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.
There is community consensus for the introduction of a requestable permission which will be required to review articles at Articles for Creation.
The participants were not asked any further questions about this permission (such as who can request it and how they do so), these questions will need to be resolved later on. Below is a summary of the primary arguments as they stood out to me, it is certainly not a complete summary of all the reasons presented in the request for comment.
The majority of the support arguments focused on the need to have experienced users checking to ensure that draft articles meet content policies and guidelines. Some supporters referred to instances of inexperienced users accepting articles, moving them into mainspace to be subsequently deleted, in some cases speedily deleted. Supporters also suggested that there have been times when submissions may have 'slipped through the net' and exist in mainspace (whether with maintanence tags or not seen at all), when they ought to have been declined and redrafted.
It provides an opportunity for hat collecting. This argument was countered with the suggestion that admins already face this is other areas [at WP:PERM] and are able to filter out the hat collectors.
Other reasons for oppose votes included that a requestable permission introduces unneeded and unwarranted bureaucracy, makes it harder for submissions by IP users and others who want someone to look at their article before it does live, and that it provides a hoop for editors to jump through in an already backlogged area which may discourage editors from helping out and reviewing submissions.
NOTE: To keep it simple, this subsequent RfC is to determine whether a more broadly represented community would support reviewer permission in principle. Please do not start alternative proposals here.
NOTE: The community is not asked at this stage how such a right or permission should be implemented and/or what its threshold should be. This should be discussed if consensus is reached. Please do not start alternative proposals here.
NOTE: This RfC is not a discussion on, or suggestions for, any other reforms to the AfC system. Please do not start alternative proposals here.
The community is asked whether the ability to review articles at Articles for Creation should be be a requestable permission.
A preliminary discussion at Make AfC reviewing a requestable permission demonstrated a consensus with very little opposition that the introduction of a user right or a permission for reviewing pages at Articles for creation might be appropriate. See that section and for more background, please consider reviewing the following related discussion threads in various other parts of Wikipedia before commenting:
Please begin your comment in the 'support' or 'oppose' sections with a hash mark (#). Please indent your replies to the comments with a colon (:) according to standard talk page threading.
This RfC will be open for 30 days, or closed earlier if an overwhelming consensus either way is obvious.
This RfC has been noted on the Village Pump, Central RfC, RfC notifications, and personally in a neutral and identical manner to all users who participated on or edited the previous RfC and who have participated in similar related discussions.
Support I proposed this on the now closed RFC and still support it. It is vitally important that newcomers get the correct guidance to get AfC submissions into mainspace properly. Ritchie333(talk)(cont) 07:25, 24 August 2013 (UTC)
Support - I also supported this strongly on the closed RfC. As a dev for the AfC helper script, I think that the ability to review should be a permission. Given the current system, there is great potential for inexperienced reviewers. APerson (talk!) 07:46, 24 August 2013 (UTC)
Support long overdue. I've seen too many editors who have no clue what they are doing in venues like AfC, and damaging the encyclopedia. --Rschen7754 08:04, 24 August 2013 (UTC)
Support - Too many AfC editors are over-eager in their enterprises, and too many show a poor grasp of policies. AfC is used almost solely by new or inexperienced editors, so they need experienced veterans to help them through the stages. If AfC is to be retained, this is a must. Lukeno94(tell Luke off here) 08:26, 24 August 2013 (UTC)
Support a light-weight review request process for this right. Technically, this could be implemented by creating an additional user right (e.g. "afc-review"), assigning this user right to a new user group (e.g. "afc-reviewer") as well as existing groups (reviewer, sysop, etc), and then restricting the use of AFC review gadgets/scripts to users with this right. — This, that and the other (talk) 08:50, 24 August 2013 (UTC)
Support. We could bundle it with Wikipedia:Reviewer - just as that right was co-opted for WP:AFT-reviewing. That would prevent the unlikely possibility of an existing cabal at AFC becoming gatekeepers to the process, if we left the granting of the permission exclusively to existing AFC regulars. (Of course, as now, regulars at AFC would be able to use the usual community processes to manage incompetent or corrupt behaviour by "qualified" reviewers.)
Reviewers of newbies' efforts must have demonstrated some understanding of content policy. I recently anonymously submitted a policy-compliant article to AFC and had it rejected, and I've seen that done to hundreds of newbies over the last couple of months. The good work of the qualified reviewers is being undermined by unready users, who simply turn up and start reviewing. --Anthonyhcole (talk · contribs · email) 11:07, 24 August 2013 (UTC)
Support. This should not be a difficult process though; maybe either use the existing reviewer right or just set an edit count. We are always short of reviewers. The main thing is that it be something that can be revoked if necessary. Creating the right in one of these two ways should help to allay Andy Dingley's concerns below. I would like to point out that not having access to the tools would not prevent users from adding incorrect templates and afc comments manually, moving articles to the wrong space accidentally, etc. Also, this wouldn't solve all of the problems with inexperience because there are so many different situations that after reviewing and re-reviewing well over 1000 articles I am still coming across things I don't know how to do and have to ask for help. —Anne Delong (talk) 12:00, 24 August 2013 (UTC)
Support - there are a few things wrong with the current AFC system but I see this as critical to resolving one of the bigger issues. The community can't give new users the assurance of "approval" for every new article, but we can give the assurance, at least, of an "approved approver". Stalwart111 12:24, 24 August 2013 (UTC)
Support per Anne Delong. Something similar to the pending changes reviewer right would IMHO go a long way towards fixing the problem of poor quality reviews. The criteria are pracically identical to what I believe are reasonable "qualifications" to review AfCs. Roger (Dodger67) (talk) 12:49, 24 August 2013 (UTC)
Support, maybe bundling with other rights. Johnbod (talk) 13:36, 24 August 2013 (UTC)
Support, I believe there is significant confusion that comes out of reviews that are improperly done-- users frequently visit the teahouse not knowing how to go about fixing articles. While this sometimes might be the fault of the article creator's competence, I suspect they are confused because of the kind of unclear advice given from inexperienced article reviewers. Alternatively, inexperienced reviewers may release malformed articles not ready for the mainspace get approved anyway. This is not only bad for Wikipedia, but a negative experience for new users who face immediate deletion despite approval in AFC. I, JethroBTdrop me a line 13:51, 24 August 2013 (UTC)
Support Regrettably, the general knowledge level of some users is deficient. There is no purpose in having an article reviewed by a user who will not produce an encyclopedic article. I can only echo the very good points made above. Dlohcierekim 13:56, 24 August 2013 (UTC)
Support: This should be a requested permission, regardless what that permission actually is. I'll spend some time researching writing a new MediaWiki Extension for AfC since this project has reached a level where there is a need for a custom namespace, additional set of userrights, more error resistant wizard, etc. Technical 13 (talk) 14:29, 24 August 2013 (UTC)
Support - At present the reviewers are just as inexperienced as the creators. Robert McClenon (talk) 15:05, 24 August 2013 (UTC)
Support, though I think we should add a new user right instead of bundling it with Reviewer or something. Knowing whether pending changes are appropriate is a different skill set than knowing if an article is ready for mainspace or not. Jackmcbarn (talk) 15:07, 24 August 2013 (UTC)
Support putting it in the Reviewer right. This allows for effective control, limited AfC reviewing to only those who are reasonably competent in our content policies, but avoids the hat-collecting that would come with a new user right. -- King of♥♦♣ ♠ 15:45, 24 August 2013 (UTC)
Support, in general, and because this right would also be very useful if the requested article process is changed (see discussion below). -- John Broughton(♫♫) 19:43, 24 August 2013 (UTC)
Support If AfC is to have credibility, reviewers must have the experience and understanding of key policies and notability guidelines necessary to differentiate between an article on a non-notable topic, and a weak article on a notable one. They must be prepared to guide the writers of the second category toward improvement needed to bring the article to the main space. I see the fallout from AfC reviews by inexperienced editors at the Teahouse and AfD quite frequently, and a poor review is unfair to new contributors, whether or not the article survives. Cullen328Let's discuss it 20:02, 24 August 2013 (UTC)
Support per above. theonesean 21:34, 24 August 2013 (UTC)
Support, with the personal caveat that I haven't read all the supporting materials. I don't think that's necessary to support the general concept that reviewing an AfC does require some clue, and it would be worth having some hurdle. I'd need to read more to support a specific hurdle, but I am in favor of having some hurdle.--SPhilbrick(Talk) 00:36, 25 August 2013 (UTC)
Weak support. I am a relatively inexperienced AfC reviewer, and had about 1,000 edits before starting to review in AfC. While I certainly had my fair share of bumps and mistakes, I'd characterize my reviewing overall as pretty good. Because I went through a formal adoption program, and had a fairly protracted debate on a userspace draft I'd written, I was better prepared for most with a similar level of experience. I think that if we implement a new userright, then we need to keep the barrier low. We must not throw the baby out with the bathwater. In addition, AfC is chronically backlogged. Without new blood coming in, we may well wind up with a situation where there is a 5-figure number of articles waiting to be reviewed. When you consider that there is a person on the other end of each review, this comes into more perspective. However, this solution is addressing a legitimate problem, and it is much better to have the reviews undone than done badly, so I have to grudgingly support this proposal. Tazerdadog (talk) 04:13, 25 August 2013 (UTC)
Reviewing at AFC is not like normal editing, it is more akin to admin work. We don't give the deletion button out lightly and we should accept that for the newbies submitting articles at AFC a decline is as "official" as a deletion at newpages. Of course not having this hat should not preclude people from improving AFC submissions and where appropriate tagging as vandalism, hoaxes and attack pages. ϢereSpielChequers 05:41, 25 August 2013 (UTC)
Support, while agreeing with Tazerdadog that the bar should be set relatively low, such as being grantable by any administrator, similar to Autopatrolled or Reviewer rights (but separate from them). The current system, where anyone can review an Article for Creation regardless of experience or clue, is unsustainable. --MelanieN (talk) 18:11, 25 August 2013 (UTC)
Support, with routes to both have it added automatically, and by request. I think the automatic addition should be at a fairly low level, but not a very low level, and be readily withdrawn from a user if it turns out that in their particular case more experience is needed. A backlog is much better than grossly incorrect reviews. — Preceding unsigned comment added by DGG (talk • contribs) 17:09, 25 August 2013
Support as per my comments on the previous discussion, and with my feelings about specifics roughly mirroring those of DGG. --j⚛e deckertalk 01:20, 26 August 2013 (UTC)
Support per many discussions. Implementation details bellow. BO | Talk 16:47, 26 August 2013 (UTC)
Support per above. The admins at WP:PERM should be able to separate the hat collectors from those who are actually interested in helping. MER-C 06:47, 27 August 2013 (UTC)
The few admins who work there seem do a pretty good job and there are relative few misfires. However, that is one possibility from the solutions that would be offered to the community if this current RfC reaches a consensus from for some kind of permission. Interestingly, WP:PERM itself is also a target for comments from inexperienced non-admin commentators who attempt to unnecessarily 'clerk' the process. Kudpung กุดผึ้ง (talk) 04:00, 28 August 2013 (UTC)
Support in principle. I am autopatrolled and have reviewed many AFCs, but I still make mistakes sometimes. It's impossible for a new editor to understand the complexities of Wikipolicy. I'd have to see the implementation details before getting too excited about it, since that would make or break the permission. Andrew327 12:45, 27 August 2013 (UTC)
Support I'll take quality control where we can get it. Jd027 (talk) 15:59, 28 August 2013 (UTC)
Support Bad reviews at AfC are enough of a problem that the issue constantly comes up - there was another thread at AN/I recently that could be added above. And either way - acceptance of inadequate articles that then get nominated for deletion and rejection of articles for invalid reasons - they are really demoralizing to mostly new editors who are trying to follow our processes. As I said in the earlier RfC, reviewing at AfC is attractive, including for well intentioned reasons; despite the backlog there, some sort of gatekeeping measure needs to be introduced (and would at least remove the faulty declines from the backlog!). Yngvadottir (talk) 20:49, 28 August 2013 (UTC)
Support. We should make voluntary contributions easy and barrier-free, where we can. Still, the combination of allowiing anyone to give seemingly definitive advice to very new users while also 'authorizing' or 'denying' their contributions in a way that bypasses new page patrol and other consensus processes is a sound basis for a minimal qualification. It disturbs me that several experienced editors and AfC reviewers have told me the reviewing process can be wildly inconsistent and or often incorrectly executed. For details, I think an existing userright likely can cover this, bundling it with Autopatrolled or Reviewer. As a sidenote perhaps we would pair it with a new page patrol in some way? We would obviously have to be mindful of backlog issues and direct extra attention to the page, at least at first. And we would have to do outreach to existing qualified AfC folks to make sure they get/have the userright so they can keep up their work with no interruption. Ocaasit | c 12:47, 29 August 2013 (UTC)
There are many parallels in AfC and NPP and their related problems. As a software solution, perhaps even the excellent WP:Page Curation tool and its feed list could be cloned and adapted as a system for AfC - it has most of the required elements. As both NPP and AfC require a good knowledge of what constitutes an acceptable article, there could be an argument for creating a 'one size fits all' reviewer permission for them both. However, as 78.26 points out below, conforming to the preamble of this RfC, these are details that can be hashed out later. Kudpung กุดผึ้ง (talk) 02:02, 30 August 2013 (UTC)
Support Just about the worst thing we can do for a new user is tell him that his article is going to be "reviewed", and then either wrongly reject it, wrongly accept it, or give bad advice. If we wrongly reject it, he may never edit again. If we wrongly accept it, it may be subsequently deleted, confusing him, making us look incompetent, and, again, potentially losing the new editor. Requiring competence for this task is the least we should do. Begoontalk 03:14, 30 August 2013 (UTC)
Support. Per all the above. Best regards, Cindy(talk) 06:33, 30 August 2013 (UTC)
Support Some of the worst new articles that I encounter came through the AFC process: unsourced, badly sourced, promotional, etc. Even worse, those articles don't show up on Special:NewPages, so often I only see them much later. Having more competent reviewing by people who have more than a handful of edits themselves would certainly be welcome. --Randykitty (talk) 15:59, 30 August 2013 (UTC)
Support closing AFCs is a pretty important task, all around. It requires logical deductive reasoning (ferreting out spam, hoaxes, libel) as well as tact (dealing with new users). Obviously, it also requires a pretty solid all-around knowledge of Wikipedia policies and community standards too. It only makes sense that it's not a reasonable task to let people dive into after 20 minutes and 3 edits. Andrew Lenahan - Starblind 23:31, 31 August 2013 (UTC)
Support. It's too easy for a clueless AfC reviewer to single-handedly "delete" an article before it even reaches mainspace. After reading the technical discussion below, I don't see any problem with the lack of 100% technical way to prevent reviews. The FA promotion & demotion decisions are taken by FA delegates which also don't have any particular technical bit attached (and their actions could technically be performed by anyone). If AfC reviewer ends up being just a social bit, it's still a good step forward. Someone not using his real name (talk) 02:17, 1 September 2013 (UTC)
Support. Most people working at AfC are dedicated and competent, but occasionally there are problematic reviews done by somebody who doesn't appear to be familiar with our expectations for new articles; this doesn't just harm article-space, it also harms the new editors trying to contribute articles. We already have the Autopatrolled flag which is used for editors who have shown that they understand how to create new articles without serious issues; I think a similar benchmark might well be used on AfC. bobrayner (talk) 17:32, 1 September 2013 (UTC)
Support Plenty of new users heading straight to reviewing AfCs. We need something like this. buffbills7701 23:47, 1 September 2013 (UTC)
Support. First, reviewing AfC submissions judiciously requires a reasonable amount of knowledge and experience of content-building. Submissions should not be approved by just anyone – brand-new users least of all. Second, if I remember correctly, there has been more than one occasion in the past where groups of users have resorted to sockpuppetry or meatpuppetry to have their poorly-written drafts promoted and moved, carelessly, into the main article space. An "AfC reviewer" permission will bring with it some much-needed additional accountability. SuperMarioMan 15:07, 3 September 2013 (UTC)
Support Too many issues with just letting anyone play with the submissions. There's been several cases in the past where this has been troubling. Taylor Trescott - my talk + my edits 22:53, 3 September 2013 (UTC)
Support as a long standing project member and one of the main developers of the helper script. We get regular problems with new (very new) editors and thus unexperienced editors trying to review articles. Finding them (wrongly reviewed submissions) is not that easy as these submissions mostly get out of the backlog and the backlog is most of the time very very backlogged (over 1000 submissions waiting for a review). mabdul 20:29, 9 September 2013 (UTC)
Support - too many examples of new users or SPAs that decide to review their own articles, each others, etc., without meeting the criteria for new articles. GregJackPBoomer! 02:21, 12 September 2013 (UTC)
Support - I will, and always have, supported this. I was, however, going to oppose this and still will if it turns out that this is another userright other than reviewer, as that just turns AfC into that hat collectors club. --Nathan2055talk - contribs 21:25, 12 September 2013 (UTC)
Support. I'm a little hesitant, as there are an obscene amount of articles in the AfC bin at any given time. However at the same time I can't tell you how many people I've seen accept articles that are wildly inappropriate to add to the mainspace- some of whom are people who have been editing for a very long time and should know what does and doesn't give notability. I think that this would be a good idea, as it'd force some of these users to actually look at the quality of the sources rather than the quantity, as their reviewer position would be revoked or hampered by admitting too many obviously nn pages. I think that the harm done by introducing blatantly non-notable pages is probably more harmful than leaving a surplus of pages in the AfC bin. It takes minutes to accept a page, but 1+ weeks to delete it. This will need tweaking, but overall I support the idea of some sort of reviewer permission for AfC. Tokyogirl79(｡◕‿◕｡) 19:06, 4 October 2013 (UTC)
Another hat for the admin-wannabees to collect. That's going to work wonderfully at attracting those who want to be An Reviewer (Hey look Mom! I've got a Job!), rather than those capable of reviewing. If this must be done, please at least make it automatic on the basis of content, not just as another popularity contest. Andy Dingley (talk) 09:18, 24 August 2013 (UTC)
I fully understand that argument. However, a closer look at the proposal will show that this doesn't necessarily have to be a 'user right' per se. There are a lot of permissions that don't get added to the hat rack, such as AWB, Stiki, various clerks, and a few others. It could be as simple as setting a threshold for experience and asking potential reviewers to request permission from the AfC team to obtain access to the reviewing tools. In any case none of that is up for debate yet. Kudpung กุดผึ้ง (talk) 09:56, 24 August 2013 (UTC).
This !vote doesn't really make sense - the whole point of such a right/setting/permission/whatever existing, and the criteria for getting it, is to to exclude the wannabes and hat collectors. Roger (Dodger67) (talk) 19:54, 24 August 2013 (UTC)
Oppose it is more wikipedia elitism that stops people without a lot of time from contributing in useful way. MarioNovi (talk) 20:10, 24 August 2013 (UTC)
I'm not really sure you understand what you're !voting on here, to be honest. Or how "elitism" comes into this. Lukeno94(tell Luke off here) 20:31, 24 August 2013 (UTC)
Don't think permission should be needed. Yes? MarioNovi (talk) 21:16, 24 August 2013 (UTC)
Oppose as requiring users to jump through hoops to "apply to be a reviewer" is going to discourage existing, experienced users from reviewing... after all, if we needed more reviewers so badly, we wouldn't be putting obstacles in their paths. All I see this proposal accomplishing is to make an already 18-day-long backlog of articles awaiting review even longer. Setting a threshold for minimum time on Wikipedia or edit count (as the 'autoconfirmed' status does) is one thing, but conducting job interviews is something only viable when there is one post to fill and multiple applicants. K7L (talk) 02:01, 25 August 2013 (UTC)
How's it jumping through hoops? It could be like asking for AWB permission, where you just put your name and wait a few hours. Jackmcbarn (talk) 02:19, 25 August 2013 (UTC)
Same could be said for anything that anyone can do now, why not require approval for Twinkle for instance, inexperienced users of it cause no end of problems... WP:Pillars #3 shouldn't be disregarded lightly, and a requirement that someone ask for permission to edit in certain ways is certainly a hoop in that light, even if its an arguably justified one. Monty845 02:33, 25 August 2013 (UTC)
People submit AfC's for the sole purpose of having them reviewed. Edits aren't made for the sole purpose of being Twinkled. Jackmcbarn (talk) 03:24, 25 August 2013 (UTC)
I will never apply for any permission on WP or Commons (and wish I didn't have any that have been auto-granted) because they turn into an opportunity for some officious twat of a teenage admin to threaten their withdrawal. If I have to have a particular gracious permission to do some task, I'm just not going to do that task. Andy Dingley (talk) 10:31, 25 August 2013 (UTC)
I sympathize with Andy Dingley's position here. I feel the same sometimes, especially when changes start happening for which I can't find any consensus discussions. I've been on the butt end of some rather patronizing discussions. However, I would like to point out that, sadly, these same admins can block or ban you from any activity on WP, whether it requires a "right" or not. Making the Afc script available after a certain number of mainspace edits, or after creating one article and passing it successfully through the process, for example, would be less offensive than having to ask permission, especially since most of the current Afc reviewers are not admins, and by the very nature of being reviewers, can be (myself included) somewhat judgemental. —Anne Delong (talk) 13:05, 25 August 2013 (UTC)
I never requested either, AFAIR they were granted as part of a bunch of users when these rights were introduced. I don't particularly want either. I don't object to having the right (it's not the right itself that's the problem), but I won't let myself be bullied again by an admin with, "Do as I say or I'll take your toys away". Banning or desysoping is at least a bit more public, then less prone to abuse. Andy Dingley (talk) 16:36, 25 August 2013 (UTC)
Autopatrolled/Autoreviewer can still be granted without a request from the receiving Editor, as it is designed to reduce the workload of new article reviewers, and doesn't really have direct benefit to the editor who receives it. As for the other, during the roll-out of Pending Changes and the reviewer permission it was being granted without request, there was a good deal of controversy surrounding it, and now is only granted on request. I think over the last couple years we have mostly moved away from revoking rights without discussion, its certainly become less common. (Also, at least for the reviewer permission, if you want it removed a request to any admin should suffice, Autoreviewer is a bit trickier) Monty845 21:14, 25 August 2013 (UTC)
I understand some of the concerns but I would assume that "requiring users to jump through hoops to "apply to be a reviewer" is going to discourage existing, experienced users from reviewing..." is probably not what is being discussed here, and that experienced reviewers would be grandfathered in to any new permission - it's almost certain that they would would already have reached any required threshold. I believe some work is going on here to update the list and review the users' performance (some have been removed already) but some reviewing is possibly being done by editors who have not put their names to the project. Kudpung กุดผึ้ง (talk) 19:47, 26 August 2013 (UTC)
I would be inclined to support this, if there was some concrete procedure to support. It has a feel of the Australian republic debate of the late 90'a about it. I think that there is a consensus of sorts for some sort of level of competency for those who review submissions what that level is, how it is measured is ultimately going to make or break this. I would rather know what the process is before being asked to support it or not. LGAtalkedits 06:04, 25 August 2013 (UTC)
There are too many possible different ways of introducing a permission to review AfC submissions, so it is logical to ask the community if it wants to entertain the idea at all. The alternative would be to do a lot of hard work elaborating systems and their templatees and have seven or eight RfC running concurrently. That would be counter productive to gaining consensus either way. . Let's decide if the community wants the motorway first before investing heavily in the land and throwing residents off it - that's the kind of mistake governments make. That's why this RfC was expressly conceived to solicit the requested opinion prior to going ahead an developing a solution.Kudpung กุดผึ้ง (talk) 22:48, 25 August 2013 (UTC)
If it was clear how the community would be given that choice then I would consider switch to support. LGAtalkedits 11:42, 26 August 2013 (UTC)
LGA, this proposal is simply checking to see if there is enough consensus to see if it is worth asking the community how such a thing should exist. Based on your responses, I'm thinking that you're misunderstanding this RfC. The only two actions that can be a result of this is either it is dropped as an idea (a substantial number of opposes over supports) or a new RfC is spawned offering the suggested options (new user right, tying in with existing reviewer right, subpage like AWB does, or whatever other options there might be (I can think of many but am trying to avoid tl;dr and confusion). So, the question here in as broad of a term as I can think of is: "Do you support the idea of AfC drafts being required to be reviewed by people who have demonstrated competency?" Technical 13 (talk) 12:36, 26 August 2013 (UTC)
Oppose, hat-collecting and bureaucracy. Stifle (talk) 20:10, 25 August 2013 (UTC)
There are ways of doing this without it becoming another hat collecting exercise, Something needs to be done to control the quality of AfC reviewing and what this RfC asks first and foremost is whether the community wants some kind (any kind) of control in the form of a requestable permission to review submissions. There are many instances where permission has to be obtained before doing certain tasks but without it being added to the hat rack. The current system of AfC is one that is already too bureaucracy-laden because it has to detect and prevent substandard reviewing. Kudpung กุดผึ้ง (talk) 23:13, 25 August 2013 (UTC)
This isn't an RfC to create a new "hat", it's an RfC to determine if the community supports AfC drafts being required to be reviewed by qualified editors instead of anyone. Existing "hats" could be used, or no hat could be used (and base it solely on an edit count or some other technical information that can be pulled from api). Technical 13 (talk) 12:36, 26 August 2013 (UTC)
Oppose per WP:NOTBURO. The problem of bad reviewers is a very important problem. It's also a relatively infrequent problem. Let's deal with these on a case-by-case basis. Creating process here is wrong. Martijn Hoekstra (talk) 13:27, 26 August 2013 (UTC)
I think it is very common. Certainly, almost every AfC review I've seen has been flawed, either with minor issues, or with major ones (copyvios missed, obviously invalid articles accepted, and the inverse) Lukeno94(tell Luke off here) 13:35, 26 August 2013 (UTC)
That's a question of the overall quality from our reviewers. I concur that we don't always review well enough. I don't think any form of this proposal would fix that. What percentage of reviewers would you say wouldn't get the reviewer right? Do you think that those in the above section agree that they shouldn't have the reviewer right? How many of those reviewers have you discussed the issue with to try and fix it? Martijn Hoekstra (talk) 13:50, 26 August 2013 (UTC)
It all depends where you set the bar. Personally, that's not something I'm entirely sure about myself; you can be a brilliant writer of articles, but useless at helping others, or even analyzing anything other than your own work; or you can be the exact inverse. I've used AfC perhaps thrice in my entire time here (none of which were to put my own articles through the process); I simply don't trust the process, or the reviewers generally. There have been several high-profile cases recently of AfC reviewers being a long way below par, but I've also seen some people decline articles for invalid reasons first-hand, like minor stylistic quibbles, or even made-up reasons. Lukeno94(tell Luke off here) 14:44, 26 August 2013 (UTC)
Really, I see and understand your issue, but more process to 'become' a reviewer will not solve this issue, and might even work counter to it. Going in to the discussion what will is beyond the scope of this RfC (as pointed out at the top), so I'll leave it at that for now. I agree we have to do something, but the something we have to do here is become more pro-active in telling each other what we are doing right, and what we are doing wrong, rather than when a problem presents itself running for the red tape and hope that with sufficient taping, codifying, approval and rubber stamps the problem will go away. It won't. Martijn Hoekstra (talk) 15:37, 26 August 2013 (UTC)
It would seem like adding more requirements to being an AfC reviewer would, in fact, solve this issue if the reviewer right got enough thought put into it to ensure that only "good" reviewers would get the right. APerson (talk!) 16:46, 26 August 2013 (UTC)
If you would want it that high, it would be so hard to obtain, we can close the entire project. (That's obviously not a good thing, but that's related to other issues of AfC which we're not discussing here.) That's definitely not in the spirit of this proposal. Martijn Hoekstra (talk) 19:18, 26 August 2013 (UTC)
That's a good point; however, there are probably requirements that both are not too hard to obtain but still ensure that there is a basic threshold of experience for AfC reviewers. APerson (talk!) 17:47, 27 August 2013 (UTC)
Oppose Though I appreciate all the work done by the current volunteers involved in AfC, I still think the AfC project should be shut down. Adding a user right as a trick to keep the number of poor reviewers down will not help reduce the backlog or help the new-article-submitters to find their way on Wikipedia. Some people like to do things by the book, and some people like to fly by the seat of their pants. The AfC process is a very effective tool to help the first group find their way, but it drives away the second group by confronting them with acronyms and policy discussions. More user rights are not needed, but a viable workflow to steer new editors to like-minded Wikiprojects or in the case of their submissions, similar existing articles as examples to learn from without human intervention. Jane (talk) 16:53, 26 August 2013 (UTC)
AfC can't be shut down until/unless something is available to replace it. That 'viable workflow' project was first introduced as a feasibility well over a year ago and is now receiving renewed attention. However, such software-driven solutions, as with the New Page Curation for example, take a long time to develop and what many regular users and the affected AfC submitters have expressed is the possible need for a quicker remedy to its downsides. The Page Curation tool has helped cut down some of the NPP backlog, say from a monthly average of around 30,000 pages to currently 12417 unreviewed pages - oldest 1587 days (4 years) - and still in the encyclopedia, while a very significant number of AfC submissions are totally inappropriate (blatant spam, vandalism, attack, and test pages) and will never see mainspace as most of the WP:G13 deletion candidates in the backlog will demonstrate. The alphabet soup you refer to is indeed daunting for new users and hence one of the reason why experienced reviewers are needed - ones who take the trouble to spell out the blue links in plain text, or a set of templates that does it for them (which will possibly be integrated into a new 'draft' namespace for AfC submissions that is not part of this discussion). Kudpung กุดผึ้ง (talk) 20:21, 26 August 2013 (UTC)
I appreciate the need for more reviewers, but I just don't think this user right will solve that problem. Instead of throwing up yet another hurdle to potential reviewers, I would rather see some practical and tailored advice being pushed to submitters whenever a new article is rejected. I am thinking of some short practical advice (without any acronyms or links) linked to the type of subject of the rejected submission of the form: (1)Please do not attempt to create an article about a living person until you have successfully added 4 articles about dead people. If you can't think who to add, focus on famous people in your family's ancestral line, or famous people from your local community such as founding mayors of your town, or founding leaders in your church. (2)Please do not attempt to create an article about a place without the proper references. For proper references, look up the places near or similar to the place you want to create and see how those articles are referenced. (3)Please do not attempt to create an article about a going concern until you have successfully added 4 articles about defunct concerns. If you can't think what to add, focus on cultural heritage in your community, perhaps basing your research on the history of the subject you are attempting to add. I think the problem is that this solution of adding a user right will only create more backlog, not less. Jane (talk) 07:55, 27 August 2013 (UTC)
Your suggestions in italics (which Shinmawa below strongly opposes) are prcisely what this need for improving reviewing quality is all about. Even if it not the actual subject of this current RfC, the idea is currently being discussed elswhere and will be offered to the community for acceptance (or rejection). We have to take things by stages, and if one stage fails, we try something else - WP:ACTRIAL was a classic example of this and as a result we got the page Creation tool. It is possible to make things happen, but on Wikipedia, they can't be rushed - one thing at a time. Kudpung กุดผึ้ง (talk) 03:42, 28 August 2013 (UTC)
I agree things should be done in order of priority and step by step. My suggestion of halting the project should help reduce the backlog, which is growing logarithmically each day. Increasing user rights is not a priority, nor will it address the backlog. I am listed here: Wikipedia:New pages patrol/patrollers, but this is not because I consider myself a new page patroller, but because I do a lot of checking and editing in "long-tail" articles. There are lots of Wikipedians like me out there who will not request this user right, so this user right will probably cause the backlog to grow faster than it is already doing. Jane (talk) 10:27, 29 August 2013 (UTC)
If AfC were to be simply shut down completely as you suggest, the burden of controlling all new pages would fall on the New Page Patrollers. NPP suffers from similar problems of expertise and backlogs and there may even have to be an appeal to introduce some kind of user permission for that too. WP:ACTRIAL was a good faith attempt to stem the creation of inappropriate pages by restricting creations to WP:Autoconfirmed users, and by doing so, alleviate some of the problems at NPP. The community was strongly in favour of the proposition but it was not known at the time that it in fact clashed with a Foundation global policy. I am sure that if this RfC closes with a consensus in favour introducing some kind of control over the quality of reviewing, those experienced users who are already doing a monumental task would be grandfathered in, so backlogs are not likely to grow any faster than they already are. Kudpung กุดผึ้ง (talk) 11:56, 29 August 2013 (UTC)
Well I guess my problem is that at the time I was against WP:ACTRIAL for pretty much the same reasons. I see no reason to "to stem the creation of inappropriate pages", because earlier this year while browsing several rejected AfC pages on a particular subject, I discovered that most of them were pretty darned appropriate. I have since understood that what I was browsing was already trimmed of blatant vandalism, but isn't that what the anti-vandalism tools are for? I even cleaned up a rejected AfC and resubmitted it, only to get slapped with a rejection myself. That was the straw that broke the camel's back and is what got me started on my one-Wikipedian "Down with AfC" campaign. But don't worry, I don't carry much clout with "the Community", for what it's worth. Jane (talk) 13:32, 29 August 2013 (UTC)
In the last ten days or so I have deleted hundreds of WP:G13 AfC submissions. There is an absolutely staggering amount of totally inappropriate pages in there. The backlog for G13 is or was around 80,000 pages and if what I have seen is representative, then I'm afraid I have a very different picture from yours. If after a clean up an article you resubmitted was slapped down again, then there is all the more reason to clean up the quality of reviewing rather than shutting down the whole process; if it were closed down, who would otherwise review the pages? AFAIK, Rollbackers, PC reviewers, and other vandalism patrollers don't touch on newly created articles. Rest assured that we all have the same 'clout', it maybe just that some are more active than others in some areas and have time to do more research and are hence simply seen around more often. Kudpung กุดผึ้ง (talk) 05:28, 30 August 2013 (UTC)
First of all, thanks for your G13 cleanup work! I still howver think that you can better turn off AfC, leaving article creators temporarily to the chaos of the recent changes queue, until these issues have been addressed. By a sound cleanup and analysis of the backlog, better decisions can be made going forward. An extra user right is in my mind the wrong direction to go. Better tooling would be giving someone like me an easy way to tag an article in mainspace with one or two tags such as 1) meets minimum stub quality with issues or better; or 2) bad but subject is encyclopedia-worthy. Anything worth deleting remains ripe for the AfD process as is. I was surprised to realise that as a NPP'er, pages that I had edited were suddenly considered patrolled, while I may have noticed several issues and just added a category with hotcat. Digging in to your WP:G13 list I found a rejected submission that I believe deserved to be rejected: Wikipedia talk:Articles for creation/Mario Zampedroni. After being declined at AfC, it was simply added as a new article here: Mario Zampedroni. This stub is now available in 8 Wikipedia languages, all based on the exact same text of the declined submission. This case is an example of where the AfC process fails to stop determined people to put up vanity pages on themselves. For more information on this case, see Wikipedia:Administrators' noticeboard/Incidents/User:SqueakBox and paid editing (again). I think someone needs to run a bot to check whether your G13 list has been created in mainspace!! Note that this example just shows how easy it is to beat the review system altogether - yet another reason to abandon this convoluted construction. Jane (talk) 11:49, 30 August 2013 (UTC)
I don't do the clean up work, I just go through the cat and do the deletions (one mouse click) but I do read each one first and that's the whole reason for my concern and this RfC. I'm not quite sure I follow your arguments any more, the very things you are mentioning seem to me to be precisely every rationale for improving the quality of AfC reviewing rather than abolishing the whole thing, and AFAIK, Recent Changes is something quite different from patrolling new pages. But I think we ought to stop there, because this discussion is not going to influence the outcome of the RfC either way. Kudpung กุดผึ้ง (talk) 13:40, 30 August 2013 (UTC)
Well thanks for taking the time to make those deletions! I just spent some time trying to delete the one I found and it's really a pain in the xxx. I just don't see how another user right will help out here, other than reducing the number of eligible people willing to get their hands dirty and risk the after-effects of "revenge-editting". Jane (talk) 06:54, 31 August 2013 (UTC)
Although it's true that Afc can't keep all of the spam out of the encyclopedia, any that it does keep out is worth it in my opinion. It seems that once an article is in the encyclopedia, it takes a great deal of effort to get it out again, and sometimes articles which were tagged as having no references to show notability or as being promotional stay for years with the tags on them. —Anne Delong (talk) 23:46, 30 August 2013 (UTC)
Anne Delong, I agree wholeheartedly, but I still don't see the added value here of the extra user right, which is what this RfC is about. Jane (talk) 06:54, 31 August 2013 (UTC)
Yes, I was responding to your comments about the usefulness of Afc in general. However, a reviewer right, even one relatively easy to acquire, would prevent a lot of inappropriate articles from getting into the encyclopedia. For example, last month we had a user who submitted a bunch of articles, then created a sockpuppet to approve them, and to hide this went around whimsically promoting and declining articles. —Anne Delong (talk) 15:53, 31 August 2013 (UTC)
Oppose. In many ways, I think that AfC has lost sight of its mission and is getting bogged down in bureaucracy. When the ability to create new pages from an IP was removed, AfC was created to fill the gap for those people who, for whatever reason, either couldn't or wouldn't create an account. The inclusion criteria was very simple: If it wouldn't get speedied, it was good for inclusion. The job of "reviewers" was actually quite straight-forward and I even created the original AfC wizard to help filter out some of the more egregious articles that would never get past a SpeedyD. In all, it was supposed to be no big deal. Somewhere along the line, this mission was utterly lost and the bar on IP users submitting articles through AfC rose much, much higher than for that registered editors. This RFC's proposal is one more step in that direction, which I strongly oppose. AfC was never intended to get articles that were perfect from the get-go. It was intended to add stubs that editors (including the original AfC submitter) could then improve upon over time. However, despite a bold statement that the criteria is "at least a 50% chance of surviving an AfD nomination" (which is higher than the original criteria of "would survive an SpeedyD"), the practical reality is that there seems to be a demand that AfC only produce damn-near perfect articles on its very first edit, which never happens anywhere else in article space. Fact is, we don't NEED this role/bit/priv. What we need is to have a better understanding of just what AfC is for and a realistic expectation of what will come out of AfC. -- ShinmaWa(talk) 17:57, 27 August 2013 (UTC)
I think the issue is much more in the other direction; incompetent people putting off inexperienced/new editors (there is a distinction) from contributing to Wikipedia, with incorrect or imperfect declines. If an inappropriate article ends up in mainspace, it can be dealt with much better than an incorrectly declined article - and without being quite so off-putting to a new editor (although the deletion of one's first article is always going to be a negative experience) Lukeno94(tell Luke off here) 18:58, 27 August 2013 (UTC)
I understand, but that is exactly what I'm talking about, Luke. There are many editors, including experienced editors and even administrators, who are proponents of or actively rejecting article submissions for reasons that are usually reserved for AfD because they misunderstand the purpose of AfC and its intended low threshold of acceptance. Within this last week, an administrator who has participated in this very discussion was lamenting on the poor state of AfC because a stub-class article from AfC landed in AfD. Well, if the article managed to survive to get to AfD rather than getting Speedied, then as far as I'm concerned AfC has done its job. AfC reviewers are NOT one-man AfDs and they shouldn't be expected to be. They are simply conduits to create articles for those who otherwise can't. Their responsibility starts and ends there. In fact, by AfC's own stated guidelines, even if half the articles end up as deleted in AfD, then AfC has done its job. (The rule of thumb is "50% chance of surviving AfD") However, reading many of the support comments above indicates to me that many misunderstand that fact and are placing a much higher standard on the articles in AfC and on AfC reviewers than intended or needed.
Some examples: "[Reviewers] must be prepared to guide the writers...toward improvement needed to bring the article to the main space." This is not the purpose of AfC. AfC simply creates the articles written by those who can't on their own. Guidance is for the Teahouse, Help Desk, Coaching, and bazillion other places. AfC even points users to those resources to use (then, ironically, there are complaints when AfC submitters actually, you know, use those resources AfC points them to!!) Another: "There is no purpose in having an article reviewed by a user who will not produce an encyclopedic article." That's definitely not the purpose of AfC. Reviewers are conduits and minor gatekeepers. That's all. They aren't responsible for writing one word in an AfC submission. When **administrators** don't understand the purpose and criteria for AfC, how is a bit going to help? As I stated before, what we need instead is education on what AfC is and, more importantly, what is AfC is not. -- ShinmaWa(talk) 01:44, 28 August 2013 (UTC)
If every borderline new article were to be sent to AfC a lot more of the quality control systems would collapse. The sheer volume of articles being submitted to Wikipedia has caused, over time, the initiation of more controls over them. This is one of the reasons why the failed WP:ACTRIAL was proposed and why AfC has evolved into more than simply a superficial 50/50 decision, and includes providing some comments for the submitters (I read somewhere that the backlog is or was 80,000 WP:G13articles pages, and that is totally unacceptable; even the 'gate keepers' at WP:NPP have problems of quality reviewing and numbers of patrollers (although their backlog is now down to a 'mere' 12,000 or so from its traditional 30,000), so throwing all the control back to them would also not be a solution. We do indeed have a plethora of help desks and advice noticeboards, but all this is largely due to the fact that even after 12 years of existence, Wikipedia still does not have a proper landing page for new users. Until it does, we have to think of something else, and while most of the volunteers at AfC are highly motivated and do an excellent job, AfC is riddled with other serious problems that need solving. What we need instead is to educate new users on what Wikipedia is and is not for - one nice word from an experienced reviewer can help enormously towards editor retention (a major concern). Kudpung กุดผึ้ง (talk) 03:31, 28 August 2013 (UTC)
I'm sorry but I don't know what "If every borderline new article were to be sent to AfC a lot more of the quality control systems would collapse." means. That sentence makes absolutely no sense to me. Did you mean AfD? If so, no one suggested that so I don't understand this sentence in that context either. Also, no one suggested "throwing all the control back to them [I assume you meant NPP]", so that didn't make a lot of sense to me either. But you make an interesting point here. Reviewers at AfC and patrollers at NPP have very, very similar jobs with very similar criteria. NPPers aren't intended to be one-man AfDs and neither should AfC reviewers. Currently, there's no bit needed for NPP and don't think they need one. For the same reason, I don't think AfC needs one either.
Yes, I apologise, of course I meant AfD. That said, as one of the initiaors of reforms for NPP, I have resisted the creation of a user right for it. Page Curation was introduced by the Foundation as a measure to improve the quality, but apart from reducing the backlog and opening up the possibility to leave comments for the creators, it hasn't done much else. Kudpung กุดผึ้ง (talk) 05:15, 30 August 2013 (UTC)
Furthermore, I would disagree with Shinmawa's opinion on what AfC should be. It might have been what they described back in 2006, but it's not that now, nor should it be. Take a look at this comment : "I have no interest in contributing to Wiki now .... For the record, the page was accepted by a Wiki reviewer and then immediately tagged (is that the right term?) and a few weeks later has been put forward for deletion - this does not demonstrate consistency." That's what you get when you take a "CSD clearance is good enough" attitude to AfC. Regarding bureaucracy, believe me I hate it and I have no issue in smacking people down who try and create it, should know better, and don't listen. I want to keep the implementation of this as simple and straightforward as you can get - little more than a "can you prove you understand the reviewing instructions?" and the main driver for it is that no matter how clearly you state "You must understand Wikipedia policies and use this tool within these policies, or risk being blocked from editing", people just don't read instructions. The alternative is to block everyone who transgresses or haul them up to ANI for a topic ban, and that does create more bureaucracy, and high dramah. Ritchie333(talk)(cont) 08:36, 28 August 2013 (UTC)
I think we'll just have to stand in opposition on this point, Ritchie. As my first sentence in my opposition stated, I think that AfC has lost its mission and is getting transformed into "Yet Another Bleepin' Teahouse". We have too many of those already. In fact, having so many different overlapping and often competing "help centres" just confuses the hell out of new users. However, I digress. Back to the subject at hand, you pointed out that an article was created via AfC, but it could just as easily been created by user registering an account and creating it himself, it got tagged, AfDed... and then he didn't want to play any more. That's nothing new. That happens for AfC, registered users, and even experienced editors. However, that's not an excuse to lump a bunch of bureaucracy and other garbage onto AfC that really doesn't belong there. It is this scope-creep at AfC, which this RFC is just one symptom of, that I oppose.
In closing my oppose arguments, the intent of AfC according the WikiProject's opening paragraphs matches nearly word-for-word what I said its intent is in my original oppose statement. This other stuff that you propose it is and the underlying reasoning behind the reviewer bit, is exactly the kind of scope-creep I'm opposing. Finally, the guideline of 50% chance of passing AfD is not the guideline 6 years ago or even 3 years ago. That is the rule of thumb for reviewers today. Thank you for allowing me to state my opposition openly. -- ShinmaWa(talk) 22:50, 28 August 2013 (UTC)
OpposeShinmaWa (oppose #8) nailed it on the head. I also agree with Martijn that the number of issues are low enough that they can be dealt with on a case by case basis. Making it more difficult to review submissions is not what AfC needs right now. Legoktm (talk) 01:58, 16 September 2013 (UTC)
The number of issues may appear to be low, but only because they are not all made aware of to the regulars at AfC. Many reviewers don't list themselves, and most queries are addressed directly on the reviewers' talk pages. Kudpung กุดผึ้ง (talk) 08:00, 17 September 2013 (UTC)
Oppose per ShinmaWa (oppose #8). This is just needless bureacracy and obstacle to participation and erodes WP:5P #3 and #5. jni (talk) 05:59, 17 September 2013 (UTC)
Oppose per ShinmaWa and Andy Dingley. Another hat just to edit Wikipedia. I have promoted an article in the past, as it was one I was going to create, but an IP and new user had done so in AfC space. Will I apply for this hat if it is implemented?, I don't know. Will not having this hat stop me and others from doing the same in the future then yes. I occasionally look through AfC, and if there is an article that I am knowledgable about, I may promote it, however most AfC articles seem to be BLPs. All this extra hat will do is make it harder for experienced editors to make occasional promotions. Those that really want an article about their cute pet cat will still create articles in main space.Martin451 (talk) 21:35, 17 September 2013 (UTC)
Oppose, largely per ShinmaWa. I'm also not entirely sure why a userright is even what's being asked for here; userrights are for access to MediaWiki-restricted content. AfC isn't a MW component in any way, shape or form. Ironholds (talk) 17:27, 19 September 2013 (UTC)
Ironholds, you would need to read the RfC proposal to understand what is being asked. It's actually quite clear, and there is no suggestion AfC is MW component. The very essence is to avoid this becoming another trophy for the hat collectors, while at the same time assuring, by some form of permission, improved quality of reviewing. A bit like NPP really, but that wasn't a MW component either if taken in the same sense as you imply, but the Foundataion did develop the Page Curation system and if I remember rightly, you had something to do with that; it still did not address the core issues - i.e. numbers of active patrollers and quality of patrolling. Kudpung กุดผึ้ง (talk) 08:25, 29 September 2013 (UTC)
I'm going to gloss over most of your comment, since it seems to alternate between assuming bad faith (or, at least, incompetence), and talking about things totally unrelated to the actual subject. In either case, I maintain a practise of not replying to such things. The proposal calls for a permission/user right that is required to review AfC submissions. Both of these terms refer to MediaWiki-recognised flags that allow (or don't allow) access to a specific area or ability, automatically. Autopatrolled, for example, is a permission/user right that automatically evaluates which status to give pages created by the user who holds that flag. This only works because 'creating a new page' is something MW understands. The same is true of autoconfirmed, administrator, checkuser, etc. AfC is not something MW understands - it's just a collection of pages. If you and others are just asking for something based on the honour system, rather than something based on the software, you need to use different terminology. Ironholds (talk) 20:30, 29 September 2013 (UTC)
Oppose Every time I've heard of/from Articles for creation in the past year or so it's been AfC begging for people to come in and help them out. I fail to see how putting up a barrier to people helping out is a good idea considering their perennial need for more people helping out. Sven ManguardWha? 00:21, 27 September 2013 (UTC)
Sven, you're probably right about why you have heard of AfC every time - it's partly due to the good faith, but liberal invitations that may have attracted inexperienced users to the task that have led to the situation. No one is 'putting up a barrier' but this RfC is to suggest that a minimum of experience is required, and that introducing some form of control, whether formal or informal, over it would improve the reviewing. Kudpung กุดผึ้ง (talk) 08:09, 29 September 2013 (UTC)
Oppose per SinmaWa and bureaucracy. The real hat will be the admin gatekeeper hat. I had trouble getting autopatrolled while my articles were stuck in the patrolling queue forever. The admin at permissions did not care that my articles were encyclopedic, notable, referenced, well-written, reviewed by plant editors and being ignored by patrollers, all that mattered was I did not have 50 new articles. Some admin will make up a gatekeeper "school" process and require mentorship to force other editors to type policies in the guise of learning. Common sense will disappear from one more arena. I submitted an AFC a while ago as an IP. It was a plant article stub, with a taxobox and categories from another species in the genus; it was sourced and linked. It was rejected by an editor who admitted he or she knew nothing about plants, thought the species did not seem notable and wanted more references than two websites and a scholarly review article and wanted me to format the citations better. A plant editor just moved the article, but I left Wikipedia instead of doing what I do now, verify and reference plant articles. The new user interface makes it appear you have to use AfC, then shoves you to not very helpful places to ask help, losing good faith reviewers would make this problem worse. Oh, this week I found a plant article title that was spelled wrong seven years ago, giving, today, thousands of google hits from Wikipedia and its mirrors. I could have found it a year or two sooner if I had not been chased away from Wikipedia by bureaucracy. No. --(AfadsBad (talk) 20:07, 2 October 2013 (UTC))
I'm not quite sure how you had an issue getting autopatrolled because you had a few unpatrolled articles (I didn't); besides, that issue is totally irrelevant to this RfC on the AfC process, not on autopatrolled user rights (or NPP). Lukeno94(tell Luke off here) 21:06, 2 October 2013 (UTC)
I have no idea what your comment means, read the following sentence if you care to understand. That issue is relevant if later it is decided admins will be granting the right. However, feel free to just assume it isn't but everything else is. Matters not to me. --(AfadsBad (talk) 21:49, 2 October 2013 (UTC)) --(AfadsBad (talk) 21:49, 2 October 2013 (UTC))
If I have read your oppose statement correctly, it appears that you are ironically stating the very reasons why better control over the quality of both AfC and NPP are required. Furthermore, from my very regular work at WP:PERM over the past couple of years, I detect there a significant misunderstanding of the 'Autopatrolled' flag. IMHO, some users certainly consider it to be some kind of merit for having created a certain number of articles. The fact that some of the requests get declined will hardly impact on relieving the work of the patrollers. That said, your personal issues concerning autopatrolled are off-topic here - AFAICS, your request for the flag was approved on 29 August 2013. Kudpung กุดผึ้ง (talk) 01:02, 3 October 2013 (UTC)
Not off target at all. If it is made a right it could become a, "oh, I can withold and remove that right" bureaucratic issue. It is not only hat gatherers as far as I can tell, but hat witholders on Wikipedia. Right now, users can create their own articles after only 10 edits and 4 days, so the current standard is not that high, so I see no reason why it needs to be a granted bureaucratic permission. I will go ahead and express my opinion here based on what I see as issues with making this a permission, thank you. --(AfadsBad (talk) 03:14, 3 October 2013 (UTC))
Strongest possible oppose If any user can create an article, any user can review one at AfC. This is the encyclopedia that anyone can edit. This 'permission' does not contain any 'tools' like other permissions, it's simply an "I'm better than you" permission. We don't have vested contributors. Simple as that.--v/r - TP 19:28, 6 October 2013 (UTC)
So what do you propose we do about the incompetents at AfC, and those that attempt to WP:OWN it? Brush them under the rug and pretend they don't exist? Carry on letting them allow clear spam into the encyclopedia, whilst they simultaneously reject good articles and drive away their contributors? Lukeno94(tell Luke off here) 21:44, 6 October 2013 (UTC)
We do exactly what we do everywhere else on Wikipedia. AFC is not a 'special project' on Wikipedia that gets its own special rules.--v/r - TP 21:59, 6 October 2013 (UTC)
So we bury our heads in the sand and pretend there isn't any issue whatsoever. Gotcha. Lukeno94(tell Luke off here) 22:23, 6 October 2013 (UTC)
To avoid creating a special class of user and violating core principals and everything this encyclopedia stands for? Abso freakin lutely.--00:10, 7 October 2013 (UTC)
As Kudpung says, "none of [how this will work] is up for debate yet". Until we know how this will be implemented (at least broadly), it's impossible to say whether it will be a good or bad thing. I'd be happy, in principle, to support a well crafted and more complete proposal, just as I would oppose one which seemed to be about setting up a walled garden of reviewers. Warden's proposal, below, of using Autopatrolled, has merit and is worthy of further exploration. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:43, 24 August 2013 (UTC)
Broadly, the solutions how this could be implemented have been brought up on the previous RfC and while different, the possible solutions aren't really that many - except for doing nothing at all. The main thing as I see it here, is to see if the community would be amenable to discussing them if this RfC reaches a consensus for having any kind of control over the quality of reviewing by implementing a permission system of some kind - that's why this RfC doesn't propose any one particluar solution. Kudpung กุดผึ้ง (talk) 12:02, 24 August 2013 (UTC)
As a matter of principal I don't like the idea, but it does seek to address a real problem, and I don't see a better solution, so I guess I end up here. (Also haven't been active at AFC recently) Monty845 12:30, 24 August 2013 (UTC)
I don't like to add more complexity to what we already have, and this proposal does that. However there are enough review problems. Even experienced reviewers get their accepted articles nominated for deletion. On the bad review side I see article nominated for speedy deletion that should just be declined. Also AFCs put up at MFD just for being written in capital letters. Nastiness to contributors is a problem here, but we also don't need nastiness to reviewers. Graeme Bartlett (talk) 08:14, 25 August 2013 (UTC)
Due to my inactivity, it seems highly inappropriate that I would vote above. However, in my time as an historic AfC editor, I would agree that at the beginning of it all, I was highly unprepared for the reality of what the job required. I made sloppy, sloppy mistakes I still haven't forgiven myself for. While I think a Userright might be a viable solution to keep the rubes, who at some point counted me among their ranks, out, I am not sure the level of response fits the scope of the problem. Most editors who stick around for more than a few reviews learn the paces and fall in line with increasingly reasonable reviews, IMO. -TIM(Contact)/(Contribs) 08:01, 1 September 2013 (UTC)
Facepalm And there you have it, proof that my opinion really should not be weighed in this matter... -TIM(Contact)/(Contribs) 08:01, 1 September 2013 (UTC)
Most, but not all. Some serious problems have been outlined in several comments in the sections above and in the discussion below that can only be addressed through great vigilance.For example, expecting the reviewers to contro each other all the time would be just as impractical as patrolling the new page patrollers - although there are a couple of admins who are still carrying out random checks there (and finding enough discrepancies), but most of the errors only come to light there when an admin comes across a wrongly tagged CSD; other issues slip through the net. Kudpung กุดผึ้ง (talk) 08:31, 1 September 2013 (UTC)
(Longer threads and/or or comments about this RfC may be moved to the talk page)
We could use the existing Autopatrolled right which is "granted to experienced users who have demonstrated an understanding of Wikipedia policy ... and have already created a good number of articles." That is the level of clue we're wanting, right? If we use this then there's no need to wait for WMF to create a new right. Warden (talk) 10:00, 24 August 2013 (UTC)
Others have pointed out in earlier discussions (linked in the above intro') that autopatrolled may be too high a bar: most of the relatively few people who have qualified for it would be unlikely to have the time or interest to patrol AFC. In those earlier discussions, Wikipedia:Reviewer was proposed as more accessible yet containing (with the possible addition of some demonstrated competence at CSD/XFD to Reviewer criteria, or simply looked-for at AFC) most of the necessary skill set.
It would be sensible to start with the weaker right, as you suggest, and see how that goes. If there are still problems, then the higher level could be tried. Suck it and see. Warden (talk) 12:21, 24 August 2013 (UTC)
Its not exactly hard to use the templates manually, what about those who don't use the script? Monty845 12:34, 24 August 2013 (UTC)
Could always use an edit filter (e.g., "user w/o userright adds declined afc template," "user w/o userright accepts afc submission," etc). Theopolisme(talk) 13:09, 24 August 2013 (UTC)
It has been suggested that reviewing should require the tool. I hadn't even thought about the edit filter idea (which is a good one) but was thinking that there could be some kind of cascaded move protection on AfC Project space and if a reviewer had the script (and appropriate user right) then it could remove the protection and move it. As of last I knew, "most" of the "How to manually review a draft" instructions have been removed from the site pushing the tool. Technical 13 (talk) 13:18, 24 August 2013 (UTC)
I wouldn't worry too much about the technical details at this time until it's decided if and what kind of solution were to be adopted, it just detracts from the RfC proposal. There's going to be a proposal for a new kind of AfC page anyway which may change all that - let's not build the overpass until we know for sure the highway is going to be constructed. Kudpung กุดผึ้ง (talk) 14:16, 24 August 2013 (UTC)
I have suggested autopatrolled in the past, although as the criterion to use CSD Helper (and CSD Helper alone), which is where I see the bulk of the issue lying. Inexperienced users approving a few articles is easily manageable, an inexperienced user mass approving multiple articles can take hours of work to repair. I saw it as the most logical as it meant that you know what articles should look like, and your work didn't have to be checked. However if starting at a weaker permission is more palatable to the masses, sure. --kelapstick(bainuu) 11:53, 25 August 2013 (UTC)
The criteria for Autopatrolled are IMHO ridiculously high - even for Autopatrolled itself. For example, I have 6 years and over 30,000 edits here but on the article creation "score" I'm only about halfway to qualifying - it's easier to become an Admin than to get Autopatrolled. Creating a large numer of three-sentence and two-refs stubs is ridiculously easy - there are thousands of villages that still need articles and for them proof of existence is all that's needed to pass WP:N. Most of my article creations contain at least three or four L2-sections with several paragraphs, an image or two and occasionally even a table. They also have relevant Categories and WikiProjects. A simplistic article count is not a reliable measure of editing experience. Wikignome work experience - the processes that keep the entire WP running - are not even considered at all. Roger (Dodger67) (talk) 14:57, 25 August 2013 (UTC)
I wouldn't really mind having this, but I wouldn't want a heavy or even medium weight process. I oppose official stamps of approval that aren't really official at all, and hat collection. I'm unsure how to streamline this, but I'd be happy with anyone who makes a quick post on the talkpage of WP:AFC, and drop a note that they're planning to start reviewing. Maybe a note "before your first review, please leave a note on the wikiprojects talkpage, and allow for a day of discussion to see if there are any concerns". The problem we have with unqualified reviewers is fairly rare (once every two or three months), but fairly severe. We want to avoid this, but we don't want a lot of red tape. Martijn Hoekstra (talk) 10:57, 24 August 2013 (UTC)
The snag there is that some reviewers just jump right in without announcing their intentions, so no one knows anything is wrong until the submitters bring it to the project's attention. Some new reviewers enter their name on the list of active reviewers, but that list isn't actively patrolled - if it were, more anomalies would come to light, so at the moment, that list (which is probably out of date) doesn't serve much purpose. It's interesting to note however, that some very recent additions have been from reviewers who clearly didn't have sufficient experience, and were getting things quite wrong. Kudpung กุดผึ้ง (talk) 12:13, 24 August 2013 (UTC)
The trouble with a really informal process is that no one is sure about what to do. I'd at least like a WP:article with a list of minimums to point to (for example, 2 months editing, 1000 edits, 3 articles created; I'm just pulling these out if my head don't take them seriously) so that it wouldn't seem as if people were being accepted or refused on a whim. —Anne Delong (talk) 12:20, 24 August 2013 (UTC)
I like this, it could even be "automatically" granted, like Autoconfirmed is. All one needs is to be registered for X months, Y edits, Z new articles and no substantial recent blocks for relevant transgressions. Make Z reasonably low, I think 3-5 is sufficient, the quality of articles created is more important than the raw number. Someone who created just one comprehensive article and brought it to GA-class is a far more experienced editor than a "stub mill" editor who has spewed out dozens of stubs that just barely scrape past the minimums. Roger (Dodger67) (talk) 15:11, 25 August 2013 (UTC)
Well, maybe the reason we have a problem every two or three months is because we have a backlog drive every two or three months. Some of the problem might be solved by limited participation in the backlog drive to those who have already been reviewing for a specific amount of time or number oft reviews. —Anne Delong (talk) 12:20, 24 August 2013 (UTC)
The backlog drives don't necessarily cause the problem, they just make them really obvious. The recent case of the three or four editors who were giving "easy passes" to each others' submissions only became obvious due to the increased reviewing rate of the backlog drive - I think if there was no drive they might not have been detected. Roger (Dodger67) (talk) 14:57, 25 August 2013 (UTC)
Change name of proposal to AfC reviewer. The term "RFC" is understood in Wikipedia and in general usage to be about Requests for comments. Or, is it meant that only RFC reviewers would be able to close RFC discussions? I am not particularly for or against. --doncram 12:50, 24 August 2013 (UTC)
A valid point, but how to do it now that it's been published in a hundred different places ? On The VP and Cent the description is however unambiguous. --Kudpung กุดผึ้ง (talk) 13:33, 24 August 2013 (UTC)
AFCH developer comment: Just wanted to chime in here with a quick note: don't worry about the implementation. There are obviously numerous ways it could be done, some more complex than others, but there are active developers (mabdul, me, and occasionally others) who can help make it happen (whether it involves checking a userright, checking a subpage, or who knows what). Theopolisme(talk) 13:10, 24 August 2013 (UTC)
Another place where a "reviewer" right would be helpful would be at requested articles, if the current process is changed. The old process was simply that anyone could post a request, with "curation" simply being deletion if an editor felt that a request was inappropriate. What is now being discussed is to continue to allow anyone to post a request, but to then have that request either "accepted" or "declined". ("Accepted" doesn't mean that an article will be created, just that the request will be added to the official list of validated requests.) By limiting accept/decline actions to those with at least a minimal level of experience, abuses (e.g., tag team, sock puppets) are likely to be minimized.
I also note that with Flow, there is likely to be more opportunities to control processes based on editor rights. We could, for example, disallow nominations at WP:RFA unless the nominator (including self-nominators) had reached reviewer status. -- John Broughton(♫♫) 19:53, 24 August 2013 (UTC)
In my eyes, we should be generous in giving the bit, to just about anyone with a few hundred real article edits. It would just make it easier to stop those who lack the experience, or those who are socking/etc. and abusing the system. As long as the bar is low enough and it is the easiest bit to get, then there is no "hat collecting" value in it. Dennis Brown | 2¢ | WER 14:40, 25 August 2013 (UTC)
A few hundred edits is all? Don't you think they should have created at least one new article themselves? --MelanieN (talk) 22:44, 25 August 2013 (UTC)
Yes, that would be a minimum, not a guarantee. Like DGG below, we have to make it so the bit isn't a trophy to get or a disgrace to lose, it is mainly about stopping some people from reviewing articles who shouldn't be reviewing them yet. It should be "no big deal", and the only way to do so is to set a low threshold. That doesn't mean that anyone with 300 article edits is qualified or should get it, just saying there shouldn't be an unreasonable threshold. Even RFA doesn't have a threshold, so not sure we even need any "official" threshold. Dennis Brown | 2¢ | WER 01:33, 26 August 2013 (UTC)
While a few hundred edits is too low, I don't think that a requirement to have created a new (non-redirect) article is appropriate either. There are users, like myself, that have 10K+ edits and have created a total of about 350 new pages (and in my case - have worked on parts of the AFCH's coding, and theonesean's mentoring program) but have never created a single non-redirect article. What you would be saying is that despite my apparent level of "clue", I would not qualify. However, as Kudpung, myself, and others have mentioned multiple other times, setting limits would be a discussion for another RfC and this one needs to try and stay more on topic of should there be some kind of verification of "clue" to review AfC drafts. Technical 13 (talk) 11:38, 29 August 2013 (UTC)
I would find something like this awefull. For all I care people who aren't logged in to an account with 3 edits on their dynamic IP can review pages, as long as they are doing it well. Someone with a ton of semi-automated edits may have no idea what they're doing. I consider every case of editcountitis for this proposal harmful. The proper way about this is having a way to get a second set of eyes on reviews, and not being afraid to tell people that we really rather have they didn't do it on an individual basis, rather than imposing an arbitrary minimum edit count. Dealing with criticism (both giving and receiving it) is genuinely hard, and hiding behind a rule/userright/editcount is far easier, but it is a bad solution. Martijn Hoekstra (talk) 12:44, 26 August 2013 (UTC)
Martijn, how are people getting tons of semi-automated edits without having at least the 500 manual edits required to get AWB or running an un-authorized bot (which should be reported to admins and blocked indef until a BAG approval is given)? Also, how would an IP review a page when they can't move or create pages and can only create talk? Technical 13 (talk) 13:08, 26 August 2013 (UTC)
Technical 13, there are loads of ways to do the former (twinkle + browserextensions). For the latter, the important part is that I would be fine with them doing it, as long as they are doing it right. That it is currently technically impossible is a technical aspect. This proposal doesn't exist in a vacuum however, and changes to that situation aren't likely to be made anytime soon. The poinst is just as valid for any autoconfirmed account. Within the current technical restrictions, an editor without an account can still review pages. They can fail them all on their own, and they can accept pages through requested moves. A little cumbersome, but perfectly valid. Editors without accounts are every bit as much editors as editors with new accounts, with autoconfirmed accounts, or with accounts that have the admin or crat rights set. Dismissing their help from the AfC process that clearly needs as many capable hands on board as it can get on the basis of not having sufficient edits under the account is nonsensical. Like I said before, this is an important problem, but this isn't a very frequent problem. We don't need process, rules, userrights and whathaveyou to deal with one bad reviewer every one or two months. Martijn Hoekstra (talk) 13:23, 26 August 2013 (UTC)
I basically agree with Dennis about this., but we should also be fairly quick to act about withdrawing the bit until the user gains more experience.Just as the bit shouldn't be regarded as a trophy, so it's removal shouldn't be regarded as a disgrace, just as a temporary measure. It should be lower than autopatrolled, but a good deal higher than autoconfirmed,; I am not sure there will be an automatic standard, as it should require quality of of work, not just number of edits. If there is an automatic standard, I think it should start at a relatively high level, and that we be then successively lower it--it would not be a good idea to have to tell people they no longer qualified if they themselves had had never reviewed (and consequently had made no errors).
It has to apply to more than just using the AfC tool, because one have have the same effect by moving pages, and I have seen many promotional editors do just that--I would have it apply to any move from WT or User or User talk space to mainspace. (as well to requested articles) . DGG ( talk ) 21:53, 25 August 2013 (UTC)
All good points and ones which I, particularly, would like to join in a discussion about if this RfC reaches a consens demonstrating the community calls for the need of a requestable permission, Kudpung กุดผึ้ง (talk) 23:29, 25 August 2013 (UTC)
DGG, your point way above of having it granted automatically after a fixed number of edits (but removable if needed) is also interesting. We agree that the problem isn't most reviewers, just a few, so the main point isn't granting the bit, but being able to remove the bit. Dennis Brown | 2¢ | WER 12:10, 26 August 2013 (UTC)
I don't necessarily speak for all my colleagues at the WMF, but commenting with my technical hat on, I think it's a big problem to have the community potentially ask for a permission that is not clearly specified. I have no idea whether it will be even possible to fulfill the request from a technical perspective, if you don't say what it is, and some theoretically logical incarnations of a permission for reviewer are in fact not really possible to implement without investing developer time (as opposed to simply making a small configuration change). For instance: most of the substance of a review is actually done just by adding/editing templates to a page. You can in fact review pages without using the AfC tool. Locking down the ability to edit a namespace or a set of pages defined by a category is theoretically possible, but you immediately run into the problem of how to also let new authors edit that space, while gating reviews. Even easier to implement would be restricting page moves within a certain namespace, but this would only prevent poor articles from being moved in to mainspace, not actually prevent poor feedback by inexperienced folks. I fully sympathize with people who want to solve the problem of poor reviewing. It hurts the encyclopedia, hurts new editors, and wastes our most precious resource (volunteer time and energy) to have shoddy reviewing go on. But I'm not sure that, within the reasonable limits of what can be done with permissions, that adding a new user right will actually be a simple solution to the problem of how to do consistently high-quality reviewing at a large scale. Steven Walling (WMF) • talk 00:30, 26 August 2013 (UTC)
Could this be solvable with edit filters to prevent theamove from Wikipedia talk to articlespace, and to prevent the addition of Declined tags? Tazerdadog (talk) 02:48, 26 August 2013 (UTC)
Yes, though I don't think this is the best way to go about it. You could easily write regular expressions necessary to match AFC templates, but it's a weird use case. Edit filters were primarily designed to stop things vandals/spammers do, not selectively prevent use of templates in wide use by good faith editors. Steven Walling (WMF) • talk 03:31, 26 August 2013 (UTC)
Well, the good article process requires reviewers to be registered users, but there's nothing in the software that enforces this... --Rschen7754 03:57, 26 August 2013 (UTC)
I think that's partially just because it's too complex for most casual, anonymous contributors to figure out. And that most people who want to review GA submissions are registered. Here we've got a different story, right? Steven Walling (WMF) • talk 05:31, 26 August 2013 (UTC)
With that being said, another example is the OTRS member flag on Commons, whose sole purpose is to set off an edit filter for editors who are not in the group and who try and add OTRS tags. I believe that is tag only, however. --Rschen7754 05:53, 26 August 2013 (UTC)
┌────────────────────────────────────────────────────────────────────────────────────────────────────┘Steven, you seem to be concerned with the technical ability to enforce the permission, is that correct? What you're saying is even if a permission is required to access the AfC reviewer tool, that since the tool isn't needed to review (outside of my comments above how an IP can't create and you can't move until you have reached autoconfirmed), it is a waste of time to require said permission? If so, as part of another proposal I've been working on (New face of AfC RfC proposal draft), I've been developing in my spare time a new extension that would allow enforcement of such a thing. The extension could create a new Namespace (whatever the community agrees on, although I like "Concept:"), create a new userright (whatever the community agrees upon, I like "concept-reviewer"), create a new usergroup (whatever the community agrees upon, I like "visionary"), enforce a process that must be gone through to create a draft (new wizard that help users get sources and build a decent draft), etc... Once completed (I'll put the extension draft up on github once I make some serious headway, I've only just begun (I'll be using this extension on another mw based wiki regardless if it is supported here or not, so development is not a waste of time). Technical 13 (talk) 13:25, 26 August 2013 (UTC)
I'm really thrilled that the Foundation is chiming in here, and I'm also very grateful for the opportunity we had to discuss this issue in person recently with the Foundation staff. AfC is a process very much in need of attention (and I believe most people are agreed on that) because: It's difficult to use; It's being abused to a very worrying extent; the effect on article submitters is far to often contrary to the retention of desperately needed new contributors to the encyclopedia. However, before we even start to look at the technicalities - none of which can be posited here yet until the case for a 'draft' namespace has also been received by the community - I wouldn't worry too much about the technical details at this time, it just detracts from the current RfC proposal. A new new kind of AfC page may well be a greater deciding factors on what technical approaches may be required. Let's not build the overpass until we know for sure the highway is going to be constructed, and let's get the policy issue for or against some kind of permission over the hurdle first. Kudpung กุดผึ้ง (talk) 15:05, 26 August 2013 (UTC)
Re Steven Walling (WMF) - per our discussion at Wikimania 2013 here follows one possibility for implementing this permission - as you can see it simply restricts access to script as is done say using twinkle. Ofcourse with real support as promised by Brandon Harris in his "flow" talk we could do even better and support draft on other wikis too. However we need to fix one small problem at a time and not all the problems togetherBO | Talk 16:58, 26 August 2013 (UTC)
The AfC reviewing workflow as I currently understand it
At Wikimania 2013 I suggested incorporating the above into the new "flow" system while Brandon Harris supported this initiative it is unclear if and when this will take place. I have been been developing a course for AfC reviewers accreditation. The course format is similar to the adoption program only it has been implemented in Moodle course-ware on an experimental lab machine at http://moodle.wmflabs.org/moodle/ and once complete should be fully automated. My vision is that:
the Moodle course ware will issue an open badge type barnstar for signing up in the course and subsequent barnstars for completion of each subsequent module. However getting these barnstar will require passing a proficiency test at a certain score.
Barn-star #1 will unlock feature 1.
access to the "canonical AfC script".
access to a new HotCat feature for AfC.
Barn-star #2 will unlock feature 2 the ability to tag copyright violations.
Barn-star #3 will unlock feature 3 will unlock non BLP articles category.
Barn-star #4 will unlock feature 3 will unlock BLP type categories.
Barn-star #5 etc. are elective units for specialise in additional areas as needed.
The course features some policy but is primarily based on past reviews of AfC drafts. The idea being that after accreditation AfC reviews by different reviewers should converge. This should also be suppored by introducing greater transparency into AfC - by adding completion bars to the review templates and links to the "flow chart" above.
I agree that we need to fix one small (actually quite big) problem at a time and not all the problems together. This flow chart provides an excellent overview of the process. However, the award system may, like a formal user right, risk attracting the wrong kind of enthusiasm (see my earlier comments about MMORPG) and be adding new layers of bureaucracy (the original WP:CVUA was a classic example how good endeavours can get out of hand). Nevertheless, the discussions with Brandon and Eric Möller were most constructive and productive, but as I've said a couple of times, let's not preempt the outcome of this RfC (see its introduction) , and if the community concludes that some kind of control were necessary, then, with the promised help of the Foundation, we should be looking at a Community-WMF collaboration on the project and not necessarily a top-down package such as the excellent Page Curation tool which still did not solve the core issues that brought it into existence.
There are a lot of good ideas coming out of this but I feel strongly that resources should be pooled, for one thing, I very much like the idea that was suggested of an interactive check list - but more about that when this RfC is over and further talk on a new RfC for a new 'draft' namespace has taken place. While Article Creation Workflow and alternatives to the Wizard should now be a high priority, both that and AfC impact enormously on editor retention and there will always be a need for experienced review of created articles, and control over who can/could/should do the reviewing. There is such a backlog at AfC that the available experienced reviewer team is not always able to monitor the quaulity of all the reviewing as well, and the instances of its dysfunction have been uncovered more by coincidence than system and may only be the tip of the iceberg. Kudpung กุดผึ้ง (talk) 19:15, 26 August 2013 (UTC)
Further up I'm told, "Relax, this isn't a hat collecting exercise". Now we're already modelling designs for five graded hats. Andy Dingley (talk) 21:14, 26 August 2013 (UTC)
I understand your concerns but you are making a mistake calling the typical AfC recruit a hat collector. I see many highly motivated but dangerously inexperienced users in need of a guiding hand. I've had great results working with them in adoption and elsewhere. It is unrealistic to expect to find enough suitable reviewers because reviewing requires an intimate level of policy that you don't get by participate at AfD or even most parts of Admin notice board. My position is as Tennyson put it by soft degree subdue them to the useful an the good. What we need to do is find out hot to:
rapidly train wannabe reviewers,
maximise their useful output from the get go
give them work according to their current capabilities
teach them the real values of Wikipedians - the ones that are missing from the policy pages.
find uncooperative reviewers and COI agent before they do serious damage.
Finally you should remember that AfC exists to help newbies write articles not to keep out bad articles - we have gate keepers, wikilawyers AfC, CSD and PROD to address that concern. AfC can be a great place to transform hat collectors into model wikipedians and Admin material BO | Talk 22:25, 26 August 2013 (UTC)
Maybe your AfC reviewers are excellent editors, carefully reviewing these contributions without thought of kudos. My concern is that hat-collecting attracts other editors, who care more for the hat.
I would support the permission discussed here, per Kudpung et al, provided that it's kept low-key and probably best granted automatically, in a similar means to Autopatrolled. The permission isn't the problem, nor is the constraint on limiting AfC reviewers (a low-bar auto-granted permission will already have been granted to them). The problem is the risk (not inevitable) of making a new hat. Andy Dingley (talk) 12:57, 27 August 2013 (UTC)
Which is why my thoughts are that stopping people who are doing it wrong is more important than granting a right that can't possibly be granted in a fair and balanced way without making it an RfA like horror. This my suggesting is we let everybody have the 'reviewer' right, and we strip the 'reviewer' right by telling them to please stop doing what they are doing. The upside is that it won't need a user right, nor approval, nor technical changes, nor a process, just editors doing the right thing. Martijn Hoekstra (talk) 16:28, 27 August 2013 (UTC)
We are currently guiding beginners to start with the "quick fail" reviews (per the flowchart) and as they gain more experience to move onto the more nuanced content review decisions. If a reviewer is not sure, just leave it for someone else to do. The "quick fail" criteria are simple and easy to apply, the "content review" criteria require more knowlege and experience. With this I'm not saying we should have two "classes" of reviewers, but the workflow chart shows how even an inexperienced reviewer can do useful work by removing the bulk of the "low hanging fruit" from the pending pile so that more experienced reviewers can concentrate on the more difficult/nuanced reviews. As long as they are aware of their own limits newbies are welcome. Roger (Dodger67) (talk) 07:04, 27 August 2013 (UTC)
The recent case of the three or four editors who were giving "easy passes" to each others' submissions isn't the only concern, there are also the cases of users who have created socks to review their own submissions. AfC is a delicate area, not unlike NPP, where serious efforts need to be made to retain editors who are creating in good faith but who are not are of the the rules, and to dissuade editors from creating pages that are blatantly unsuitable and will never ever reach mainspace. Problem is that we can only guide beginners whom we know to be in need of guidance. Some ask for help and that's fair enough, but there are many who have hardly even edited the encyclopedia who jump right in and start reviewing AfC and patrolling new pages. If a submitter contests an AfC rejection, it is handled on the creator's talk page and as often as not we never get to hear about it. Apart from going manually through the 1,000s of reviews, there is no possible way of applying any metrics to the situation; what we do know is that AfC is not a place for raw beginners anymore than NPP is, or Rollbacker and PC reviewer are. Any admin who has worked regularly enough at WP:PERM knows that 90% of requests are denied - and those are rights that already have a low threshold - and it's difficult enough keeping the raw beginner wannabe admins away from unnecessarily 'clerking' the PERM pages too, so we also need to be aware of any bureaucracy becoming a meritocracy and considered a stepping stone to early adminship. Kudpung กุดผึ้ง (talk) 08:12, 27 August 2013 (UTC)
Hi Kudpung. The story about the three little piggies is sweet but not particularly relevant since those editors could have created thier articles directly in mainspace and by bypassing the review process are really only putting thier own work at risk or deletion by the first RCP who comes along. While I could develop a nice little sql script to catch this type of collusion and I might add it to a similar tool I'm working on. Anyhow I don't see how it actually break some policy. BO | Talk 14:46, 29 August 2013 (UTC)
Again, all good grist to the mill. There is of course NPP that catches articles that are immediately created in mainspce, but the system also suffers from the same problems as AfC. A couple of things are in the pipeline that could put an interesting new slant on th AfC process as a whole. Kudpung กุดผึ้ง (talk) 05:01, 30 August 2013 (UTC)
I understand that this is an attempt to gauge consensus before coming up with specific details, but I would hope that autopatrolled users would automatically be given whatever type of user right is involved. I agree that it would be an unfairly high bar for all AFC reviewers, but it would be useful for people who have proven their ability to understand notability and other policies to be able to review AFCs. Andrew327 13:00, 27 August 2013 (UTC)
From what I've seen, a userright with a higher level than autopatrolled is highly unlikely. While it might not automatically be given, it would probably be a one-line post to a board, saying "hey, I'm autopatrolled, and would like to help out. However, we are probably putting the cart before the horse now.Tazerdadog (talk) 15:10, 27 August 2013 (UTC)
I agree that it's too early to get into the technical details, but I figure it's a non-controversial suggestion. Andrew327 15:14, 27 August 2013 (UTC)
For what it is worth, 75% (165 out of 219 according to Dispenser's tool) of AfC Participants already have reviewer or higher. So, this is possibly less of a hat collection opportunity than some may believe. Technical 13 (talk) 23:36, 4 September 2013 (UTC)
Interesting, thanks for this. What was the percentage of users with rights who registered in 2013, and how many reviews wererthey responsible for? What does the 'active' column represent? What was the total percentage of reviews made in 2013 by users with rights against users with no rights? Kudpung กุดผึ้ง (talk) 04:24, 5 September 2013 (UTC)
Bear in mind that some editors choose not to list themselves as being part of a project even if they are active there. Monty845 04:29, 5 September 2013 (UTC)
That's right, and it's one of the main reasons why this RfC was launched. Kudpung กุดผึ้ง (talk) 07:53, 5 September 2013 (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.