autopatrolled that have issues that need to be addressed. If it's widespread enough of an issue, maybe some announcement should be made that due to updates in standards, permissions granted more than x years ago will be reviewed and if the editor isn't creating a number of articles that would be burdensome to the NPP process, the perm will be removed and can be reapplied for if the editor believes that their articles being autopatrolled would be a benefit to the project. Perhaps this would be less likely to offend. Natureium (talk) 18:01, 28 July 2018 (UTC)

Maybe we should revoke everybody's flag, and request them to apply after creating one or two articles (after revocation). We have gone through backlogs of 21,000; and we have handled "broken autopatrolled" situation as well. I think we can easily handle the intentionally broken autopatrol. —usernamekiran(talk) 19:11, 28 July 2018 (UTC)

Seems kind of drastic - has a purge of this caliber ever been done before in any of the other user right pages? (Admin, page mover, rollbacker, etc). I'd be interested in hearing @TonyBallioni:'s thoughts on this since he's a big driving force behind granting the privilege in the first place. SEMMENDINGER (talk) 19:49, 28 July 2018 (UTC)

There are ~4,000 autopatrolled accounts. Dealing with regranting that while checking who merits it might be more hassle than benefit. I’m not an admin though, so I shouldn’t presume to know it’s thats something they would want to take on. Natureium (talk) 20:01, 28 July 2018 (UTC)

I'd view it from a cost-benefit standpoint. There is a large cost here (time reviewing 4000 user's articles and hurt feelings) and I'm not sure the benefit of making sure that all of these articles get review outweighs the cost of the effort. I think Joe makes a valid point, though. If this were a bolded thing, I'd be neutral leaning oppose. TonyBallioni (talk) 21:35, 28 July 2018 (UTC)
Why don't we have an option in the new pages feed that allows people to view articles created by autopatrolled editors? I think that making autopatrolled, say, temporary, would be a bad idea. While there certain are some editors who shouldn't be autopatrolled, those who are sometimes have a very high output. I recall, for example, looking at the reviewed queue, and it was basically taken up by Notafly's creations. There are a good number of other editors that have an interest in biology and create a lot of articles on species. If these were visible, even for a day or too, in the unreviewed queue, we would be overwhelmed. RileyBugz^{私に叫ぼう}_私の編集 00:45, 29 July 2018 (UTC)

@RileyBugz: If you use the [tools nppbrowswer] there is a filter for autopatrolled. Best, Barkeep49 (talk) 03:15, 29 July 2018 (UTC)`

Well, I did say temporary by default. If someone is maintaining a high rate of creations indefinitely then, sure, give them autopatrolled indefinitely. I think that there are relatively few of those, editors though. For the vast majority of us, our activity waxes and wanes. The main aim is to have some sort of systematic review and avoid the current situation, which is essentially, "Created 25 decent articles? Sure, have immunity from community review forever!" – Joe (talk) 12:47, 29 July 2018 (UTC)

That's why I'm suggesting we have the option (on the New Pages feed itself) to view creations by only autopatrolled editors. If we could do that, then there should be more people paying attention to those creations, just on the fact that you can easily switch it on, and it's not very hard to do. I do agree with the fact that removing the userright should be no big deal, but I don't think making it temporary for most people is the solution. Again, even one day without autopatrolled for a prolific creator will hurt us badly. RileyBugz^{私に叫ぼう}_私の編集 14:22, 29 July 2018 (UTC)

I fully concur with Joe's concerns. Over the years I've come across plenty of autopatrolled users who I thought should no longer have the right but I usually let it go–I'm not as mean as some people claim ;). We admins have to take enough flak as it is for just doing our job, and flak is what you get when you take away a user right, even in the case, for example, of some New Page Reviewers who get it wrong too often or who have simply never used the right. Quis custodiet ipsos custodes? Well, as Tony says, it's a cost/benefit issue; autopatrolled is supposed to alleviate the task of patrolling new pages, not add to the work. All minor rights should be 'easy come; easy go'. If we come across such autopatrollers by coincidence, sure, do something about it, perhaps put them on time limited probation now that the feature exists, but I would prefer our New Page Reviewers to concentrate on keeping the backlog down which is already rising again sharply. Kudpung กุดผึ้ง (talk) 11:24, 29 July 2018 (UTC)

Well put. I think dealing with it when it pops up is less time consuming than the other options. Removing all autopatrolled rights would just make the application process flooded as everyone scrambles to reclaim it - and exponentially more so if the right is only temporary and the same users are requesting it every X amount of days. The other solution of going through 3000 user contribs to see who should retain the right is also ridiculous. That's a lot of behind the scenes time which would only fractionally impact the encyclopedia at large. I'd rather those man-hours go towards mainspace edits than a wikipedia talk page. SEMMENDINGER (talk) 16:41, 29 July 2018 (UTC)

There is already a filter for autopatrolled pages in the New Pages Feed preferences, RileyBugz and Barkeep49. It's called Reviewed pages . Probably my fault if it's not sufficiently recognisable - I worked (as a volunteer) with the WMF team that developed the software 7 years ago. It can easily be changed to something more obvious, but as previously stated, I would guard against turning it into a witch hunt. -Kudpung กุดผึ้ง (talk) 01:44, 30 July 2018 (UTC)

Yeah, I know about the reviewed pages thing, but that's just reviewed pages in general, not autopatrolled articles in the specific. If there was a separate thing for autopatrolled articles, it would be easier for NPRs to make sure that autopatrolled users are creating quality articles; the workflow of just marking an article as "unreviewed" and "investigating" (in the broadest sense) an editor in addition to unreviewing their article is different. RileyBugz^{私に叫ぼう}_私の編集 01:48, 30 July 2018 (UTC)

I'm sorry to disagree with you RileyBugz, but it lists Autopatrolled pages and that is its purpose. There is an extremely small number of non-autopatrolled pages in that list but that is a technical issue due probably to a short server lag in updating the feed. They should certainly be gone as soon as the feed is refreshed. If not, let me know and I will file a bug report in the appropriate place. I will remind again however, that patrolling new pages is far more important than going on a counter-productive witch hunt - the purpose of Autopatrolled is precisely the opposite. There is nothing non-admins can do if they come across poorly performing autopatrollers, except perhaps notify an admin. Thank you. Kudpung กุดผึ้ง (talk) 02:34, 30 July 2018 (UTC)

Oh. Yeah, that should probably be clarified. I'm not proposing to go on a witch-hunt (I certainly wouldn't take part in it; too much work and no benefit!), but I'm saying that if this is in fact a problem, it would be useful to be able to have people being able to take a second look at autopatrolled articles if they want. Thanks for the clarification, though. RileyBugz^{私に叫ぼう}_私の編集 02:38, 30 July 2018 (UTC)

@Kudpung: Thanks. I had thought that displayed both autopatrolled and pages reviewed by NPP. I don't tend to patrol there but it's good to know. Best, Barkeep49 (talk) 12:47, 30 July 2018 (UTC)

A possible 'soft' solution

By checking out the filter preference I mentioned above, I found that the situation is probably worse than Joe imagined. My suggestion is to send a mass messaage on the NPR newsletter template to all autopatrolled users on the lines of:

Hi, {{BASEPAGENAME}} thank you for creating so many new articles. Concomittant with new developments for the New Page Review system, the Autopatrolled user group is currently under discussion. This right is not a 'right' per se; it is designed to reduce the workload of our New Page Reviewers and does not affect the actual work of the Autopatrollers or accord them additional editing privileges or tools. Autopatrolled pages nevertheless do appear in the New Pages Feed and are subject to random reviews. Users found to be not, or no longer observing the high standards required for Autopatrolled may find that the right has been removed at administrator discretion. This message has been sent to all 3,874 members of the Autopatrolled user group.

As this doesn't address anyone personally, no one could complain, and it might just have the required knee-jerk. Kudpung กุดผึ้ง (talk) 02:06, 30 July 2018 (UTC)

Can autopatrol right be removed?

I came across a number of articles that were autopatrolled consisting of one, two, or three sentences, and no sources. They were created by the same person, can someone who keeps creating such articles have their autopatrol rights removed? Hzh (talk) 13:12, 19 August 2018 (UTC)

Absolutely. Which user are we talking about? – Joe (talk) 15:02, 19 August 2018 (UTC)

The user is DilletantiAnonymous. Example are Thomas More, Lord High Chancellor of England and A Knight at Prayer in a Chapel, Preparing Himself for Combat which I have nominated for deletion. The editor has created a large number of articles on paintings, many of which are no doubt notable, but it is hard to tell for some without sources e.g. Young Woman with a Blue Choker, some such as Young Woman with a Lapdog the sourcing is at best perfunctory and just simple listings. Is there a procedure for nominating someone for removal of right? Hzh (talk) 15:35, 19 August 2018 (UTC)

Thanks. I'll take a look at their creations.

As for the procedure for removing the right, that's a bit unclear; see the discussion immediately above this one. The only guidelines are at WP:PERM, which says you should raise concerns about advanced rights at WP:ANI. I find that a bit drastic and in other contexts (e.g. NPR and AfC) reviewing/removing rights is handled "internally" on the relevant project talk page. So I think it would be reasonable to continue using this talk page or WT:NPR to ask an admin to review autopatrolled creations. – Joe (talk) 16:32, 19 August 2018 (UTC)

@Hzh and Joe Roe: See the thread above. The right can be removed on admin discretion. The people at NPR have no juridiction. Kudpung กุดผึ้ง (talk) 23:33, 19 August 2018 (UTC)

Slow down, everyone. Have any of this editor's creations been deleted as nonnotable? E Eng 23:48, 19 August 2018 (UTC)

@EEng: I've already removed the right. Just creating articles articles on notable subjects is not enough; autopatrolled is only given to editors who regularly and reliable write completely "clean" articles (i.e. with no cleanup tags, in this case {{unsourced}}). – Joe (talk) 05:27, 20 August 2018 (UTC)

This seems heavyhanded given that AFAICT there has been zero attempt to help the editor understand what's expected, but of course this affects him not at all. It's NPP that the burden falls on, and hey, what's another several hundred or thousand articles in the queue? E Eng 05:33, 20 August 2018 (UTC)

Their talk page is a wall of people trying to talk to them about their creations, from both before or after they were given autopatrolled, few of which they have responded to.

Of course this affects him not at all – exactly. We need to get away from the idea that autopatrolled is a badge of good-editorship. It isn't. It is purely a tool for NPP, which comfortably handles hundreds of articles a day. Even a prolific editor like DA does not have an appreciable impact on it individually. – Joe (talk) 06:14, 20 August 2018 (UTC)

I'm sure most reviewers don't look at articles marked as autopatrolled, therefore a lot of the articles of dubious notability would have passed under the radar. At the least for an editor like DilletantiAnonymous, someone else can review their output. I for one wonder what the point is of having three articles on different painting variations - The Satyr and the Peasant (Jordaens), The Satyr and the Peasant (Jordaens, Brussels), The Satyr and the Peasant (Jordaens, Munich), each barely viewed by anyone and they are unnecessary duplications (the way to do variants should be like Madonna (Munch painting) where they are all placed under the same title). Given a free rein, the editor might well create separate articles for all seven variations. Perhaps with someone looking over what the editor did, it might actually reduce the number of pointless articles created. Hzh (talk) 09:26, 20 August 2018 (UTC)

Revising the minimum criteria

The current minimum criteria for autopatrolled is the creation of 25 'clean' (i.e. notable, not tagged or deleted) articles. Originally, the bar was considerably higher: 75 articles. Reviewing the discussions above, it was lowered to 50 and then 25 primarily because of a shift in emphasis, from reserving autopatrolled for "highly active" editors, to granting it to any "trustworthy" editor.

Given ongoing concerns about problematic autopatrolled creations, I think this standard is too low. It has resulted in autopatrolled being given to many editors who may be trustworthy, but don't actually create many articles. This was perhaps justifiable when NPP was in crisis, but we now have a healthy community of reviewers, so giving the right to an editor who creates a few articles a month has no appreciable impact. And since we can't one hundred percent-reliably identify who is "trustworthy", nor guarantee that they will remain so forever, each editor who is granted autopatrolled when they don't really need it represents a small—but wholly unnecessary—risk to the integrity of the project.

To address this, rather than simply raise the bar, I propose that we split it into two new criteria, to assess both the trustworthiness of an editor and whether there is a genuine need for autopatrolled. The exact standard can be tweaked, but something like:

Regularly creates new articles: an average of 3 or more per month over a period of at least one year
Reliably creates 'clean' articles: less than 5% of creations have been deleted or tagged for cleanup

I think this also more closely matches what admins here actually look for when granting the right, so there should be less surprises for people requesting it. – Joe (talk) 15:59, 14 September 2018 (UTC)

Would this have helped us avoid some of the misses? I can agree that the bar is too low now but I think finding a balance so that some of our prolific content creators don't end up going through the queue while running out a clock is important. I think the community of reviewers is in a good place but we recently went through a month where the queue was growing rather than stable or decreasing but keeping people in the queue who would have been taken out under other criteria will help us stay in that good place. Best, Barkeep49 (talk) 16:10, 14 September 2018 (UTC)

The ultimate goal of this proposal would be to (moderately) reduce the number of autopatrolled editors, which by definition will reduce the number of mishaps.

NPP will always have ups and downs, but if you compare the situation now to where we were two years ago, it's like night and day. I wish I had some numbers to back it up, but my intuition is that there is a "long tail" of low-activity autopatrolled creators that have little to no bearing on the size of the queue. – Joe (talk) 16:24, 14 September 2018 (UTC)

If you get your own CSD do you keep autopatrolled?

User:The Transhumanist is an auto reviewer/auto patrolled Legacypac (talk) 00:05, 12 March 2019 (UTC)

I started to respond when I remembered that this is not a user conduct page. Best, Barkeep49 (talk) 00:13, 12 March 2019 (UTC)

According to discussions on this page this is the correct place to raise possible user right removals. Legacypac (talk) 00:18, 12 March 2019 (UTC)

I went ahead and did it as a unilateral admin action. Anyone is free to restore via a PERM request, but I don't see further ANI action as being helpful to anyone, and I think having a CSD criteria devoted to your creations in any namespace raises sufficient concerns that revoking is warranted. TonyBallioni (talk) 00:24, 12 March 2019 (UTC)

Notification of related bot request

Wikipedia:Bots/Requests for approval/DannyS712 bot 38 § DannyS712 bot 38 DannyS712 (talk) 21:27, 14 May 2019 (UTC)

"Autopatrolled" listed at Redirects for discussion

A discussion is taking place to address the redirect Autopatrolled. The discussion will occur at Wikipedia:Redirects for discussion/Log/2020 June 11#Autopatrolled until a consensus is reached, and readers of this page are welcome to contribute to the discussion. Hog Farm (talk) 18:32, 11 June 2020 (UTC)

Creation by admin not autopatrolled

Am I missing something in this article? I just marked that article as patrolled, and then I realized it was created by an admin. I thought autopatrol is included in the admin package? --Dps04 (talk) 10:40, 16 May 2020 (UTC)

@Dps04: What you are missing is that setting up or changing a redirect (the first and second edit) don't trigger the new article process. It was the third edit that created the article and that was not done by an admin. Schwede 66 11:26, 16 May 2020 (UTC)

@Schwede66: Thanks for telling me that. Guess I got confused cause page curation told me that the admin was the article creator. Good thing I didn't send a message to them through the tool or else it would have reached an unintended target. Thanks. --Dps04 (talk) 11:33, 16 May 2020 (UTC)

@Schwede66: I have noticed the same thing. Blackcurrant production in the United States was created by administrator Dumelow and appears on the new page feed and Mandai Wildlife Bridge was created by administrator Rlendog and also appears there. Cwmhiraeth (talk) 08:44, 13 June 2020 (UTC)

Limit to third party nominations?

Rosguill and I were discussing the latest UPE who got AfC/NPP. Ros expressed some skepticism about the usefulness of Autopatrol at least when it comes to evaluating new people for the PERM. I countered that many of the editors who best qualify for the PERM already have it so we shouldn't get rid of it. But maybe, we should limit the ability requests for the PERM to third parties only. This could potentially discourage hat collectors while also perhaps better fulfilling the purpose behind this PERM (to reduce the burden on NPP). What do others think? Best, Barkeep49 (talk) 19:03, 10 October 2020 (UTC)

Well I see this didn't generate much discussion here (perhaps it did elsewhere?) but I'm inclined to agree. The people who should be requesting autopatrolled are NPP regulars requesting the "permission" be added to third-party accounts to ease the NPP burden. Limiting it to such won't remove the possibility of folks gaming the system, but I think it would cut down on spurious requests, and save the admins who watch Wikipedia:Requests_for_permissions/Autopatrolled some time. With that in mind, if you're interested in generating more discussion, perhaps a post at Wikipedia talk:Requests for permissions would catch those who watch the PERM queue? Ajpolino (talk) 19:11, 27 November 2020 (UTC)