Wikipedia talk:Reviewing pending changes/Historical/Poll on autopromotion

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.

---

This poll is now closed. The majority view (32 to 22) result was no autopromotion

The reviewer usergroup is described here. For general documentation on flagged protection and patrolled revisions, see this page.

Please note: The software will automatically promote users satisfying those criteria, but the rights can also be requested at WP:RFPE, and those criteria does not apply there, but rather administrator's discretion, with specific requirements possibly clarified by a guideline.

Choices

There are 3 choices here. You may number your supports by preference. If you oppose any autopromotion, please use the Oppose any autopromotion section.

Choice 1: Time since registration

Time since the user's account registration date.

1 day

1. Why delay? -- Taku (talk) 00:24, 18 April 2009 (UTC)

   Users need to be at least autoconfirmed, so 4 days / 10 edits is minimal. Cenarium (talk) 22:45, 19 April 2009 (UTC)

   Let's autoconfirm new users in 1 day then. -- Taku (talk) 10:44, 26 April 2009 (UTC)

4 days

1 week

1. 4 days is too little, but I think this is good enough to keep out most vandals and I suppose the flagging itself will be reviewed at some point, at least by the administrator removing the protection. —Admiral Norton ^(talk) 17:32, 12 April 2009 (UTC)
2. Seems fine, especially if we require a certain number of edits (25ish?). Vandals will either not make enough edits, or will be caught. I find the possibility of "sleeper" vandals unlikely, easily dealt with, and not able to be deterred by any reasonable wait period. --Falcorian ^(talk) 19:27, 12 April 2009 (UTC)

   I don't see how the possibility is 'unlikely', we already have tons. — neuro^{(talk)(review)} 23:29, 12 April 2009 (UTC)

   I would love to see the data. --Falcorian ^(talk) 18:00, 13 April 2009 (UTC)

3. Knowing that what you write is seen by the whole world immediately is an important motivating factor for editors. It is what makes Wikipedia successful. Let's not erase that factor for a small benefit. Really, imagine a guy would make good edits during one week and then suddenly start vandalizing pages: He would just get banned and reverted, not a big loss, and very few people would take the pain to do this. With this "one week" setting we eliminate 99% of the vandals, and we keep the motivating factor for 99% of the editors. With a "3 months" setting we eliminate 99.5% of the vandals, but we also eliminate the motivating factor for 50% of the editors. (figures are just my guess) Nicolas1981 (talk) 07:03, 16 April 2009 (UTC)

2 weeks

1 month

2 months

# I don't see why three months is really needed; support two as long enough that vandals should be caught but not too long that good users feel "left behind" for too long. –Drilnoth (T • C) 14:42, 12 April 2009 (UTC)

3 months

Support, vandals won't normally have enough patience for this but to commited editors it's nothing.-The Iceman (talk) 22:27, 9 April 2010 (UTC)

1. Long enough to gain some experience and too long for vandals to bother waiting. --Apoc2400 (talk) 09:39, 12 April 2009 (UTC)

   — neuro^{(talk)(review)} 13:37, 12 April 2009 (UTC)

2. This plus the edit count of something like 500 (or close) would be good enough to find the people that can be trusted with it. Chamal ^talk 14:25, 12 April 2009 (UTC)
3. Reasonable period. --Hans Adler (talk) 14:34, 12 April 2009 (UTC)
4. Shows if they will be an Active Editor or an Active Vandal. Exit2DOS2000^•T•C• 17:28, 12 April 2009 (UTC)
5. Plus 500 editcount, or 200 editcount non-automated. //roux   18:35, 12 April 2009 (UTC)

   I don't think the software can weed out automated edits. Stifle (talk) 09:05, 14 April 2009 (UTC)

6. Definitely not less than 1 month. The vandals have been learning as we put up more obstacles. This and 200 edits should weed them out, or at least force them to contribute positively for a longer period before switching. Trevor MacInnis (Contribs) 00:17, 13 April 2009 (UTC) Changed my mind. No autoconfirms. - Trevor MacInnis (Contribs) 22:02, 16 April 2009 (UTC)
7. Support fahadsadah (talk,contribs) 18:23, 13 April 2009 (UTC)
8. Just right. utcursch | talk 07:17, 14 April 2009 (UTC)

   While I oppose the concept of flagged revisions as adding too much maintenance overhead (and it's not like we don't have enough of that), if it is going to happen, this is the best selection. Stifle (talk) 09:05, 14 April 2009 (UTC) Changed my mind. Stifle (talk) 08:08, 17 April 2009 (UTC)

   Three months seems like a good distinction between users with experience and users without. The Earwig ^{(User | Talk | Contributions)} 16:23, 14 April 2009 (UTC) I change my mind: autopromotion is a bad idea. The Earwig ^{(User | Talk | Contributions)} 16:26, 14 April 2009 (UTC)

9. Sounds good as long as this is being done not by itself but also with a minimal number of edits. JoshuaZ (talk) 19:46, 15 April 2009 (UTC)
10. Seems reasonable to me.--- mav (talk) 01:57, 17 April 2009 (UTC)
11. Reasonable enough. Long enough to be a good editor, yet not too long, as they might as well be admins by then.--Unionhawk ^Talk 00:20, 18 April 2009 (UTC)

6 months

1. Standards should be high (but still less than that of becoming a sysop, which is normally around 1 year). It should be longer than those who operate on a Tor network, which is currently 3 months. MuZemike 19:52, 13 April 2009 (UTC)

   Excuse my ignorance, what is a Tor network and why is this relevant? AndrewRT(Talk) 21:54, 13 April 2009 (UTC)

   Tor is an anonymity network designed to bounces and relays your signals across various routers and networks throughout the world. It gives you more security by preventing others, by the method I just explained, from finding your physical location. (I would not be surprised that many serial sock puppets use this network to otherwise vandalize the enycyclopedia or otherwise engage in other disruptive editing.) Anyways, WP:UAL dictates that users who use this network require 90 days and 100 edits just to reach autoconfirmed status - far higher than the 4 days/10 edits for non-Tor users. For me, it would make sense that this be addressed, such as making such users wait far longer (i.e. 9 months or 1 year and 1000 edits) to be granted Reviewer status, just as we do with autoconfirmed status. MuZemike 02:34, 14 April 2009 (UTC)

2. 500 edits over 6 months ought to be enough to prevent gaming the system and allow the older accounts to help ponctually without having to formally request the right (a manual check of the accounts newly meeting this criteria could also be done). A procedure should be created to grant it earlier, though. -- lucasbfr ^talk 16:41, 16 April 2009 (UTC)
3. Editors should have had enough time to get a strong understanding of Wikipedia before they upgrade to the "Reviewers" class. SyG (talk) 18:28, 18 April 2009 (UTC)

9 months

1 year

1000 years

1. As an alternative to "never". ++Lar: t/c 11:39, 15 April 2009 (UTC)

Choice 2: Total edits

Total number of edits by the user.

10

25

1. 25 edits and 1 week seems to be what I've set in my mind. This should weed out any vandals, and let new users start editing in full quickly (and therefore hopefully continue to do so). --Falcorian ^(talk) 19:29, 12 April 2009 (UTC)
2. Knowing that what you write is seen by the whole world immediately is an important motivating factor for editors. It is what makes Wikipedia successful. Let's not erase that factor for a small benefit. Really, imagine a guy would make 25 good edits and then suddenly start vandalizing pages: He would just get banned and reverted, not a big loss, and very few people would take the pain to do this. With this "25 edits" setting we eliminate 99% of the vandals, and we keep the motivating factor for 99% of the editors. With a "200 edits" setting we eliminate 99.5% of the vandals, but we also eliminate the motivating factor for 50% of the editors. (figures are just my guess) Nicolas1981 (talk) 07:06, 16 April 2009 (UTC)

50

1. That's the number I could rack up in one month and it's a lot for a new user. One can't expect new users to huggle to prove their worth. —Admiral Norton ^(talk) 17:33, 12 April 2009 (UTC)

100

1. Together with 3 months. Just enough to prevent sleeper accounts. --Apoc2400 (talk) 09:40, 12 April 2009 (UTC)

   Agreed. It should take no more than 100 edits to see if a user is a vandal or a legitimate editor. Perhaps require 50 of them to be mainspace edits also, to ensure that they aren't just working on their user page? –Drilnoth (T • C) 14:42, 12 April 2009 (UTC)

2. If they're rouge and can work huggle, they can pile up hundreds of edits in a short time; that is a given. Hence, no limit on edit count is enough to stop them and we should discard our attempt to disrupt their concentrated edit-count attack on autopromotion and focus on letting the good guys in - 100 + 3 months would be my preferred choice. - Jarry1250 ^{(t, c)} 14:49, 12 April 2009 (UTC)
3. Combine with 3months fahadsadah (talk,contribs) 18:22, 13 April 2009 (UTC)

200

— neuro^{(talk)(review)} 13:37, 12 April 2009 (UTC)

1. Choice #1 if there is a way to count only non-automated edits within MW software, else 500. Either way in combination with three-month tenure. //roux   18:37, 12 April 2009 (UTC)
2. This one, as long as their non-automated. - Trevor MacInnis (Contribs) 00:18, 13 April 2009 (UTC) Changed my mind. No autoconfirms. - Trevor MacInnis (Contribs) 22:02, 16 April 2009 (UTC)
3. Sounds about right. Walkerma (talk) 04:11, 14 April 2009 (UTC)
4. Ok as long as it is with 3 months. JoshuaZ (talk) 19:46, 15 April 2009 (UTC)
5. II | (t - c) 16:24, 17 April 2009 (UTC)

300

500

1. Comment: WP:RFP requires 500 edits for someone to be given AutoWikiBrowser or NewPageWatcher functionality
2. Anyone can pile up a few hundred edits using something like huggle in a matter of days. But that wouldn't give him an understanding of the problem here. I think we should allow at least 500 edits to get enough experience (+3 months, as I have said there) so that the user has time to look around the place and understand the Policies and Guidelines properly. Chamal ^talk 14:22, 12 April 2009 (UTC)
3. I agree with Chamal N. --Hans Adler (talk) 14:37, 12 April 2009 (UTC)
4. 500 Article edits should be enough to show knowledge of Policies and Guidelines. Exit2DOS2000^•T•C• 17:30, 12 April 2009 (UTC)
5. Choice #2 if there is no way to separate out automated from non-automated. In combination with three-month tenure.//roux   18:37, 12 April 2009 (UTC)
6. Sounds fair. Combine with my proposal for 6 months, it shows to those "slow editors" or those editors who emphasize on quality of edits over quantity of edits that they are more than capable of reviewing edits. MuZemike 19:56, 13 April 2009 (UTC)
7. Seems a reasonable amount of editor experience to know WP policy and to keep trolls out. --Cybercobra (talk) 20:23, 13 April 2009 (UTC)
8. Seems just right. utcursch | talk 07:18, 14 April 2009 (UTC)
9. Will be enough to discourage vandals, but enough to encourage real editors to stay. Darrenhusted (talk) 14:20, 15 April 2009 (UTC)
10. 500 edits over 6 months ought to be enough to prevent gaming the system and allow the older accounts to help ponctually without having to formally request the right (a manual check of the accounts newly meeting this criteria could also be done). A procedure should be created to grant it earlier, though. -- lucasbfr ^talk 16:41, 16 April 2009 (UTC)
11. Sounds good if it is 500 non-reverted edits. --mav (talk) 02:00, 17 April 2009 (UTC)

    I don't think we can do that. -- lucasbfr ^talk 08:23, 17 April 2009 (UTC)

12. Sounds good to me. 500 edits is a good standard.--Unionhawk ^Talk 00:20, 18 April 2009 (UTC)
13. 500 edits is not that much if you are a newbie that needs to get accustomed to the inner behaviour of Wikipedia, and understanding its most basic guidelines. SyG (talk) 18:30, 18 April 2009 (UTC)

1000

1. I believe it really took me a year to fully understand and appreciate the various duties, issues and obligations that being an editor entails. We need to give people time to trip over the various discussions, policies, standards and practices... -- Mjquin_id (talk) 01:52, 18 April 2009 (UTC)

   500 'PAGES' edited, not edits...and only AFTER they received Rollback...(which is the first 500); but NOT counting RVV type edits with tools like Huggle (doable=yes) AND the editor in question must have NO RVV's on their record...Surely you would not want a 500 page vandal...!

   OR 25 new articles

   OR 500 "Reviewed pages"?

1500

2000

10,000,000

1. As an alternative to "never". ++Lar: t/c 11:40, 15 April 2009 (UTC)

Choice 3: Oppose any autopromotion

Add your !vote here if you oppose any autopromotion for the reviewer usergroup.

1. STRONG oppose Any automatic system can be gamed, and collecting the necessary minimum number of time/edits is always possible. As soon as one account has reviewer status, it can start approving edits from a vandal sock, and because it is reviewed, the chances of finding vandalism in poorly patrolled BLP's is even worse. I know, almost every system can be gamed, but a human check before giving someone this tool is just a good precaution, and once the initial round has been done, not that much work. -- Kim van der Linde ^{at venus} 14:52, 12 April 2009 (UTC)
2. Oppose I want human review of reviewers, like we did for Rollbackers, but I also want IAR to be firmly treated like a menace and plague when it comes to this: You need a minimum of 2-3 months, PLUS a human OK in a PUBLIC venue. No backdoor dealings with BLP, no "old users" sneaking in on new names. Let the sun shine in. rootology (C)(T) 15:12, 12 April 2009 (UTC)
3. Oppose The whole point of this system is to ensure quality in articles, and if some automatically promoted user begins to run amok, it'll a) reduce quality and b) reduce confidence in the system. Colds7ream (talk) 16:07, 12 April 2009 (UTC)
4. Oppose (switched from support). It seems that I'd missed this section earlier. –Drilnoth (T • C) 16:08, 12 April 2009 (UTC)
5. Oppose. I'd prefer even a cursory review of potential reviewers, for the sake of example I would say a month and a hundred edits should normally suffice. More importantly I'd like reviewers to ask for the permission, as this then means that they know what it means and what their responsibilities are. -- zzuuzz ^(talk) 16:09, 12 April 2009 (UTC)
6. Oppose I have to agree with some of the concerns raised above. I'm not too comfortable automatically giving a user this status. If it's used improperly, it's an easier way for vandals to bypass the page protection. An administrator review will allow a better degree of control to ensure that a user wouldn't just go off on a tangent. Icestorm815 • Talk 17:28, 12 April 2009 (UTC)
7. Oppose - Manual is the best way to promote but some users ie. admins should be given the right naturally. Once again sometime I get this confused with sighters, but manual should be the way to go. --Marcusmax(speak) 17:48, 12 April 2009 (UTC)

   If I'm not mistaken, I think this table says that admins will naturally have the right, similar to rollback. Icestorm815 • Talk 17:57, 12 April 2009 (UTC)

   Oh ya your right didn't see that before. --Marcusmax(speak) 18:01, 12 April 2009 (UTC)

8. Oppose. You guys invented FR – you believe it may have some benefit to it - excellent, do it yourselves, recruit dedicated volunteers, why would you need auto-recruitment of others? NVO (talk) 18:03, 12 April 2009 (UTC)
9. Comment: I favor auto-promotion of rollbackers and admins; and manual approval on request of others, with the same prereqs as for rollbackers. Pete Tillman (talk) 18:41, 12 April 2009 (UTC)
10. Oppose, This is very powerful stuff we got here. This should be manually given out. Should be given out as part of the rollback privileges with the entire role renamed to Reviewer, cause I agree about our administrative systems already being like a complicated layer cake. ViperSnake151  Talk  19:13, 12 April 2009 (UTC)
11. Oppose, manual approval is the safest route here. I imagine, that a user with 100 edits can be a reviewer, if their edits are mature - so manual evaluation is needed. feydey (talk) 21:32, 12 April 2009 (UTC)
12. Oppose - until I see how this performs in the wild, I want eyes on anyone seeking reviewer status. It works for rollback, it'll work for this. Xymmax ^{So let it be written} _{So let it be done} 00:06, 13 April 2009 (UTC)
13. Oppose, editors should demonstrate a superlative understanding of the biographies of living persons policy before they are granted a right which is restricted primarily for the purpose of stopping WP:BLP violations. This evaluation of reviewer candidates is necessarily outside of the competence of the autopromotion software. Erik9 (talk) 00:25, 13 April 2009 (UTC)
14. Oppose. Requests for permissions works fine. This can be given out like Rollback. Autopromotion is too risky, and would negate some of the benefits of the flagged revisions/protection. --Patar knight - ^chat/_{contributions} 00:43, 13 April 2009 (UTC)
15. Prefer to use WP:RFPERM like rollback, but with higher standards. -Atmoz (talk) 00:44, 13 April 2009 (UTC)
16. Oppose; RFP is better because they need to be vetted to look after RFPs. Sceptre ^(talk) 01:22, 13 April 2009 (UTC)
17. Oppose - just like rollback; the tool can easily be vandalised when an account reaches a certain time frame or has sufficient edits. MathCool10 ^{Sign here!} 02:22, 13 April 2009 (UTC)
18. Oppose This is more serious that rollbacker, not sure why this even a debate... BJ^Talk 09:22, 13 April 2009 (UTC)
19. Oppose - You guys are right, this needs a moderated system. — neuro^{(talk)(review)} 09:25, 13 April 2009 (UTC)
20. Oppose auto-promotion. Needs to be reviewed by a human, most likely an admin. A similar system to the rollbacker rights, or maybe combining this with rollbacker rights as suggested above, is necessary. Auto-confirm is not a good idea for reviewing flagged revisions. The Seeker 4 Talk 15:38, 13 April 2009 (UTC)
21. Oppose auto-promotion. I don't think it's a good idea to have it. We should give it out manually and/or give it it to rollbackers, who already revert vandalism, so they can see if an edit is vandalism or not and flag it as such. Techman224^Talk 21:45, 13 April 2009 (UTC)
22. Oppose auto-promotion. I suggest the system at WP:RFR with a rebuttable presumption that a user with, say, 200 edits and three months' history be approved in the absense of any signs to the contrary. Having a human intervention to approve the rights would seem sensible to prevent the most obvious abuse. AndrewRT(Talk) 22:06, 13 April 2009 (UTC)
23. Oppose Automatic system? Too easy to be abused. OhanaUnited^{Talk page} 02:40, 14 April 2009 (UTC)
24. Oppose. Beats the purpose. utcursch | talk 07:19, 14 April 2009 (UTC)
25. Oppose. I changed my mind: autopromotion is a bad idea. Essentially we would be handing out permissions to users that haven't been checked. What if the user has existed for a while and has vandalized? Just because someone has made a large number of edits and has been here for a long time doesn't mean that they can be trusted. We don't auto-promote our rollbackers. We don't auto-promote our admins. The Earwig ^{(User | Talk | Contributions)} 16:30, 14 April 2009 (UTC)
26. Oppose We have lots of editors here who have had tons of edits, and have still become problem editors. I don't see any reason why this should be any less restricted than rollback. These are pretty powerful rights after all. I would however support 500 edits/3 months as a minimum guideline for the manual approval of the right.--Terrillja _talk 16:49, 14 April 2009 (UTC)
27. Oppose any automatic system can and will be gamed. There isn't any problem with feasibility, as we managed to give out thousands of rollback flags without problems. Hut 8.5 19:54, 14 April 2009 (UTC)
28. Per everyone... autopromotion can be gamed. strong oppose any autopromotion scheme. Also, why is this taking so long to implement. Let's get on with it. ++Lar: t/c 11:34, 15 April 2009 (UTC)

    Oppose for the same reason. Brain checks are good. -- lucasbfr ^talk 12:47, 15 April 2009 (UTC)

    Striking in favor of a very high threshold, with a procedure to grant it manually. -- lucasbfr ^talk 16:41, 16 April 2009 (UTC)

29. Oppose, as per Pete Tillman, bundle with rollback. Tim Vickers (talk) 18:04, 15 April 2009 (UTC)

    Rollback is sometimes given when the user is not even autoconfirmed... And rollback is sometimes removed due to misuse in disputes. Thus we shouldn't bundle them, nor establish an inclusion between them. Although we could grant them together at the request of one or the other, at the admin's discretion. Cenarium (talk) 10:38, 16 April 2009 (UTC)

30. Oppose; a human hand in the loop is essential not only to provide a dose of sanity, but to bring a bit of compassion to the process. This is an increasingly huge project: it's too easy for newcomers to feel that they're all alone in the woods. A requirement to 'get in touch' with the Wikipedia community is an encouragement to get involved with the Wikipedia community, which helps to ensure that there is a Wikipedia community. I made a rather lengthy post on this somewhere in the first FlaggedRevs poll; I'll see if I can dig it out. Happy‑melon 17:45, 16 April 2009 (UTC)
31. Oppose auto promotions. There should be a user review, although the requirements should be kept low. Definitely lower than what the RfA has become. - Trevor MacInnis (Contribs) 22:02, 16 April 2009 (UTC)
32. Oppose auto-promotion; grant by admins on reasonable request. Stifle (talk) 08:08, 17 April 2009 (UTC)

Other criteria and notes

Criteria such as total content edits, unique content pages, recent content edits and whether to include or exclude deleted edits will be precisely determined after the first phase of the poll.

The previous poll on the conduct of a trial is here, and closed with 259 supports, 61 opposes and 4 neutrals.

Discussion

Letting this open for one week should be enough ? Then we could discuss the secondary criteria. Cenarium (talk) 13:52, 12 April 2009 (UTC)

One week is plenty. As has been said before, the long votes were only needed on the broader issues, whereas this technical/procedural nitty gritty doesn't need 1,000 users to all weigh in. rootology (C)(T) 16:24, 12 April 2009 (UTC)

• • Disagree. The exact details matter a lot. Too far in one direction and this becomes toothless. Too far in the other direction and we damage the wiki. 2 weeks seems good. JoshuaZ (talk) 21:58, 13 April 2009 (UTC)
• Dont bother with a 'reviewers noticeboard' for removal of rights ... go direct to WP:ANI, do not pass Go. Dont create another layer of administration (appointed, responsible and administered by whom?)... use what is already in place till such time as they decide that it is overwhelming them. Exit2DOS2000^•T•C• 17:40, 12 April 2009 (UTC)

  It's not for removal of rights, it's for any problem that would be specific to reviewing, for example contested flaggings, etc. Cenarium (talk) 10:56, 16 April 2009 (UTC)

Ok, now really not happy with the choice system. I'm either going to favor no autopromotion or favor an autopromotion that has both a time and an edit component. The way it is now written that's not an option. JoshuaZ (talk) 05:41, 14 April 2009 (UTC)

It can also check the spread of the edits, IP sharing, # of reverted edits, # of patrolled edits, and if the user was blocked. Aaron Schulz 01:03, 15 April 2009 (UTC)

Of course, it is, the criteria include both components, and can include others, as noted in #Other criteria and notes. Cenarium (talk) 10:56, 16 April 2009 (UTC)

Stupid brain check: if we set it up to autopromote, do we have a way to give it or revoke it manually for specific users? I'd support a high threshold (a few months and hundreds of edits) if we could do that but I fear we can't. -- lucasbfr ^talk 15:27, 15 April 2009 (UTC)

Yes, to both. Aaron Schulz 19:42, 15 April 2009 (UTC)

I thought it was straightforward and didn't mention it here, but it's in the intro of Wikipedia:Flagged protection and patrolled revisions, since the creation of the page. Cenarium (talk) 10:56, 16 April 2009 (UTC)

Should we auto-promote anyone who is an admin on another project? JoshuaZ (talk) 22:00, 13 April 2009 (UTC)

I don't think so. Such admins will not necessarily possess the minimum language and comprehension skills, or familiarity with local policy. Ask yourself if you would be happy being automatically assigned reviewer status on the Finnish Wikiversity (assuming you're not a Finnish Wikiversity editor). We can promote them more readily, but shouldn't do so automatically. -- zzuuzz ^(talk) 17:07, 16 April 2009 (UTC)

What about autopromoting users to a group 'established' with only the additional permission to confirm on a semi flag protected page (but not on pages with intermediary flag protection), see #Established usergroup intermediary between (auto)confirmed and reviewer. Autoconfirmed users are already auto-confirmed on semi flag protected pages: when a latest revision is confirmed, a new revision by an autoconfirmed user is automatically confirmed. So an established user could not do much more harm than any other autoconfirmed user, and we really need to identify users who would make good reviewers in order to limit the backlog. It would also allow to reduce the cases where an autoconfirmed user edits an unconfirmed latest revision and cannot confirm the new revision. Established users without reviewer rights could not access reviewing special pages or patrol. Since users would be autopromoted with a requirements of recent content edits, Special:Listusers/established would be progressively filled and admins could patrol the list and see whether it's appropriate to grant the user reviewer rights or remove established rights, or just let as is. Cenarium (talk) 11:36, 16 April 2009 (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.

Promotion criteria

The agreed time of a week has passed with a fairly good response and a fairly clear view that there should be no auto-promotion. Therefore I've closed the poll - hope you agree that was the right thing to do - with a conclusion of no autopromotion. After the trial is over I'm sure there will be an opportunity to re-open this debate should anyone want to.

Looking at the manual promotions via WP:RFPERM there is often a presumption that users that don't have a certain number of edits and/or a certain length of time editing should not be promoted. I suggest we should do another the same for the reviwer group - and I suggest a limited poll on what those limits should be - perhaps another week, say starting Friday 24th April? AndrewRT(Talk) 00:29, 20 April 2009 (UTC)

I'm fine with that.--Terrillja _talk 00:38, 20 April 2009 (UTC)

We should continue the discussion on the reviewer usergroup, yes, maybe a poll. And I'd also like to make a separate discussion-poll on an established usergroup. Cenarium (talk) 00:47, 20 April 2009 (UTC)

In addition to a minimal number of edits and account age to qualify as a reviewer candidate, we also need to decide who will have the ability set a reviewer flag. --mav (talk) 03:39, 21 April 2009 (UTC)

Does set a reviewer flag mean the same as promote an editor to the Reviewer usergroup? If so, admins have been discussed - are there any other options? AndrewRT(Talk) 23:10, 21 April 2009 (UTC)

Not really, there isn't any other group large enough to be able to handle the throughput of reviewers we need to keep this place running. Happy‑melon 09:20, 22 April 2009 (UTC)

Sure hope this doesn't get very far if it doesn't have autopromotion. Sounds like we've got crowd which really wants to turn Wikipedia into a cabal. Anyway, about 22 in favor of autopromotion and 31 against suggests to me that we should defer to the position most open, unbiased, and consistent with Wikipedia policy, which is autopromotion. II | (t - c) 23:50, 21 April 2009 (UTC)

41% is a clear consensus in support of autopromotion? I fear you will be in quite a minority to think that. Try and see how that statement looks from the other side of the fence: it's hard to escape the conclusion that you are ignoring a quite distinct consensus (in an approval poll at that, don't forget this is essentially "31 in favour of no autopromotion, 22 in favour of any other option") simply because you disagree with it. That's not a tenable position, so I hope it's not the case. However, I can't see any tenable argument to overturn the consensus above, so unless you can provide one, I expect we shall proceed as planned. Happy‑melon 09:20, 22 April 2009 (UTC)

I admit to a certain amount of flabbergastification at ImperfectlyInformed's suggestion to overturn consensus in this matter, and at the basis for it. In fact I'm significantly more than 41% flabbergasted. ++Lar: t/c 12:18, 22 April 2009 (UTC)

"suggests to me that we should defer to the position most open, unbiased, and consistent with Wikipedia policy, which is autopromotion.". That's your opinion; in my opinion having administrators look at each request is more in line with Wikipedia policy than autopromotion is. That was the purpose of the poll... to see what people's opinions on what system would be most inline with policy and just make the most sense in general. 41% shows that a number of people favor autopromotion, but nowhere near the number needed to have a clear consensus against it. Regardless, this is for the trial only... once the trial is over, people may or may not have different opinions on this point having seen it in practice. –Drilnoth (T • C • L) 14:08, 22 April 2009 (UTC)

This was strongly controversial to begin with, as ~39.9% voted against even the trial (I'm actually supportive of the idea). Was this promotion criteria poll advertised anywhere? Although I haven't checked how many of the editors commenting here against that vote, I suspect there could be a self-selection bias, especially since around 50 voted here and around 700 voted on the trial. It bothers me that everyone else could wake up one day and find that Wikipedia is no longer really the encyclopedia anyone can edit.

Anyway, a lot of people say "if it's auto then it could be abused". But I don't see anyone saying how it could be abused. Wikipedia polls are, in theory, "not voting", meaning that we provide reasons and debate. In fact, #3 on WP:VOTE is that "voters often expect that a majority or supermajority will automatically win the argument, or that the result will be binding — which is not the case". A vandal account usually won't survive 10 edits, much less 500 edits. Admins should not have to review the contributions of everyone who asks for something like this; that would become a bureaucratic nightmare. Assuming good faith and reducing bureaucratic overhead means we should default to letting editors edit until they've been shown not to deserve it by actual evidence. II | (t - c) 16:52, 22 April 2009 (UTC)

It doesn't seem to have "become a bureaucratic nightmare" for Rollback, IPBE, AWB or any of the other permissions given out by admins. If you seriously believe that your statemnt "a vandal account usually won't survive ten edits" is true, please search the blocklog for the summary "page move vandal" and check their edit contributions. There are innumerable ways for a large number of edits to be clocked up discretely and in an automated fashion without any inconvenience for a vandal. You are correct that polls on wiki are different to straight votes, and quality of argument is equally important. Your 'argument' that there is not consensus on this idea seems extremely weak. Now, can you provide an argument to convince the 60% of participants who argued against autopromotion that your statement "the position most open, unbiased, and consistent with Wikipedia policy... is autopromotion" does not merely represent your own personal opinion? To which you are entitled, but which others are equally entitled to disregard if it is not supported by evidence. Happy‑melon 17:47, 22 April 2009 (UTC)

The argument is that we assume good faith and avoid turning into Wikipedia into something it is not. Additionally, I imagine requests for review will dwarf rollback requests. What you can do with rollback (revert vandalism) can be done easily with regular tools or Twinkle, which is why I've never requested it. Let's try to avoid the usual Wikipedia habit of abusing the word consensus, which Wordnet defines as "agreement in the judgment or opinion reached by a group as a whole" [1]. Clearly there's a majority against autopromotion, but clearly it's not a consensus. I'll admit that I'm not terribly good with block logs; most of the vandalism I revert comes from IPs. How do I generate that summary log from the public logs page? II | (t - c) 18:10, 22 April 2009 (UTC)

Doesn't have to BE a consensus, it was an A/B choice. We chose A. "none of the above" wasn't a choice. See below. ++Lar: t/c 20:18, 22 April 2009 (UTC)

Seems to me the only argument against autopromotion raised was that it could be abused. Please let me know if I'm missing an argument. Happy-melon says that "there are innumerable ways for a large number of edits to be clocked up discretely and in an automated fashion without any inconvenience for a vandal". I wouldn't be surprised if Happy-melon knew more than myself about fighting vandals. I'd like to hear at least one of the ways that a vandal could rack up a large number of edits over a month or two without being noticed. Although I said below that I don't mind if manual promotion is used for the trial, I would like engagement on my questions here. II | (t - c) 21:09, 22 April 2009 (UTC)

Off the top of my head, methods I recall include adding and removing userboxes on their userpage, changing between British and American spellings, and converting between similar but distinct typographical variants such as 'curly' and 'straight' quotation marks. All of which are discrete enough not to attract attention, yet can be done entirely by an automated process. Although the programs I recall being used were very simple, I can imagine it would be trivial to obfuscate the script to add random edit delays, variant edit summaries, and other tricks to obscure the automated nature of the edits. Since any autopromotion criteria will be added to publically-viewable config files, such a program can also be tailored to the exact requirements of the software. It would not be difficult to do. Happy‑melon 21:58, 22 April 2009 (UTC)

Not having an autopromotion for reviewers will considerably increase the backlog of old confirmed pages, but there's a way to counter that. Since reviewers can both confirm and patrol, the latter being more sensitive (not mentioning confirming on intermediary flag protected pages to simplify), as I explain here, a usergroup intermediary between autoconfirmed and reviewer could alleviate the issue, while not giving out more sensitive rights automatically (autoconfirmed users being already auto-confirmed anyway). Cenarium (talk) 01:03, 26 April 2009 (UTC)

Poll results ?

I looked over the poll, seemed pretty clear cut to me that Autopromotion didn't have consensus, so I updated the page to reflect the results. ImperfectlyInformed reverted that update with the comment rvt the autopromote is rejected statement - isn't this a bit extreme and premature after a 60-40% breakdown from 50 editors in only a week of open voting?. Seems to me SOME change to the page is warranted, since it says "we will choose based on a poll" and we had a poll. So I tried a somewhat more caveated version. I invite discussion if that's not satisfactory but the page should reflect what's transpired so far (or should at least not speak in future tense of something that happened already). It has a small error which I will fix, I was confused about duration... sorry. ++Lar: t/c 19:31, 22 April 2009 (UTC)

I did not participate in the poll, but I agree that at this time there is a clear consensus against using any form of autopromotion. That consensus could change if the underlying facts change (e.g., we get through the trial period and there is subsequently a consensus to use reviewing on a "permanent" basis) or if a new and compelling argument arises. (ImperfectlyInformed's argument is not new, and the poll demonstrates pretty conclusively that it is not compelling either.) I had considered reverting ImperfectlyInformed myself, but that would have been sterile edit warring. GRBerry 19:58, 22 April 2009 (UTC)

My personal opinion is that we should not use autopromotion for the trial because of the poll here. After the trial, this topic can and should be revisited on a much larger scale. We need to remember that this is for a 2 month trial and things don't need to be set up in the trial the same way that they will in the full implementation, if there is consensus after the trial to fully use the system. –Drilnoth (T • C • L) 20:06, 22 April 2009 (UTC)

I'd agree (with both of you) that the trial and the final result after the trial may well be different. As you say, the point of the trial is to discover some things... then we take them on board and decide what to do next. To ImperfectlyInformed's objection: The poll, in my view, was (at the highest level) a choice between two alternatives rather than a choice among "alternative A but only if it has consensus", "alternative B but only if it has consensus" and "something else if neither has consensus". ... Something else wasn't one of the choices. B won the poll and A is out. End of story. Suggesting differently has an overtone (to me anyway) of using consensus, or the lack thereof, as a weapon to thwart progress. Sometimes you have to make A/B choices even if it didn't come out 90% A. ++Lar: t/c 20:16, 22 April 2009 (UTC)

This is the clearest explanation of the situation I can imagine; I agree totally. Happy‑melon 21:00, 22 April 2009 (UTC)

• The new statement looks fine. I'd caution against using the word consensus to describe a 60% majority in any situation, since a 60% majority is not consistent with the definition of consensus and implies that the opinions of the other 40% don't count. Anyway, I don't mind if it's used for the trial on the condition that the discussion on autopromotion is advertised more heavily and adequately discussed when flagged revisions become permanent (something which seems inevitable). II | (t - c) 21:09, 22 April 2009 (UTC) II | (t - c) 21:09, 22 April 2009 (UTC)

  The relevant poll for flp/pr is here and closed with 259 supports, 61 opposes and 4 neutrals. Cenarium (talk) 17:38, 23 April 2009 (UTC)

• What's the difference between that poll and this one, which had many more opposes as a percent? II | (t - c) 17:44, 23 April 2009 (UTC)

• I think that was for "standard" flagged revisions, and the more recent poll was for the flagged protection and patrolled revisions form of flaggedrevs, which functions drastically differently and just functions off of the same basic code. –Drilnoth (T • C • L) 17:55, 23 April 2009 (UTC)

Yes, I think we will have to concede that most do not want automatic promotion for the trial. --Apoc2400 (talk) 23:48, 22 April 2009 (UTC)

I'm also fine with admins granting reviewer flags per some criteria we set here for the trial. But I'd like us to revisit the idea of auto-promotion after we digest the results of the trial. --mav (talk) 00:22, 26 April 2009 (UTC)

There's clearly no way that autopromotion is going to get a consensus (that's clear to me and I'm in favor of autopromotion). The sooner we get flagged revs dealt with the better so I'm really not inclined to keep arguing over the autopromotion issue when it clearly isn't going to happen. JoshuaZ (talk) 01:14, 28 April 2009 (UTC)

I support autopromoting. Issues of socks and gaming the system I do not see as frequently occuring. Most vandals just add swear words to pages or write hi notes. These are the ones I wish flagged revisions would reduce as they are a drain on current editors.

We do not need great bureaucracy and will need lots of reviewers if this rolls out to the entire project. Also I think we need to WP: Assume Good Faith until good faith should no longer be assumed. We need to be open and welcoming and make joining and becoming a part of Wikipedia as easy as possible.Doc James (talk · contribs · email) 05:40, 29 May 2010 (UTC)