Wikipedia:Requests for comment/Wikipedia:Requests for adminship

Closing comments

I am closing this RFC. Any further discussion should be directed to Wikipedia talk:Requests for adminship.

The RFA process involves three groups of people: candidates for adminship, voters, and bureaucrats. Suggestions for reform have focused on each of these three groups. These suggestions can be summarized in the following chart:

Community group	Reform proposal	Comments
Candidates for adminship	Require reconfirmation and recall.	Voluntary systems are in place, but it will not become a requirement.
Voters	Change the format of discussion.	Some ideas have been tried and rejected, but new ideas should be considered.
Bureaucrats	Change the definition of consensus.	No consensus for change.

Recall and reconfirmation

Candidates for adminship can be reformed by changing the definition of adminship. Most people believe that it is too hard to become an administrator, so there needs to be an easier way. (I personally share this view.) Jimbo famously wrote that "adminship is no big deal," but perhaps it has become a big deal.

Consequently, Friday suggested that adminship should be easy to remove. Under the current system, desysopping can only be accomplished by ruling of ArbCom. The community has decided not to create a process outside of ArbCom to remove adminship (see the rejected proposal Wikipedia:Community enforced administrator recall). However, suggestions for optional recall are still being discussed. Category:Administrators open to recall has survived a deletion discussion. Wikipedia:Admin accountability alliance is another attempt to develop a workable recall system.

Many users, including TomStar81, suggested that admins begin with a "trial period," after which they will be reevaluated on their performance. Some have suggested that administrators be reconfirmed once every year. Some smaller wikis do require administrators to be reconfirmed periodically. For example, Wikisource requires that each admin be reconfirmed once a year, or whenever three users requests a "vote of confidence" (see the Wikisource administrators policy). However, there are too many administrators on English Wikipedia to evaluate each of them efficiently without creating unnecessary drama. Also, inactive administrators are not reconfirmed under such a review system, but they remain administrators indefinitely on English Wikipedia.

Another way to change the definition of adminship is to create an intermediate level of access between ordinary users and admins. For example, a person could be enabled to delete pages but not block users. This idea has been rejected. Wikipedia:Rollback for non-administrators has also been rejected (see Wikipedia:Requests for rollback privileges). [N.B. Rollback for non-admins became policy a few days after I wrote these comments. Shalom (Hello • Peace) 22:07, 6 March 2008 (UTC)]

Consensus

^demon's opening comment suggested that Wikipedia:Consensus contravenes the current RFA system, in which a candidate's success or failure can be predicted based on vote percentages. By established precedent, 75% support in RFA or 85% support in RFB is usually needed. Every request for bureaucratship includes a discussion about the proper definition of consensus as applied to RFA.

It has been suggested that consensus has nothing to do with vote percentages; and even if it does, there should not be any difference between RFA and RFB. Conversely, some users question the judgment of any bureaucrat who deviates significantly from the vote percentages. Ironically, there is no consensus about the meaning of "consensus."

For now, the voting benchmarks will remain in place. If anyone wishes to change them, make a separate proposal and talk to the bureaucrats. The bureaucrats are reasonable people, and they do what the community has appointed them to do. If the community changes its definition of consensus, the bureaucrats will cooperate.

Discussion format

Currently, the discussion at RFA consists of optional questions at the top of the page, followed by sections labeled "Discussion", "Support", "Oppose", and "Neutral." All of these sections are open simultaneously for seven days.

Several suggestions have been made for changing the format. Removing the "Support", "Oppose" and "Neutral" sections in favor of an open-ended discussion, like an XFD, has been tried at Wikipedia:Requests for adminship/Moralis. It is considered impractical because of the large volume of RFA comments. Some have suggested to remove the "Support" section so that only "Oppose" votes are allowed. The idea is to appoint administrators who have minimal community opposition. Experience has shown that candidates who are heavily opposed can still pass an RFA and become good admins, so I don't expect this proposal to gather any momentum.

TomStar81 and others have suggested to create a discussion period before the voting period begins. This idea did not receive much attention, but I believe it would solve a number of problems. First, a candidate should have the opportunity to answer questions before people oppose on the basis of those questions. Under the current system, someone writes "Oppose per [diff] etc." and many other people oppose on that basis before the candidate has a chance to respond, yet the "votes" still count. Conversely, a candidate may receive nearly unanimous support before a critical issue is raised, yet all of the "support" votes still count. A preliminary discussion period would give voters a chance to ask important questions, and the candidates could answer those questions, before the votes count.

There are other reasons to consider this proposal. Please see the proposal at Wikipedia:Requests for adminship/Proposal to add a discussion period before voting begins.

Conclusion

This RFC has not produced a clear consensus to require administrator reconfirmation and recall, or to change the definition of consensus, or to change the format of discussion. However, at least one proposal should be considered independently. Other proposals are always welcome at WT:RFA. Shalom (Hello • Peace) 18:10, 1 January 2008 (UTC)

---

Please note, that this was originally part of an MfD discussion, which was closed as speedy keep.

Wikipedia:Requests for adminship (edit | talk | history | links | watch | logs)

Statement of the dispute

Desired outcome

Rfa to be reformed as described below.

Applicable policies and guidelines

• WP:RFA
• WP:ADMIN
• WP:CONSENSUS

Description

This process has failed the community and needs to be shut down. We are no longer operating on a system of consensus, we are creating an environment in which people vote because of their own arbitrary standards, rather than judging the candidate on his own merits and abilities pertinent to adminship. Look at the vast majority of administrators. How many of them spend time writing content? Most administrators spend their time combing through the backlogs, trying to get slightly caught up. We’re busy mediating disputes, deleting, blocking and protecting. These are the administrative functions of Wikipedia necessary for it to continue to function, and require no article writing experience at all. Those people who would make great mop-wielders because they’re meta-involved, but not content-involved are shot down. This process is constantly called “Not a vote, it’s consensus,” but everyone involved knows it’s a percentage system. Those people who say it’s not a vote are the first to complain if someone with 69.99% approval is promoted. How can there be consensus by percentages? How can there be a 55% consensus? Consensus means a general agreement on something, NOT a percentage of votes. Other issues plaguing it are favoritism, elitism, personal attacks, assumptions of bad faith, and a general un-wiki-like attitude. We empower the bureaucrats to use their discretion, yet beat them up when their discretion isn’t how we agree. I propose to destroy the current RfA process. Instead, I suggest we allow a system in which candidates either nominate themselves or are nominated. If any user has opposition to their becoming an admin (keep the new system at a week-long process), they voice their concern. Provided there is no significant opposition, the closing Bureaucrat promotes in a week. I know many people say that MFD is not the place to discuss policy or to suggest shutting things down, but I have come to the conclusion that this cannot be discussed at the Village Pump, or at WT:RFA, or at any of the other places we’ve tried before. MFD garners attention and promotes discussion, and that’s what I’m trying to do, promote discussion, so please don’t shut this down prematurely.

Users that endorse this summary

1. ^demon^{[omg plz]} 16:20, 4 October 2007 (UTC) 16:20, 4 October 2007 (UTC)
2. ➪HiDrNick! 17:02, 4 October 2007 (UTC)
3. Kbdank71 17:28, 4 October 2007 (UTC)
4. --Mschel 18:05, 4 October 2007 (UTC)
5. - Philippe | Talk 23:13, 4 October 2007 (UTC)
6. Cheers,JetLover (Report a mistake) 03:46, 5 October 2007 (UTC)
7. Laleena 12:50, 5 October 2007 (UTC)
8. Bobsbasement Bobsbasement 15:55, 12 October 2007 (UTC)
9. K. Scott Bailey 23:43, 31 October 2007 (UTC)

Comment on this summary

If a few cases attract the kind of negative attention as described above, that only demonstrates that the systen is not perfect, not that is broken. A wiki is messy, because it is a wiki; it attracts best behaviors and awful ones ... What's new? It would be interesting to run some stats in say, the last 200 RfAs and see how many of these fall within the description on this summary. ≈ jossi ≈ (talk) 23:32, 6 October 2007 (UTC)

If there's no system to easily take out the tools, as suggested below by Friday, people will continue opposing for minor worries, because an eternal status that the community doesn't have the power to change is a big deal. a.z. 00:36, 7 October 2007 (UTC)

There is no such a thing as eternal adminship, as proven by the de-adminships that have been sanctioned by the ArbCom in the past. Do a user RfC is there are concerns about abuse of admin privileges, or post at WP:AN/I for discussion. ≈ jossi ≈ (talk) 01:37, 7 October 2007 (UTC)

The number of de-adminships has been far too low, and the process is far too arduous, especially since admins tend to support each other blindly against uppity users - challenging an admin can bring significant consequences. How many recent de-adminships haven't involved wheel-warring? Argyriou (talk) 20:23, 15 October 2007 (UTC)

"This process has failed the community and needs to be shut down." This is the starting point of this summary, and it isn't based on facts! I think it's quite surprising that so many esteemed editors discuss this here without anyone asking "where's the evidence?". From my experience in participating in RfA, I have the impression the process works reasonably well. We had many almost unanimous positive evaluations of longtime editors recently, as well as near unanimous rejections of candidates who lacked the necessary experience with rules and guidelines (one editor started in WP two weeks ago). We had candidates who came back for their second RfA and who were able to show that they took the advice from their first try to heart and that they had gained valuable experience since then. Of course, those were rewarded with the mop. Sadly, there were also some examples where the candidate's view of his work didn't really correlate with the view of the community. That's when the procedure is taking a toll on the candidate. But if a candidate can't stand this, how is he supposed to keep cool in heated debates? However, imho RfA does exactly what it's supposed to do: Give the tools to editors who are able to use them reasonably. And I think this shows in the numbers of successful noms. What do the critics want, that everybody gets a positive vote? Than everybody would self-nom, in sheer self-defense! No, the problem here is that there are not enough good candidates for all the work that needs to be done. Why don't critics spend more time recruiting experienced editors and move them to nomination? Now that would really be a huge help for getting more admins! Gray62 14:10, 14 October 2007 (UTC)

• Comment As I understand it, the proposal states that (a) Too many people are unfairly denied administrator status by a process requiring a ~70% supermajority vote, and (b) we should replace this process with one which promotes only if there are no significant objections. It seems to me that this would have one of two results. Either bureaucrats would have much greater discretion to decide whether they think objections significant or not and hence have much greater control over the process, or else we'd in effect be raising the approval requirement from a mere super-majority to near-unanimity and fewer people would become administrators. I am not convinced that appointing a dictator results in a uniformly fairer process than a democracy. I also don't believe the average Wikipedian participating in an RfA has the time to participate in a substantial dialogue of the sort that would change minds and generate concensus. A jury-room model, where there is a back-and-forth exchange of views that results in genuine changes of opinion, requires an intense, demanding, time-consuming process that cannot be used for every decision. Most people have time only to drop in and offer an existing opinion. Like it or not, this time availability situation more resembles what happens inside a balloting station than what happens inside a jury room. Best, --Shirahadasha 22:16, 15 October 2007 (UTC)

View by Ryan Postlethwaite

When I first saw this go up I though oh dear, but ^demon raises some very good points here. The whole process goes against what wikipedia is about. Talking in percentages, how can not promoting a candidate with 70% support follow consensus? How can the levels be different for RfA and RfB if we are supposed to be following consensus? The major problem with the whole process is that we can't ever discuss it. WT:RFA turns into a war field the minute a change is suggested and we never properly get to discuss new methods.. I'd say this is a reform comment, as I generally agree with the nomination statement, but perhaps not as strongly as ^demon.

Users that endorse this summary

1. Ryan Postlethwaite 16:29, 4 October 2007 (UTC)
2. User:Hiberniantears - Thanks to Ryan for moving this here. To begin, rather than letting this turn into a process where many well intentioned editors and admins begin talking past each other in an effort to describe their perfect vision of a process to grant the mop to editors, perhaps we should turn this into a series of mini-debates on the components of the existing RfA process. For example, an individual debate on the percentage issues, an individual debate on legitimate criteria, and individual debate on what consensus is, etc. I sense that many frustrations with the RfA process spring from the very fact that it is simply a very contradictory beast, that can leave many editors very confused as to just why they actually failed. Hiberniantears 17:00, 4 October 2007 (UTC)
3. nat^tang 17:12, 4 October 2007 (UTC)
4. J-ſtan^Talk_Contribs 17:20, 4 October 2007 (UTC)
5. Tbo ¹⁵⁷(talk) (review) 17:31, 4 October 2007 (UTC)
6. SamBC(talk) 18:08, 4 October 2007 (UTC)
7. --Rocksanddirt 18:24, 4 October 2007 (UTC)
8. Adrian M. H. 22:24, 4 October 2007 (UTC)
9. --Hirohisat ^紅葉 01:24, 5 October 2007 (UTC)
10. нмŵוτнτ 02:11, 5 October 2007 (UTC)
11. - Zeibura ^(Talk) 02:46, 5 October 2007 (UTC)
12. Lara❤Love 19:02, 5 October 2007 (UTC)
13. IvoShandor 05:07, 6 October 2007 (UTC)
14. Camaron1 | Chris 13:37, 6 October 2007 (UTC)
15. Pedro :  Chat  14:43, 6 October 2007 (UTC)
16. — Dorftrottel⁠, was: aldebaer 10:02, 7 October 2007 (UTC)
17. Phgao 15:48, 10 October 2007 (UTC)
18. I agree, both with concerns about consensus, and with the difficulties of trying to talk about it on WT:RFA. Discuss adjustment/modifications/changes, sure. Destroy? no. - jc37 00:27, 11 October 2007 (UTC)
19. Bobsbasement 15:56, 12 October 2007 (UTC)
20. Perfect Proposal ^{Speak out loud!} 22:39, 14 November 2007 (UTC)
21. Agree. Cirt (talk) 15:40, 12 December 2007 (UTC).
22. --Malleus Fatuorum (talk) 19:01, 12 December 2007 (UTC)

View by User:Deskana

(not an "outside" view since I'm so involved in the process)

Many people desire that RFA be changed, to improve Wikipedia. Change can be good, and it can be bad. What a fair chunk of the people that argue for the reform of RFA don't seem to appreciate is that it isn't "broken". That's not to say that it can't be improved. Broken would imply that it's not working at all, which it is. Last month (September 2007), 34 users were approved to become administrators. That's more than one a day. There's no way this process can be broken. However, it is entirely possible that there are potential improvements to be sought. RFA can be vulnerable to comments (both oppose and support in nature) that have little basis on the candidate's suitability to be an administrator. Ironing out such support votes would be difficult but ironing out opposes is potentially easier. I await the outcome of this RFC eagerly. --Deskana (talk) 17:19, 4 October 2007 (UTC)

Users that endorse this summary

1. --Deskana (talk) 17:19, 4 October 2007 (UTC)
2. J-ſtan^Talk_Contribs 17:22, 4 October 2007 (UTC)
3. ^demon^{[omg plz]} 17:29, 4 October 2007 (UTC)
4. SQL^{(Query Me!)} 17:30, 4 October 2007 (UTC)
5. Mtmelendez ^{(Talk|UB|Home)} 17:52, 4 October 2007 (UTC)
6. Chaz ^Beckett 17:54, 4 October 2007 (UTC)
7. Tbo ¹⁵⁷(talk) (review) 18:03, 4 October 2007 (UTC)
8. — Moe ε 18:18, 4 October 2007 (UTC)
9. Sam Blacketer 18:26, 4 October 2007 (UTC)
10. Fram 18:27, 4 October 2007 (UTC)
11. WjBscribe 18:30, 4 October 2007 (UTC)
12. RfA is the worst system in the world, except for all of the others which have been proposed. MastCell ^Talk 18:59, 4 October 2007 (UTC)
13. I agree with this. If RfA is broken, the page wouldn't be up and new admins would not be constantly created. The system is flawed just like every political system in the world (and yes it is political, as polis has to do with the will of the people). But like MastCell said above... Keegan^talk 19:18, 4 October 2007 (UTC)
14. Kusma (talk) 19:32, 4 October 2007 (UTC)
15. - auburnpilot talk 20:11, 4 October 2007 (UTC)
16. True. What MastCell said. - TwoOars (Rev) 20:44, 4 October 2007 (UTC)
17. People support and oppose for silly reasons. Support should be the default !vote with opposes only for good reasons--Phoenix 15 20:48, 4 October 2007 (UTC)
18. Lemonflash_{(O_o)} 20:52, 4 October 2007 (UTC)
19. There is always room for improvement. LessHeard vanU 21:05, 4 October 2007 (UTC)
20. Hit the nail on the head. —O (说 • 喝) 21:11, 04 October 2007 (GMT)
21. Wizardman 21:24, 4 October 2007 (UTC)
22. — xaosflux ^Talk 00:36, 5 October 2007 (UTC)
23. The RfA system is not broken. It makes mistakes. It can be chaotic. It is prone to clique-like behavior and irrationality. What else is new? Welcome to the wiki. But all in all, it does a decent job and all evidence suggests that it only promotes candidates which the community overwhelmingly trusts to handle admin duties responsibly. Pascal.Tesson 00:56, 5 October 2007 (UTC)
24. --Hirohisat ^紅葉 01:26, 5 October 2007 (UTC)
25. Agree. Dureo 01:46, 5 October 2007 (UTC)
26. RFA is not broken. --JayHenry 04:27, 5 October 2007 (UTC)
27. ➔ REDVEЯS was here 07:57, 5 October 2007 (UTC)
28. At least until I see evidence that RFA is broken, rather than assertions. Neil ム 09:27, 5 October 2007 (UTC)
29. Sjakkalle (Check!) 09:47, 5 October 2007 (UTC)
30. Maxim_{(talk) (contributions)} 12:31, 5 October 2007 (UTC)
31. Orderinchaos 13:17, 5 October 2007 (UTC)
32. Agree. The counter-argument seems to be "RFA isn't perfect so it needs to be deleted". — iridescent (talk to me!) 15:52, 5 October 2007 (UTC)
33. Sχeptomaniac^{χαιρετε} 16:07, 5 October 2007 (UTC)
34. -- _{Escape Artist Swyer} ^{Talk to me} _{Articles touched by my noodly appendage} 17:38, 5 October 2007 (UTC)
35. Homestarmy 18:54, 5 October 2007 (UTC)
36. Lara❤Love 19:03, 5 October 2007 (UTC)
37. Eye of the Mind 23:46, 5 October 2007 (UTC)
38. · AndonicO ^Talk 00:10, 6 October 2007 (UTC)
39. -- Anonymous Dissident^Talk 01:28, 6 October 2007 (UTC)
40. I agree with Deskana. — Thomas H. Larsen 02:07, 6 October 2007 (UTC)
41. I have always thought adminship could be improved, but I have never considered it broken, the current system itself does have good points. Camaron1 | Chris 13:26, 6 October 2007 (UTC)
42. Chris B • talk 15:32, 6 October 2007 (UTC)
43. Agree. There are voices that say that the RfA is broken, but there is no such evidence. We can improve the process as and if needed. ≈ jossi ≈ (talk) 22:57, 6 October 2007 (UTC)
44. Duh. Titoxd^{(?!? - cool stuff)} 02:30, 7 October 2007 (UTC)
45. --Quiddity 06:57, 8 October 2007 (UTC)
46. Carlosguitar 07:01, 8 October 2007 (UTC)
47. I agree. Mastcell's comment above says it all. --A. B. ^(talk) 00:13, 9 October 2007 (UTC)
48. agree. Rlevse 16:25, 9 October 2007 (UTC)
49. It isn't broken. It isn't perfect. It is not going to be perfect, so long as we allow humans to participate in the process. GRBerry 19:12, 9 October 2007 (UTC)
50. I've never seen a better solution proposed. Espresso Addict 19:13, 10 October 2007 (UTC)
51. - jc37 00:27, 11 October 2007 (UTC)
52. Xoloz 12:03, 12 October 2007 (UTC) MastCell is correct. The process has problems, but it has fewer problems than any yet conceived.
53. RfA is only broken to the extent that the community is broken. The process isn't in any sense overloaded, so if you want more people to get adminship, nominate them and support them. If people refuse to support editors who deserve the tools, that is a problem that can only be "fixed" by rejecting input from the community - a "solution" that no one would advocate. Dekimasuよ! 04:11, 14 October 2007 (UTC)
54. Rudget _{Contributions} 13:56, 24 October 2007 (UTC)
55. I echo GRBerry: Broken? No. Imperfect? Naturally. Room for improvement? Certainly. Mastcell's channeling of Winston Churchill is right on... — Scientizzle 15:21, 1 November 2007 (UTC)
56. It's not perfect, but this place will never be perfect. - KrakatoaKatie 22:36, 2 November 2007 (UTC)
57. There are many improvements that could be made, but the system as it stands does at least work. Deb 12:48, 6 November 2007 (UTC)
58. What GRBerry said. Angus McLellan (Talk) 13:43, 6 November 2007 (UTC)
59. A reform may not be needed, because RfA isn't entirely broken, but there are large cracks in it and it needs mending. We need to discuss this seriously. — Coren ^(talk) 21:32, 11 November 2007 (UTC)
60. Per Coren.Perfect Proposal ^{Speak out loud!} 22:40, 14 November 2007 (UTC)
61. Rusty, but it works. Master of Puppets ^{Care to share?} 04:30, 16 November 2007 (UTC)
62. This is a Secret ^account 18:40, 16 November 2007 (UTC)
63. Room for improvement =/= broken. - Kathryn NicDhàna ♫♦♫ 20:51, 20 November 2007 (UTC)
64. There is room for improvement, but it does work. Midorihana_{(talk)(contribs)} 00:02, 23 December 2007 (UTC)
65. Summarized perfectly. --D_arkF_alls ^talk 01:25, 31 December 2007 (UTC)
66. *standing ovation* —Kurykh 05:30, 31 December 2007 (UTC)

View by User:Warlordjohncarter

I actually was going to suggest that the MfD discussion remain open before it was closed in an edit conflict. I know that this point has probably been discussed to death, but is there any possibility between differentiating between the "maintenance" functions of an admin (speedy deletion, maybe editing some protected pages as per request on talk pages, maybe closing some discussions), and the "executive" fuctions, like blocking or banning and actual page protection itself? If it were possible to somehow "tier" the admin functions, I think that there would probably be more people selected to perform some of the maintenance functions than there are now, and possibly lessen the load per admin regarding such activities. Anyway, I do like the idea that the discussion is being allowed to continue.

Users that endorse this summary

1. John Carter 17:39, 4 October 2007 (UTC)
2. Adrian M. H. 22:26, 4 October 2007 (UTC)
3. -- _{Escape Artist Swyer} ^{Talk to me} _{Articles touched by my noodly appendage} 17:41, 5 October 2007 (UTC)
4. — Dorftrottel⁠, was: aldebaer 10:05, 7 October 2007 (UTC)

Comment on this summary

A two-tier system would be beneficial. If a limited range of "safer" tools are on offer, it could allow more streamlined reviews to take place. Admins with limited powers could deal with some of the more mundane aspects of mop work and spread the workload to allow the fully enabled admins to be less bogged down. Last month may have seen 34 new admins, but if admins are still overworked and stressed, as has been said elsewhere, then surely we need more of them to share the burden. Adrian M. H. 17:03, 5 October 2007 (UTC)

• • I do not see there are two tiers. I think the harm done by incorrect speedy deletions is a great as the harm done by incorrect blocking. As it stands, it's no harder to get a block reversed than an article restored. Probably admins should however be encouraged to specialize, but I do not see a reasonable way to institutionalize this. DGG (talk) 04:19, 7 October 2007 (UTC)

    Aboslutely, when I visit CSD redirect/merge candidates often "vanish" while I'm reading them. Rich Farmbrough, 11:25 15 October 2007 (GMT).

In the past, I suggested that the ability to block/unblock be assignable the way that checkuser is assigned. I think that that alone would be enough for what you suggest. - jc37 00:27, 11 October 2007 (UTC)

View by Friday

There's one specific sacred cow that needs slaughtered to fix this problem: We need a reasonably easy way to remove adminship from those who demonstrate poor usage of the tools. There's an unfortunately large amount of cultural inertia behind "adminship is forever" and "only arbcom can remove it, and only in extreme cases." This attitude is harmful. Almost the entire rest of Wikipedia works on a simple "What's easily done can be easily undone" basis, and this works surprisingly well. We already have crats who judge consensus; let them do the whole job instead of only half of it. Then, we can hand out the tools fairly liberally, knowing they can be removed when needed.

Clarification: (endorses before this time don't address this, of course) I don't necessarily want removing adminship to be too easy- hopefully it wouldn't happen often at all. Still, if people knew it could happen, it may drive home the point that all editors and particularly admins need to be responsive to feedback, even when that feedback is critical. Friday (talk) 23:46, 4 October 2007 (UTC)

Users that endorse this summary

1. We shouldn't liberally take adminship away, but a way for the community to remove adminship would be a step forward. Ryan Postlethwaite 18:06, 4 October 2007 (UTC)
2. Addhoc 18:14, 4 October 2007 (UTC)
3. --Mschel 18:15, 4 October 2007 (UTC)
4. ➪HiDrNick! 18:17, 4 October 2007 (UTC)
5. With Ryan's caveats above. SamBC(talk) 18:18, 4 October 2007 (UTC)
6. Tintin 18:19, 4 October 2007 (UTC)
7. — Moe ε 18:20, 4 October 2007 (UTC)
8. ...with reservations. Adminiship is a responsibility, not an award or recognition. It is provided by the community by means of consensus in an effort to improve Wikipedia; as such, it should also be taken away by the community, by consensus when it so chooses. My reservation, in line with Ryan's comment above, is how such consensus could be determined. Mtmelendez ^{(Talk|UB|Home)} 18:26, 4 October 2007 (UTC)
9. The problem is that once adminship is handed out, it can't be taken away without outright abuse by the admin and months of wikilawyering at arbcom. That's what makes people so reluctant to support a candidate. Please see the discussion here. Melsaran (talk) 18:31, 4 October 2007 (UTC)
10. — Edokter • Talk • 18:41, 4 October 2007 (UTC)
11. I'm sympathetic to this idea; after all, we sometimes see Dr. Jekyll in the RfA and Mr. Hyde afterwards, and more accountability to the community would be good. But is the current method of going through RfC/ArbCom really broken? MastCell ^Talk 19:01, 4 October 2007 (UTC)
12. Weak support of this. I'd support based on "Admiship is no big deal", but I've also seen vandal patrollers (and others) spammed mercilessly by those who oppose some admin action they performed. So yes, it should be at least somewhat easier than it seems to be now, but it shouldn't be "easy". - jc37 19:43, 4 October 2007 (UTC)
13. I'll repeat what I said at WT:RFA: "What shouldn't be presumed is that everyone who passes RfA will make a good admin. There should be a lot more "you are getting admin stuff wrong" self-policing within the admin community, and similar criticism from non-admins should be listened to as well." Maybe not a full-blown de-adminning process, but a lot more training after RfA and making it easier to review an admin's record. When vandals are given warnings, an admin can review contributions and warning templates. It should be easier for us all to review the record of different admins. That way it will become obvious which ones are not doing a good job. Carcharoth 19:52, 4 October 2007 (UTC)
14. Making adminship less serious is key to any reform. Nihiltres^(t.l) 20:12, 4 October 2007 (UTC)
15. Agree with this too. - TwoOars (Rev) 20:45, 4 October 2007 (UTC)
16. Agree, but (per Ryan) in moderation. High standards can be poisonous for unreasonable or narrow definitions of "standard". It is, in my opinion, best to lower standards for adminship to "can this person better Wikipedia as an admin?" (what the question should be anyway), and simplifying deadminship requests might be one way to accomplish this. Gracenotes^T § 20:49, 4 October 2007 (UTC)
17. Not easy, or even reasonably easy, but the ability for the community to decide upon desysopping is required. LessHeard vanU 21:07, 4 October 2007 (UTC)
18. The easier to take a tool away, the easier to give it out. Sasha Callahan 21:19, 4 October 2007 (UTC)
19. If adminship is truly no big deal, then perhaps removing it should be no big deal, too ... with the caveat that community de-adminship shouldn't be so easy our admins fear desysopping every time they use their buttons. szyslak 21:33, 4 October 2007 (UTC)
20. Weak Endorse, mirroring Ryan Postlethwaite's comments. I can just see the corrupt support votes "If he screws up, we'll just take it away". I think some have this idea that if we have an easier removal process, we should be more liberal about giving it out. While not perfect, RfA standards are pretty good. Maybe a start would be giving bureaucrats the ability to remove the mop. J-ſtan^Talk_Contribs 21:43, 4 October 2007 (UTC)
21. (e/c)Agreed. As it stands, it is near impossible to have the tools removed except in an absolute emergency or after persistent, documented, and serious abuse. As a result of this, RFA is (and with the current process for de-sysopping, almost should be) extremely intensive, so that anyone who might have any potential of abusing the tools won't get them because it will take so long to desysop if they do abuse the the tools. This super-critical process discourages many good candidates from applying. On the other hand, the process can't be too easy or otherwise-good admins will have the extra stress of knowing that a screw-up or 2 might lead to a lynch mob calling for a "Request for de-adminning." Mr.Z-man 21:44, 4 October 2007 (UTC)
22. Absoloutely, give the community the power to remove adminship without having to resort to the long legalese of arbcom. Viridae^Talk 22:00, 4 October 2007 (UTC)
23. Endorse, with Ryan's caveats. Expanded: I believe there should be a mechanism to remove adminship, but it should NOT be an easy mechanism. I'm not too opposed to the current ArbComm method, but I think perhaps they're the wrong body. Perhaps a sampling of current admins or other trusted users? - Philippe | Talk 23:09, 4 October 2007 (UTC)
24. Weak endorsement: while I agree that something easier than ArbCom should be found, I am concerned that any requests for de-adminship via the community would turn into a troll/grudge-fest. It's bad enough some people don't pass their RfAs because of trolls or grudge-opposers without perfectly good admins losing their adminship due to trolls or grudges. After all, RfAs for former admins attempting to be resysopped bring out the trolls and the grudge-opposers, and a community-based desysopping process would most likely do the same. Acalamari 23:14, 4 October 2007 (UTC)
25. I like this, as do I most proposals regarding RFA that aren't prefixed by "RFA IS BROKEN!!!". --Deskana (talk) 23:16, 4 October 2007 (UTC)
26. Carbon Monoxide 23:36, 4 October 2007 (UTC)
27. Cla68 00:38, 5 October 2007 (UTC)
28.  — Dihydrogen Monoxide (H₂O) 01:16, 5 October 2007 (UTC)
29. Well said. But the only part I'm worried is a the same thing that's going on at Rfa right now with the De-sysoping system; Drama. People who aren't really favoured or not well known might have disadvantage in de-sysopping system than others who have the popularity. This is a total violation of WP:AGF to me, but just to point out. --Hirohisat ^紅葉 01:23, 5 October 2007 (UTC)
30. this seems to be the major factor in voting for some, would help immensely I believe. Dureo 01:48, 5 October 2007 (UTC)
31. нмŵוτнτ 02:13, 5 October 2007 (UTC)
32. Absolutely - having the extra buttons is not a big deal. Mistakes can be undone, and continuous mistakes should be prevented by taking the buttons away through a less bureaucratic process than an ArbCom case that takes over a month for anything to happen. If the current RfA process is to stay as it is then this would be a great starting point for reforming the system. - Zeibura ^(Talk) 02:49, 5 October 2007 (UTC)
33. WODUP 05:25, 5 October 2007 (UTC)
34. maclean 07:34, 5 October 2007 (UTC)
35. If the tools could be taken away easily, they could be given out easily. A community-based desysopping process is exactly what we need - an "easy sysop, easy desysop" culture. Walton^One 09:34, 5 October 2007 (UTC)
36. -- lucasbfr ^talk 09:49, 5 October 2007 (UTC)
37. Strongly agree. In part it is such a damaging punishment at the moment that it would equate to being removed from the project. Being de-admined should be on the scale of getting a 3RR block, and shouldn't necessarily require a vote to be being restored to the fold. Spenny 12:52, 5 October 2007 (UTC)
38. Agree if a way anyone can find a way to do it — iridescent (talk to me!) 15:53, 5 October 2007 (UTC)
39. Lara❤Love 19:05, 5 October 2007 (UTC)
40. I whole-heartedly believe the bureuacrats should have more discretion when it comes to judging RfAs. I don't see what's the point of giving bureaucratship to someone if they are only going to press buttons, and not really use their judgment. Nishkid64 (talk) 21:19, 5 October 2007 (UTC)
41. Eye of the Mind 23:50, 5 October 2007 (UTC)
42. Mailer Diablo 05:09, 6 October 2007 (UTC)
43. Yes: promote many more admins; take away just a few more adminships; the net result is a win. (sdsds - talk) 05:38, 6 October 2007 (UTC)
44. The key to solving any problem is putting as much power as possible into the hands of the people, and taking it out of the hands of the bureaucracy. A stricter interpretation of WP:COI should sort out any problems that would come out of a potential troll fest. The Hybrid 10:11, 6 October 2007 (UTC)
45. Adminship is supposed to be "no big deal", so I support some kind of fair process where the community could remove adminship. ArbCom does a good job - but that is de facto only for serious abuse of the tools. Camaron1 | Chris 13:33, 6 October 2007 (UTC)
46. Partial support. De-adminship could be accomplished by a consensus vote of no confidence from fellow admins, but I would argue that ArbCom has and can continue to arbiter these cases. ≈ jossi ≈ (talk) 23:00, 6 October 2007 (UTC)
47. I think anyone at any time should be able to start a "request for confirmation" to confirm the admin status of someone. If the request is nonsensical, then confirmation will have a lot of supports, and this won't be something to worry about. I wrote an essay saying there should be an easy system for taking the tools away. a.z. 23:44, 6 October 2007 (UTC)
48. Any reasonable admin who's clearly screwing up, faced with a reasonable request from reasonable people to step down, will already do so (even if he's not in the "optional" hey-trolls-please-desysop-me category). —Cryptic 01:56, 7 October 2007 (UTC)
49. I've said this before, but I'll say it again: the perceived elitism in RFA is mostly a counter-reaction to the difficulty in removing the sysop bit from the so-called "bad eggs". Titoxd^{(?!? - cool stuff)} 02:32, 7 October 2007 (UTC)
50. — Dorftrottel⁠, was: aldebaer 10:05, 7 October 2007 (UTC)
51. Carlosguitar 06:51, 8 October 2007 (UTC)
52. Jeeny ^(talk) 23:30, 8 October 2007 (UTC)
53. I agree as long as it's not too easy. As to concerns about a process like this causing Wikidrama -- it will, which is why it should be reserved for admins that might cause still more Wikidrama while in office.--A. B. ^(talk) 00:24, 9 October 2007 (UTC)
54. Support, as this is in the spirit of Wikipedia. I do not see why it is so hard to remove adminship; everyone can revert edits and this does not lead to chaos. -- Casmith_789 (talk) 15:05, 9 October 2007 (UTC)
55. agreeRlevse 16:27, 9 October 2007 (UTC)
56. Support the principle. Davewild 17:13, 9 October 2007 (UTC)
57. Support in principle, but the problem that many administrative actions are unpopular means the details of such a system would be the tricky bit. Espresso Addict 19:18, 10 October 2007 (UTC)
58. Xoloz 12:01, 12 October 2007 (UTC)
59. The current system is akin to the House of Lords. --Nydas^(Talk) 18:18, 12 October 2007 (UTC)
60. Totally agree with Friday. Along the same lines #View by Irpen below. --Irpen 23:24, 13 October 2007 (UTC)
61. I agree. A more streamlined process to remove 'bad' admins would considerably lower the hurdles towards becoming admin. It's my opinion, too, that many editors are overly critical in the RfA process, because they fear of becoming stuck with an abusive admin that is very difficult to remove in the ArbCon procedure. That's only self-defense. It seems to me that ArbCon decisions often don't meet the expectations of the broad majority of editors. This begs the question, why such a complicated 'trial' at all? If adminship is "no big deal", removing the tools shouldn't be a "big deal, either. So, I'm totally with Friday in demanding that the process for removing admins has to change. Why not simply copy the RfA procedure for a new RfR (Request for Revocation) process? Gray62 13:40, 14 October 2007 (UTC)
62. Absolutely. shoy 19:06, 15 October 2007 (UTC)
63. 'Zactly. The argument that always comes up against this proposal is "vandals and trolls will deadmin users who opposed them", but do we have no means of discounting such votes? There are plenty of non-admin users who do vandalism patrol and have no trouble gaining admin status subsequently. Respectable editors should have every right to say "If I had known they'd act this way, I'd never have given them the tools." The community giveth, but can never taketh away is the core problem with our adminship process.--Father Goose 20:44, 16 October 2007 (UTC)
64. Agree. And strongly endorse Friday's caveats). --Bfigura ^(talk) 17:53, 23 October 2007 (UTC)
65. I agree that the community should get a say on whether an admin keeps their tools or not. But I don't think desysopping should occur frequently and it shouldn't be based on factors such as the popularity of an admin. Tbo ¹⁵⁷(talk) 17:40, 29 October 2007 (UTC)
66. Per AB. Adminship should be able to be commented on by the community. But per Tbo, the process desysopping is not the only answer.
67. — xDanielx ^T/_C\^R 07:00, 23 November 2007 (UTC)
68. Agree. Cirt (talk) 15:42, 12 December 2007 (UTC).
69. It would have to be 75% of the community (or more), and would certainly have to be a tough thing to do. Malinaccier (talk) 00:04, 15 December 2007 (UTC)
70. If becoming an admin is "no big deal", then the removal of the tools shouldn't be a big deal either. Guest9999 (talk) 06:12, 2 March 2008 (UTC)

Comment on this summary

• I can't imagine an "easy" deadminning process ever being anything other than a venue for grudges to be aired and a bunch of Wiki-lynch mobs. Admins make enemies - they are forced to make judgement calls based on (hopefully) what is best for the encyclopaedia. The busiest (and thus most important to the well-being of the Wiki) admins make a lot of enemies, who would love the opportunity to pillory someone who had dared revert them / block them / protect a page they were edit-warring on / warned them / deleted a page they had worked on. It should be difficult to take away admin rights - otherwise those who make tough calls receive no protection for doing so, and we will have a situation where people shy away from making such calls. Admins who are truly incompetent or malicious end up having their tools removed at the moment - by all means, let's establish a straightforward and defined process to do so, but it shouldn't be easy. Neil ム 09:35, 5 October 2007 (UTC)
• I tend to agree with Neil on this one. Perhaps more contentious RfA's should be adminned for 3-6 months and then reconfirmed (in a manner not dissimilar to WP:PROD), with a new RfA if not automatically reconfirmed. This is not dissimilar to the way it.wikipedia does things, although they take it to the bureaucratic extreme of applying it to *every* admin and I don't think that is necessary on en. Having seen some admins who do deal with very controversial sections of the encyclopaedia many of us would not dare to go near, who would probably fail such a reconfirmation solely because of the types of people they deal with, I would hate to think of them losing their rights because of a new failed community process. Orderinchaos 13:19, 5 October 2007 (UTC)

These are problems we already have. We're already good at distinguishing frivolous complaints from legitimate ones. We depend on crats to read consensus for promotion- I would surely think we could depend on them to know whether complaints are for good cause. Someone shouting "desysop him! he deleted the article on my dog!" will find that their ridiculous opinion is not taken seriously. Friday (talk) 14:39, 5 October 2007 (UTC)

It is very easy to make something seem worse than it is, through selective use of diffs, quotes, and the like - there's been plenty of RFCs where that has happened. If someone is truly motivated or is a popular editor, they can easily enough whip up a lynch mob over something relatively minor or infrequent. I wasn't suggesting the stupid complaints are the ones that would be an issue - more the malicious ones. Neil ム 16:07, 5 October 2007 (UTC)

Yes. Junior-high-school-style politics is also a problem we already have. But I don't see how RFA reform can particularly make that problem better or worse. A lynch mob can already try to derail an RFA for a stupid reason, for example. Friday (talk) 16:19, 5 October 2007 (UTC)

• In practice, this would probably work more like a 'block' than a de-adminning. It might be a bit strange for "admins" to be without their tools for a few days, but maybe some of them would learn from it. If someone was blocked from using the admin tools for a week for a poorly judged block, that might seem like a punishment ("but I want to help clear CSD"), but it should be considered more along the lines of a permanent record of the misjudgements. This would be no different to editors who get blocked early on in their editing career before they realise how things work around here. The problem is, this would require too much technical intervention to implement, so will never happen. Carcharoth 17:08, 5 October 2007 (UTC)
  • Hmm. Admins restoring their admin rights might be more of a problem here. There is a reason why only a limited number of users turn admin rights on and off. Probably best not to do it like this. Carcharoth 17:10, 5 October 2007 (UTC)
• Why does a cow need to be slaughtered? That goes against my religious beliefs. I believe a chicken (or goat) will be fair game for everybody. Nishkid64 (talk) 21:14, 5 October 2007 (UTC)
• Agree per Neil. Who is going to do the hard stuff and nasty blocks just to build up a grudge group who will try to desysopp at the first stumble? Resolute vandals and trolls will quickly establish a greater presence, and will likely target deadminning as a damage limitation limitation excercise. LessHeard vanU 13:50, 6 October 2007 (UTC)
• I've created a draft proposal for (another) de-sysopping procedure. Please discuss it and improve it. Mr.Z-man 01:13, 7 October 2007 (UTC)
• I can't imagine supporting anything which makes it easier to become an admin unless it becomes much, much easier to remove an admin. Admins once made, are essentially immune from removal, and for some, that immunity leads them to grossly abuse the trust placed in them by the community. So long as they don't attack fellow admins or the friends of influential admins, bad admins can abuse other editors with impunity. Only if it become practical to remove such admins would it be acceptable to make it easier to create admins. Argyriou (talk) 16:12, 15 October 2007 (UTC)
• To summarise why I think this proposal is misplaced:

  Dispute resolution, our fundamental community-based process for dealing with conduct disputes, can already remove the sysop bit. Nobody has demonstrated that there exist or are likely to exist significant numbers of admins who should be desysopped and yet are not.

  Nobody has demonstrated a mechanism by which a further community-based process for desysopping would cause potential administrator candidates to volunteer in higher numbers. It is not so much that RFA doesn't sysop enough candidates, but that not enough candidates volunteer.

• In other words, it isn't so much that RFA directly hampers production of sysops. It either doesn't encourage Wikipedians to feel that they would have more to offer with those extra powers, or it actively deters significant numbers of those who think they might. I cannot guess why this proposal is attractive, but I can state with confidence that it doesn't address the problem and is unlikely to deliver the many more admins we need. --Tony Sidaway 19:46, 25 October 2007 (UTC)

  Well, if we lowered our RfA standards it seems likely that more candidates would pass and more would come forward. There's no guarantee that instituting a "recall" process would prompt people to employ more forgiving standards, but I know I personally would be far, far more comfortable giving out the tools if I knew I could revoke them when needed. At the present time I usually don't participate in RfAs because I never want to have to think, "Crap, why did I give this person the tools?" At the same time, I don't want to hyper-demanding -- so I choose not to particpate at all.--22:59, 25 October 2007 (UTC)

Well we already know that the tools can be revoked in a matters of minutes. This isn't hypothetical, it's been done several times in recent months. We also know that dispute resolution can permanently desysop someone, and that in cases of serious abuse it's quite common for a misbehaving admin to be pre-emptively desysopped and then taken to arbitration.

The reason for my concern about RFA is that we've been promoting administrators at the rate of about 380 per year for the past three years, with no sign of any substantial increase (in fact it dipped a bit in 2006). The vast majority of all administrators just quietly get on with their work, you'll never even notice them, and consequently the number who have been desysopped is minuscule. The number of active editors has exploded in that time, and the workload has increased accordingly, but the low promotion rate has resulted in this tiny corps of admins having to deal with the growing workload.

And it really isn't about low pass rates, either. Let's get that out of the way. Over the past six months 202 out of 279 applicants have been promoted: a 72% success rate, and the rate over the past twelve months has been 51%. We'd hardly need to "lower our standards." Passing more than half of all applicants but greatly increasing the number of applicants would be fine.

It's just that this whole "we'd feel a lot nicer if we knew we could recall an admin" thing defies logic. Firstly, we can and do recall admins, secondly we don't need to pass a greater proportion of admin candidates than we do. What we need to do is to persuade more of the thousands and thousands of good, experienced, dedicates, trustworthy editors, that are editing Wikipedia that adminship isn't such a big deal, but it's not only quite useful to have but, in a way, as janitorship, essential to Wikipedia's organisation. --Tony Sidaway 01:09, 26 October 2007 (UTC)

The means to desysop people in extreme and/or emergency cases work fine (stewards & arbcom). But there are plenty of cases where we would choose to revoke adminhood from people who abused their powers in less dramatic ways. Until we have that power, we will only hand out the tools to editors with extensive and glowing records instead of to anybody who doesn't seem insane. Anybody who doesn't have such a record generally gets rejected; the remainder have good enough sense to avoid applying.

The barrier to acceptance is very high. Make us more comfortable accepting less predictable users and we'll lower the barrier. Lower the barrier and more people will jump over it.--Father Goose 08:25, 26 October 2007 (UTC)

I don't understand. What kind of "less dramatic" abuse would not merit desysopping by the arbitration committee but would lead to desysopping by some other (as yet undescribed) process? Surely if the administrator's actions are damaging Wikipedia and the only way to stop him is to desysop him, it wouldn't take more than a couple of minutes to demonstrate this to the arbitration committee and have the fellow pre-emptively desysopped pending arbitration. Anything other than that, go to the less formal methods of dispute resolution. Why would we want to desysop someone who didn't need to be desysopped?

And again, let me repeat: this isn't a matter of the barrier being too high. RFA already passes over 72% of applicants at the moment. The standards are fine as they are. But RFA as a process isn't attracting enough applicants. --Tony Sidaway 18:25, 26 October 2007 (UTC)

Admins can block users and protect or delete pages to enforce their views in ways that are not necessarily "against the rules" but nonetheless can form a long-term history of POV-pushing or bad faith. Arbcom will not act against these abuses except when a line has been crossed (usually wheel warring). Until admins have more direct accountability to users (not just arbcom), we won't take anyone who looks even slightly suspicious.

Frankly, I'm not comfortable having more admins until they have more direct accountability to the user community as a whole (not just arbcom). If, however, an effective community-driven desysopping process were in place, I would embrace your "give the tools to just about every experienced editor" proposal. In that vein, maybe we should invert RfA: give the tools automatically, and only hold a vote when they should be revoked.--Father Goose 06:46, 27 October 2007 (UTC)

Making these wild allegations won't do. Put up or shut up. If you have any evidence of an editor abusing his admin powers to push his point of view, and arbitration committee refusing to take action to stop him doing so, present that evidence at once, or stop making such claims without evidence. --Tony Sidaway 10:18, 30 October 2007 (UTC)

We've had a one year time-restricted adminstratorship system on the Swedish language wikipedia since 2006 and it is working quite smoothly so far. A few admins have been voted out, and a few have chosen not to continue, but the vast majority of sysops are re-elected for a new period of one year every time. How does it work in reality? Well, to help the stewards, and to gather the many candidates, the administrators are divided into groups (according to what quarter they were elected). These are then gathered into a large voting round which is held on the RfA page. The administrators in question are first notified (about a month in advance) that their administrative period is about to end, and are questioned whether they would like to stand candidate for another period or not. If an administrator wishes to do so, he or she will file a nomination or accept a nomination for administratorship. If no renewed election is kept, or if the candidate fails the election, then the users administratorship ends by the end of the month. If an administrator is to lose his/her privileges, then there will be filed a request at Meta so that a steward can control that everything is OK and proceeds with the de-sysoping. We use the same voting rules as in the RfA process, i.e a 75% majority for the candidate is needed to be re-elected.--MoRsE (talk) 00:52, 30 December 2007 (UTC)

View by User:WJBscribe

There are a lot of processes in the real world that aren't perfect - national election models, juries etc. They are often criticised, have inherent biases, and come to the wrong results. We stick with these systems for want of a better alternative. I feel RfA is another such system. It has flaws and limitation and proposals for change/replacements always merit full consideration but we've never thought of anything resoundingly better.
Is it broken?
Often I find that criticisms of RfA are actually criticisms of how people participate in RfA; that people are supported/opposed for the wrong reasons and that such reasons are given too much weight. I'm not sure RfA can be expected to magically eliminate these perceived "wrong opinions" of the Community - changes in process are unlikely to create very different results without addressing underlying attitudes as to what makes a good admin. I suspect working towards altering attitudes will be more effective an effort at reform than cosmetic changes to the process. If an oppose rationale is thought weak, it should be challenged and the contrary position advanced forcefully. But I do think it wrong to blame RfA for the perceived misdeeds of its participants - it is fairly refective of prevailing views as to what makes a good admin and flexible to changes in those views. I also think RfA creates a good balance between valuing the opinions of individuals (whether someone has or lacks another's trust is an important element to consider) and giving low weight to weak reasons (no need for the tools, low portalspace edits etc.) through using crat discretion guided by the weight of opinion on both sides to determine the consensus of the discussion.

Users that endorse this summary

1. WjBscribe 18:13, 4 October 2007 (UTC)
2. --AnonEMouse ^(squeak) 18:15, 4 October 2007 (UTC)
3. Addhoc 18:18, 4 October 2007 (UTC)
4. Mtmelendez ^{(Talk|UB|Home)} 18:19, 4 October 2007 (UTC)
5. Grandmasterka 18:20, 4 October 2007 (UTC)
6. — Moe ε 18:24, 4 October 2007 (UTC)
7. Nihiltres^(t.l) 20:13, 4 October 2007 (UTC)
8. Agree in principle but I am not so sure about giving low weight to weak reasons (no need for the tools...'. I mean, while you and I think "no need for tools" is a weak (or I'd like to say silly) reason to oppose, we don't really know whether any of the crats think so. This is the problem with so-called "crat discretion". We don't know what their own requirements for adminship are, so we don't know what they think is a weak reason and what is not. In the absence of a demonstrable community consensus on what is a reasonable oppose and what is silly, I don't see how crats exercise their discretion except by following their own biases and interpretations. - TwoOars (Rev) 20:56, 4 October 2007 (UTC)
9. The Basic model of RfA is possibly one of the best possible in theory: get input from the community on whether that person would be a good sysop, evaluate those arguments and the consensus, and then come to a decision. I doubt we'd be able to find a system better than that. The problems (whether real or perceived) are the reasons on which people decide their !vote. No change to the system will keep the good parts of RfA (an open invitation for community discussion) and be able to reform the process. The way to Reform RfA is to reform the way people !vote. Think the standards are currently wrong? Vote yourself. Decisions are made by those who show up. If the current standards for adminship are set by those voting, then they're the only thing we have that can act as a gauge on what the community values in an admin. WjBscribe is right: the process isn't broken, and no amount of tinkering will improve it. --Ybbor^Talk 22:55, 4 October 2007 (UTC)
10. I agree that much of RfA's problems can't be fixed by changing the process. Switching the way RfA works won't change the reasons people oppose for. Captain panda 23:45, 4 October 2007 (UTC)
11. Couldn't agree more. The flaws of RfA are best addressed by educating participants and making sure that people understand both the purpose of RfA and the importance of choosing quality admins. Pascal.Tesson 01:00, 5 October 2007 (UTC)
12. Echo Pascal's comments. — Dihydrogen Monoxide (H₂O) 01:16, 5 October 2007 (UTC)
13. Dureo 01:50, 5 October 2007 (UTC)
14. нмŵוτнτ 02:14, 5 October 2007 (UTC)
15. Most of the supposed problems with RFA would exist under any possible system of requesting adminship. --JayHenry 04:27, 5 October 2007 (UTC)
16. Sjakkalle (Check!) 09:46, 5 October 2007 (UTC)
17. I try to stay away from RfA because I often strongly disagree with the arguments I see there. We need all kind of contributors, the mop can be useful in many ways. For some you need a strong article writing experience (most of the dispute resolution process), for some you still need it, but less strongly (3RR, AfD, ...). But we have many good admins that work on the vandal/deletion/copyright side of Wikipedia, and that are seriously in need of help. Seeing contributors willing to help on this side be turned down because they didn't write a Featured Article (and don't plan to) is really depressing. -- lucasbfr ^talk 09:55, 5 October 2007 (UTC)
18. Orderinchaos 13:20, 5 October 2007 (UTC)
19. I don't think problems with the participants preclude discussing (and possibly trying) improvements, but we should recognize what the problems are. Sχeptomaniac^{χαιρετε} 16:18, 5 October 2007 (UTC)
20. Endorse, and raise you 2p: There's no perfect way of selecting anyone for anything; there are different ways to the way we do it now that can be explored, but the baby appears to be going down the plughole with the bath water with most other options being presented on this page. The RfA process can be poisonous; but this isn't the fault of RfA, this is the fault of the voters. More 'crat discretion, I say. ➔ REDVEЯS was here 22:33, 5 October 2007 (UTC)
21. · AndonicO ^Talk 00:10, 6 October 2007 (UTC)
22. Gizza^{Discuss ©} 11:31, 6 October 2007 (UTC)
23. Camaron1 | Chris 13:40, 6 October 2007 (UTC)
24. Indeed, the voting process isn't broken. Actually, it is in favor of 'Support', because editors don't need to do any reseach or effort to write "confident won't misuse the tools". Only if you oppose, you better have good arguments ready. What's needed are not new rules, but more and better candidates. Admins and Bureaucrats should actively search renowned editors as candidates. Gray62 02:19, 10 October 2007 (UTC)
25. -Icewedge 05:15, 27 October 2007 (UTC)
26. It isn't broken This is a Secret ^account 18:40, 16 November 2007 (UTC)
27. Don't blame the process for the actions of some of the editors who participate in the process. - Kathryn NicDhàna ♫♦♫ 22:43, 20 November 2007 (UTC)

Comment on the summary

I think, as with any process on Wikipedia, the process itself cannot be fully separated from the patterns of practice. The sub-optimal ways that people !vote and !discuss at RfA are learned behaviors that evolved along with the formal rules of the process. The assumption that the same problems would come up under any system is, I think, wrong; even rerunning the same system starting from scratch (i.e., with a fresh set of participants who had no prior experience with RfA) would probably result in a significantly different set of problem behaviors, since the issues Wikipedians are most concerned about now are different than when many of the bad conventions developed.

Even if the process itself has fewer theoretical problems than other systems we could imagine, starting anew with a virgin process and well-defined expectations for what the process is meant to do (i.e., what criteria it is to judge potential admins on) could alleviate some of the behavior problems. (But then again, it might not.)--ragesoss 16:43, 6 October 2007 (UTC)

View by User:Moe Epsilon

It seems that we are learning how to run before we learned how to crawl. I too believe requests for adminship is broken in the way ^demon described it and I believe that a new method needs to be created to contrast a difference from current methods.

However, I also believe a closure of requests for adminship right now is premature, at best. There is currently no other method of obtaining adminship at the present time. If we were to close RFA right now, we would have to formulate a new method of users being evaluated for adminship in a timely manner. Since that is constantly an issue among the community, frequently discussed at Wikipedia talk:Requests for adminship (and no changes are ultimately made), I don't believe the community can gather consensus without taking a year of debate to do so. When a new method for requests for adminship gains consensus from the community first, that is when the move to close RFA should be made, because without RFA, we are without any process at all. And I don't know about others, but I would rather have a process right now than have no process for a long period of time.

Users that endorse this summary

1. Moe ε 18:17, 4 October 2007 (UTC)
2. The timescales are open to question, but the basis that the alternative needs to be not only running but fine tuned before closing down the existing system is fundemental. LessHeard vanU 21:11, 4 October 2007 (UTC)
3. Unavoidably, it will take some time and patience. Adrian M. H. 22:32, 4 October 2007 (UTC)
4. · AndonicO ^Talk 00:12, 6 October 2007 (UTC)
5. We shouldn't be hasty on such a major issue. Gizza^{Discuss ©} 11:33, 6 October 2007 (UTC)
6. One of the reasons we don't get consensus for any change is that those desiring change are in radical disagreement on what change to make, and the current system is a reasonable compromise between the "pure vote, no discretion" camp and the "no vote, decision by the 'crat" camp. GRBerry 19:15, 9 October 2007 (UTC)

View by SamBC

RfA is one of the most inexplicable processes on wikipedia. It's true that a lot of the reasons people give for opposing, or refusing to support, a nomination aren't really relevant to either the general work of an admin, or the work the prospective admin intends to undertake. However, a system based on opposition only would certainly lead to fewer successful "promotions" (or mop-bestowings). As a user who would certainly like to be able to help out with menial admin stuff some time in the future, it certainly seems rather unpredictable what people will choose to object over, and asking for objections only isn't likely to lead to fewer objections. So yes, there should be some sort of reform, but not what's suggested by the submitters above. I can't say what the right reform should be, but I know it's not that. There should be some sort of process of the community, as many people as are interested, looking and trying to discern a way to "fix" (or rather "improve") RfA. The problem then is that any attempt to get wider community consensus once that idea is worked out by those who choose to get involved will probably have a lot of trouble, because it will almost certainl seem terribly radical.

Users that endorse this summary

1. SamBC(talk) 18:18, 4 October 2007 (UTC)
2. Ryan Postlethwaite 18:20, 4 October 2007 (UTC)
3. SQL^{(Query Me!)} 18:44, 4 October 2007 (UTC)
4. нмŵוτнτ 02:15, 5 October 2007 (UTC)
5. Gizza^{Discuss ©} 11:35, 6 October 2007 (UTC)

View by Grandmasterka

You say RfA is arbitrary - who's criteria are we going to use to determine who gets adminship? Often one person's arbitrary oppose is another person's legitimate concern (and the bureaucrats are smart enough to know a really arbitrary concern), and in reading a large portion of the RfA archives, no-one has ever adequately explained how changing the system we use will change people's views on adminship. Also, special cases often arise and we need to have a very flexible system for when those come up, like the one we already have. As for needing a quick desysopping process - unless you provide a specific example of someone who should have been desysopped right away using the system you propose, the discussion is pretty useless... Everyone wants quick de-adminship but no-one wants to provide a specific example of how their system would have worked. I've never fully understood all the fuss over RfA, especially when we have a ~97% success rate according to a recent thread on WT:RFA.

Users that endorse this summary

1. Grandmasterka 18:35, 4 October 2007 (UTC)
2. --AnonEMouse ^(squeak) 18:55, 4 October 2007 (UTC)
3. -- LessHeard vanU 21:14, 4 October 2007 (UTC)
4. -- Captain panda 23:59, 4 October 2007 (UTC)
5. Sjakkalle (Check!) 09:45, 5 October 2007 (UTC)
6. Arbcom can rule, and has ruled, to suspend someone's admin rights for a period of time - perhaps that system could be streamlined in cases where abuse of admin privileges is fairly clear. Aside from that, I agree. Orderinchaos 13:22, 5 October 2007 (UTC)
7. Gizza^{Discuss ©} 11:37, 6 October 2007 (UTC)
8. Indeed, this discussion is useless. RfA is working much better than the critics say. Many condidates who took the advice to heart and succeeded in their second RfA are evidence for this. Editors who invest considerable time and effort in this important task shouldn't be blamed for the lack of good candidates! The critics themselves should do more to interest editors in the job and to nominate those who have the right skills. Gray62 13:48, 14 October 2007 (UTC)

Comment on this view

The specific abusive desysop is well handled via a vandalism prevention type arrangement (which we sort of do now). The type of deadmin that the community might go for is one where the admin actions are short of instant action, and short of arbcomm specific incident action, but are enough that the community has lost an amount of trust. Some type of forum for dealing with this wether actions are taken or not is of needed in my opinion. --Rocksanddirt 23:20, 4 October 2007 (UTC)

Take a look at WP:DESYSOP for examples of users who were desysopped by ArbCom or for some other reason, or resigned as a result of controversy. Shalom (Hello • Peace) 14:49, 7 October 2007 (UTC)

View by Kusma

Before we dismantle RFA, let us stop and think what we want RFA to do. I want RFA to be an efficient and simple way to promote sufficient numbers of administrators while keeping out unsuitable candidates. The current system creates some unnecessary heated discussions (efficiency could be improved, but is better than in the RfC-style systems we saw), is reasonably simple for participants to use (just voting is still allowed), does not promote all that many candidates (number of admins grows slower than amount of admin actions that needs to be done), and rarely promotes unsuitable candidates. So the system works in a way, but could be improved. The problem is how to improve it. Unfortunately WT:RFA shows that there are many ideas, but philosophical differences between the Wikipedians involved have so far kept us from reaching a consensus. I won't offer a proposal here (although I can think of several very different directions we could take), but would like to see what other people think RFA should do: is your priority to promote as many as possible, or to keep all unsuitable candidates out at all cost? Kusma (talk) 19:30, 4 October 2007 (UTC)

Kusma is right. Before we get into an RfC frenzy and RfA overhaul, we need to answer some basic questions which however simple they may appear are probably quite subtle. What is the purpose of RfA? What do we need the end result of this process to be? How successful is it currently with respect to these objectives? Or even more basically, how can we even measure whether it's successful in attaining these goals? Until we get that sort of data, we will only get into shouting matches which result in a big hoopla and no change. For instance, there is a fairly prevalent feeling that the number of admins is growing slower than the amount of admin work to be done. Is this really the case? I happen to think some good candidates avoid RfA because they want to avoid that sort of drama. Is there any basis to that hunch, beyond anecdotal evidence? Let's agree about facts before we try to agree on reform. Pascal.Tesson 18:49, 5 October 2007 (UTC)

View by jc37

In reading the several views above, I think RfA may need "adjusting", but isn't "broken". The problem lies more with Wikipedians' expectations of adminship, than with the process in granting the mop.

Users that endorse this summary

1. - jc37 19:44, 4 October 2007 (UTC)
2. Yes. It certainly isn't clear to me that the criteria by which users assess candidates at RfA are always entirely relevant to the criteria which make a good administrator. Some of them undoubtedly are, but many are not. Sam Blacketer 19:50, 4 October 2007 (UTC)
3. --Rocksanddirt 20:06, 4 October 2007 (UTC)
4.  — Dihydrogen Monoxide (H₂O) 01:17, 5 October 2007 (UTC)
5. Orderinchaos 13:24, 5 October 2007 (UTC)
6. Strongly agree. To read some of the oppose votes, you get the feeling people think they're selecting the President of the United States. — iridescent (talk to me!) 15:57, 5 October 2007 (UTC)
7. Tbo ¹⁵⁷(talk) (review) 17:55, 5 October 2007 (UTC)
8. · AndonicO ^Talk 00:14, 6 October 2007 (UTC)
9. Captain panda 02:57, 6 October 2007 (UTC)
10. The current process is not perfect, but by no means terrible. Gizza^{Discuss ©} 11:39, 6 October 2007 (UTC)
11. The current system does have good points, it is how people use the system which is the problem a lot of the time. Camaron1 | Chris 13:44, 6 October 2007 (UTC)
12. Chris B • talk 15:44, 6 October 2007 (UTC)

Proposal by jc37

Perhaps one way to deal with the many concerns on this page would be to remove the Support section of an RfA. Those who support are obviously welcome to discuss with those who oppose or are neutral. This would eliminate the appearance of poll-like "voting" (since typically a support vote is just a signature, or a humourous comment, or a statement of "meeting my personal requirements", or even a refutation of one or more oppose comments), and would lean more heavily on Bureaucrat discernment.

Users that endorse this view

1. - jc37 19:44, 4 October 2007 (UTC)
2. This is something I could agree with, definitely a step in the right direction. I think the process should rely more on "Why shouldn't we make this person an admin?" rather than "How many people think they should be one?" ^demon^{[omg plz]} 20:05, 4 October 2007 (UTC)
3. I agree with ^demon that this idea is a good one. How to counterbalance opposition, though. It's quite easy for participation in RfA to vary from forty to two hundred in a single nomination on both sides of the fence so the question of judging consensus without a measurement of agreement is rough. Nice starting point, though. Keegan^talk 20:14, 4 October 2007 (UTC)
4. In my opinion, this is a step back to the "being an Admin should not be a big thing" ideal. After some threshold of edits & time spent active on Wikipedia, anyone who wants to be an Admin should be one -- unless they are clearly a crank, troublemaker, or otherwise a liability to the project. And the RfA process should be adjusted to be just that, not some mix of "vote" & "consensus". -- llywrch 20:24, 4 October 2007 (UTC)
5. My concurrence goes with this one. When you support them, what really do you support them for? Because they generally are great? But when you opose someone, you oppose for a specific reason. Supporting on RFA is typically a bunch of superfluous perfenander. MessedRocker (talk) 20:53, 4 October 2007 (UTC)
6. Agreed. — Moe ε 22:15, 4 October 2007 (UTC)
7. Absolutely...the best proposal so far, I think.  — Dihydrogen Monoxide (H₂O) 01:17, 5 October 2007 (UTC)
8. Yes, this idea seems really practical and not too much of a change that it will completely scare off people. нмŵוτнτ 02:16, 5 October 2007 (UTC)
9. Any objection would need to be supported with objective evidence - even on marginal cases, that should be straightforward to unearth. Spenny 12:55, 5 October 2007 (UTC)
10. Same as my rationale in my view below, but I agree totally. Spawn Man 09:52, 6 October 2007 (UTC)
11. Completely agree, this proposal fosters consensus rather than !vote-counting. shoy 15:16, 6 October 2007 (UTC)
12. I agree. We should have an "objections" heading and solve problems there. It should not have the appearance of a vote. -- _{Escape Artist Swyer} ^{Talk to me} _{Articles touched by my noodly appendage} 16:51, 7 October 2007 (UTC)
13. Carlosguitar 07:50, 8 October 2007 (UTC)
14. This sounds like an absolutely brilliant proposal! Atropos 20:44, 10 November 2007 (UTC)

Comment on this view by User:Sxeptomaniac

• However, this would replace a process with at least some positives into a pretty much purely negative one. I fear that those who already see RfAs as too negative will only be further turned off by something like that. Sχeptomaniac^{χαιρετε} 16:31, 5 October 2007 (UTC)

Users that endorse this summary

1. How does one determine community support if showing support is mostly disallowed while showing opposition is encouraged? In the current process, one may get 20 comments in opposition. If they get 50 comments in support, they will likely fail, if they get 100, they will succeed. If there is no real way for those 50 or 100 supporters to indicate their support besides waiting for someone to oppose and then giving a rebuttal (which could turn a controversial RFA into an even bigger battleground than it would be now), how does one determine the difference between those 2 cases? Mr.Z-man 16:52, 5 October 2007 (UTC)

Another comment on this view

• No, no, no. This is the worst possible way to approach it. The outcome of an RfA should depend 100% on the strength of community support - that's what "consensus" actually means, in the real world. Taking away the Support section would deny good-faith users a chance to air their legitimate views or to concur with the nominator (which is a valuable thing to do, as it shows that the nom's statement enjoys consensus). This seems like a step towards giving the bureaucrats more power, which is a very, very bad idea. Walton^One 10:19, 6 October 2007 (UTC)

Users that endorse this summary

1. I agree with Walton's take on consensus-building. While consensus is not a vote and arguments are meant to be assessed on merit, numbers do matter; that's why we have the whole "ask the other parent" business of involving wider community input on disputes. :) Support statements not expanded seem to me generally to indicate that an editor has read the nominator's statements and agrees. Opposition necessarily needs more detail, since while RfAs start with a "+" column, there isn't a "-" column from the start. Opposers, in essence, assemble that. It would be nice if every supporter could throw out a couple of diffs to build and expand the "+" column as, in practice, opposers are encouraged to build the "-" one. I suspect that a change of this nature would not only give the bureaucrats more power, but also give them more headaches by creating more opportunity for dissatisfaction in the community with RfA outcomes. Hypothetically, User:IDontLikeThisGuy may throw up 10 diffs to explain why the nominators are wrong about an admin hopeful. User:ILikeThisGuy may be the only one who steps up to rebut him, but there could be 99 other users who endorse the rebutting view. Unless they start piling on in an unhelpful "dispute per ILikeThisGuy" way (which would essentially duplicate the "support section", only under every "oppose") they can't make their opinions known without redundancy. --Moonriddengirl 13:02, 7 October 2007 (UTC)

Response to the oppositional comments above

How would this work? Initially the same way that indefinite blocking works. If there is no one willing to unblock, then they are by default "blocked by the community". Same with this. If no one opposes, then their RfA is successful. If there are opposes, then go in and discuss.

What this also does is make editors a bit more responsible for their comments, and removes a fair chunk of the "drive-by voting". (Yes, supporters may actually have to type a bit more than a single line comment in support of a cantidate - they may actually have to support them, by defense of their edits/contributions.) This also means a fair chunk of the appearance of a "popularity contest" should also diminish as well.

And by the way, this would not "deny good-faith users a chance to air their legitimate views or to concur with the nominator" - There is still a discussion section. It just wouldn't be filled with non-researched/non-explained "votes". So now, people would have to actually read what's being discussed, rather than just ignoring it, dropping some pithy comment, and being on their way. I'm not saying that that's what everyone does, but I think if we went over the last 100 RfAs (or more) we'd find that most of the "support votes" fall under one of 4 types:

• Just a signature.
• A humourous comment.
• A statement of "meeting all or most of my personal requirements", or "no concerns", or something similar.
• A refutation of one or more oppose comments.

So this proposal cuts out the first three, but still leaves the fourth. And thereby fostering discussion by default. (Consensus is, after all, the weighing of the comments/responses, not counting votes, as I presume we all know by now.)

The more I think about this, the more I think that this deals with a large chunk of the complaints about RfA. And personally, I don't see how this is much different than the discussions about good or featured articles. Or even talk page discussions about article content.

The "negativity" that one poster above is talking about is inherent in this system anyway. Why? Because unlike most situations here, we're discussing the person (or at least their actions/edits/contributions) as well as the content. And as I note in a response below, if they "can't handle the negativity from a single discussion, how will they respond down the road [when an admin] in their 5000th discussion? Or when they're the one determining consensus and closing discussions?"

And I think we should be able to trust the discernment of the bureaucrats, the same way that we should be able to trust the discernment of admins, though "even more so" since they've been trusted with "even more" tools. However, if you're saying that we can't trust the bureaucrats (or admins for that matter), then let's please start a discussion about how to fix that, because that's more dire a situation than concerns about the current state of RfA. - jc37 14:21, 6 October 2007 (UTC)

• No. Support votes which carry no clear rationale are still useful, because they demonstrate the weight of community support behind the candidate. They don't need to present arguments; the nominator has already done that, and a Support vote signifies that there is no reason to object. Numbers are important. If we only count objections, we will never know how many people actually support the candidate - and that is extremely important IMO, because the basis of RfA is that each user weighs up the arguments and votes accordingly, thereby exercising the community's right to choose its own administrators.
• Your point about "negativity" is also not quite correct. As a current admin, I can attest that I have never (on Wikipedia, at least) gone through a harsher and more negative experience than my RfA. Yes, some admins (who work in controversial areas) get more abuse after becoming admins than they do at RfA. But for the majority, getting through RfA is the worst part of our wiki-lives.
• To answer your final point, you've slightly misunderstood the nature of "trust". Yes, bureaucrats are "trustworthy" users, in that they've won enough community respect and support to reach their position, and are unlikely to go insane and sysop Willy on Wheels. However, the idea of "trusting" someone does not mean that I want to surrender the community's decision-making powers to them. It is possible to consider someone a trustworthy individual, and still disagree with them in good faith; and where there is good-faith disagreement, the community's will (expressed, where necessary, through a vote) should always prevail. Walton^One 16:56, 6 October 2007 (UTC)

• Numbers are not important. Else we need to poll all 5 million+ registered accounts, and an unknown myriad of anonymous IPs in order to truly find community consensus.
• And this also goes to a crux of one of the issues with RfA being a vote: "If we only count objections, we will never know how many people actually support the candidate..." - We're not counting objections either. We're asking for the community to point out any objections that they may have. BIG difference. Again, this isn't a vote, it's about determining whether or not the Wikipedian may be trusted with the extra tools. And it's up to you and me and the rest of the community to help by contributing to that discussion, should we choose to.
• Which directly leads us to your statement of "trust". I am, well, rather stunned by the comment: "However, the idea of "trusting" someone does not mean that I want to surrender the community's decision-making powers to them." - Then perhaps we should do away with blocking, CSD, and all things which involve administrative discernment. And remember, currently, every Bureaucrat is also an admin. So they've also already been trusted to use such discernment.
• As for being "correct" about negativity, that's subjective for every person. There's no "right" or "wrong" when considering how someone may have "felt", then or now. Though, as an aside, I wonder if "taking the RfA discussion personally" might be a sign that maybe the person shouldn't be an admin, or at least maybe spend some time in personal reflection. - jc37 17:19, 6 October 2007 (UTC)

• I don't understand how you can argue that "numbers are not important". Every established Wikipedian's opinion, given in good faith, counts equally. Your argument about polling all 5 million+ registered accounts is a straw man; just as suffrage is denied in the real world to minors, there would be no good reason to poll an account with no edits. And in almost all democracies (in private organisations as well as nation-states) there is no obligation to participate in a vote in which one is not interested. So you've distorted my argument somewhat; those who advocate numerical polling do not advocate some form of compulsory universal suffrage. But every established user who decides, in good faith, to offer an opinion in an RfA, should be counted equally. Those people who make edits to Wikipedia - the community - are the people who have a moral right to make decisions, and to choose administrators.
• As to the issue of trust, this is once again a straw man argument. Yes, administrators already have discretion in blocking and deletions; however, the exercise of these powers is heavily circumscribed by strict policies and guidelines, approved by the community as a whole. This is not the case with RfA; there is no policy page which gives "criteria for adminship". Users disagree in good faith about what makes a good admin, and promotions should therefore reflect the majority opinion of the community.
• With your third point, I apologise if I sounded arrogant in using the word "correct", and you're right that different people experience different emotions. However, I can attest that I have suffered no objectively serious conflict or abuse since becoming an admin - that is, no conflict that an average person would find emotionally distressing. In my experience, the experience of passing RfA is ten times more stressful than the experience of being an admin. Some will have had different experience, if they work in controversial areas, but I'm trying to bring the benefit of my own experience into this discussion. Walton^One 17:34, 6 October 2007 (UTC)

• Let's see:
  1. Wikipedia is not a democracy. There is no "suffrage", just merely joining in on the discussion. Which is one (actually two) of the Five pillars. This is the encyclopedia that anyone can edit. That goes for project pages too, obviously. And Discussions are resolved by finding concensus, not voting.
  2. Yes there is such a page. See Consensus. That covers all such discussions. (and has several related pages as well).
  3. I think the key words in your comments are: "In my experience, the experience of passing RfA is ten times more stressful than the experience of being an admin." Your experience is framed by the the choices you make, and the world around you. As you note, those who involve themselves in "controversial areas" tend to have to deal with more stress. But one cannot always determine what may be considered "controversial". And who knows what may happen tomorrow? One need merely to look over the archives of WP:AN/I to see who the "spotlight" of scrutiny can suddenly come upon an admin without warning. One difference to note, however, is that RfA scrutinises all of a nominee's edits/actions, as opposed to just a single action or set of actions.
• - jc37 17:59, 6 October 2007 (UTC)

• In regard to point 2, you've misunderstood what I was trying to say. I was saying that there is no policy page saying "Anyone promoted to adminship must have X number of edits, fewer than Y blocks, and N months on the project". In contrast, deletions and blocks etc. have clear policies, and while people express their opinion on an XfD, the eventual outcome has to be consistent with policy.
• The question of the relationship between voting, discussion and consensus is more difficult, and I think we'll have to agree to differ on that. I've always understood "Wikipedia is not a democracy" to mean that, as Wikipedia is a privately owned organisation with legal obligations and basic foundation issues, there are some aspects of the site's running that are non-negotiable (e.g. copyright policy, BLP, NPOV) and couldn't be changed by a poll; also, I agree that a discussion resulting in a compromise tends to be better than an adversarial vote. However, look at this in the narrow context of an RfA. There are only two possible outcomes on an RfA: pass or fail. Thus, there's no possibility of reaching a compromise through discussion. Somewhere along the line, someone has to make a decision - and IMO it has to be the community that makes that decision. Walton^One 18:06, 6 October 2007 (UTC)

• Actually, as I recall, RfA itself has a couple sub-pages talking about criteria, and "tendencies" in why people oppose. However, most of those fall under: "Needs more experience" or "Needs to more fully understand the policies and guidelines" or "needs to more fully follow the policies and guidelines". What are the demarcation lines of "more"? That's determined on a case-by-case basis and typically comes out in the discussion.
• If that's your opinion of consensus, then I sincerely suggest that you may wish to re-read Wikipedia:Consensus, as well as Consensus and Consensus decision making. I will note that I feel that the confusion of consensus with voting is one of the key misunderstandings of editors on Wikipedia, and I have a current request at Template:Welcome to include it.
• And the results of RfA are not pass/fail. They are "sucessful" and "not successful" nominations. - jc37 18:25, 6 October 2007 (UTC)

• I don't understand the relevance of most of this. As you correctly point out, the criteria for RfA are determined on a case-by-case basis and come out in the discussion - and in the discussion, people disagree in good faith over what constitutes "enough" experience, since there's no absolute benchmark. Therefore, where there is genuine and well-reasoned disagreement, we should go with the opinion that is supported by most editors. As to your second point, I am quite offended at your patronising manner. Yes, consensus decision making involves discussing and reaching a compromise, which is normally how things are done on Wikipedia (for instance, at AfD). But despite your attempt to obfuscate the issue with your third point (which again is not relevant), there are only two possible outcomes on an RfA. So there is no room for reaching a compromise, or taking minority views into account. Basically, the only sensible way to close an RfA is by the numbers; otherwise it simply involves the bureaucrats exercising power and closing according to their own opinions. Walton^One 22:48, 8 October 2007 (UTC)

• Again, I think you're misreading my comments, as no "patronizing manner" was intended. I was responding to: "I've always understood "Wikipedia is not a democracy" to mean that, as Wikipedia is a privately owned organisation with legal obligations and basic foundation issues, there are some aspects of the site's running that are non-negotiable (e.g. copyright policy, BLP, NPOV) and couldn't be changed by a poll; also, I agree that a discussion resulting in a compromise tends to be better than an adversarial vote." - That we apply consensus has little to nothing to do with that. We could "vote" on everything here, and the lawyers, etc would still exist. Hence why I suggested that perhaps you might want to read/re-read Wikipedia:Consensus and the articles related to it, because you don't seem to understand it, or perhaps you do, but it isn't something you favour. I can't tell. That said, I've noticed several other proposals you've made, and it seems to me that your main complaint/concern is with "The powers that be". Whether it be arbcom or bureaucrats, or whomever. I'm sorry to hear that you've lost faith in those whom the community entrusted with more tools and/or responsibility. Obviously, based on that, you would oppose any suggestion which places the responsibility of discernment in the hands of the bureaucrats or arbcomm. So, I suppose there's nothing more to be said about that. - jc37 00:27, 11 October 2007 (UTC)

A modification to the above proposal - by rocksanddirt

My suggestion (along the lines of jc37's and demon's suggestions would be to 1) do away with the sections on support, oppose and neutral instead request people make their arguement similar to the Xfd's, 2) do away with the vote count section as well. This would put a bit more emphasis on the discussion of qualifications and put a bit more work on the closer to sort it out, but might help reduce the arguementation on "frivolous" opposses.

In addition, I would like to see a community based deadmin process, though the use of arbcomm as the main deadmin is not "broken".

Users who endorse this summary

1. Rocksanddirt 20:14, 4 October 2007 (UTC)

Comment on this view by Keegan

I can see the point in this application of our other methods of maintaining the encyclopedia. This view actually serves to highlight what the major differences are between RfA/RfB, and the XfD processes.

The deletion system is set up to get community consensus on what is/is not encyclopedic and what does/does not serve toward that goal. Ultimately it is a cut and dry procedure from the outset based on policies, guidelines, precedent and the intended goal of the free encyclopedia that anyone can edit.

RfA/RfB, on the other hand, concerns the users that maintain the project. Opinions expressed are not rooted on any precedent aside from "Adminship is no big deal." Therefore intertwining the support, oppose, neutral comments would not serve to provide the consensus that the community needs because it would do nothing to benefit the process. RfA is a mixed environment of all the lot of Wikipedia contributors and it searches for candidates that come from all expertise.

In summary, these sorts of ideas are what need to be bandied about to institute not change, but a modification to fit the environment. RfA itself in Support/Oppose is not necessarily the problem but that the forum is used for platforming and grandstanding even if most nominations go the way they should. Keegan^talk 05:27, 5 October 2007 (UTC)

Comment by Walton

• I think this approach is the worst possible (no offence to anyone). As Keegan points out above, RfA is totally different from XfD. XfD outcomes have to be in line with the narrow parameters of deletion policy; also, many XfDs attract so little participation that closing them as a "vote" would be ludicrous. RfA, on the other hand, is where we, the community, exercise our legitimate right to choose our administrators. It is not the bureaucrats who have the right to choose administrators, nor should the 'crats "weigh up the arguments" and make a decision. Each individual user should weigh up the arguments and make a decision whether to Support or Oppose. RfA is a discussion, but it is also a vote, and this is not a bad thing. Walton^One 10:22, 6 October 2007 (UTC)

Comment by Deb

I really don't understand how this proposal (or the one above it) is supposed to work. Deb 12:51, 6 November 2007 (UTC)

View by Mr.Z-man

If RFA was totally broken, it would not work at all, however, I can't help but wonder how many good editors decide against nominating others or themselves (when they could be good admins) or accepting a nom because they are afraid of the process; the immense amount of criticism some people get (so much do that some users quit the project), the thought that one minor screw up will be used against them if someone can find it in their contribs, the thought that if they fail they might not be able to run again for another 4 months or more, the en-vogue reason for opposing changes so often, the ever increasing expectations of RFA voters, they must have an almost perfect contribution record (but not so perfect that it looks they have made adminship as their goal), etc. While the current RFA process is not (yet) broken beyond repair, reforms that, while maintaining the current thoroughness, would make the process (for lack of a better term) less scary and more uniform would be extremely beneficial.

Users that endorse this summary

1. Mr.Z-man 20:07, 4 October 2007 (UTC)
2. ➪HiDrNick! 20:09, 4 October 2007 (UTC)
3. Davewild 20:11, 4 October 2007 (UTC)
4. Very well said! SQL^{(Query Me!)} 20:12, 4 October 2007 (UTC)
5. Rocksanddirt 20:15, 4 October 2007 (UTC) - under the current system, I would not be nominating myself, or accepting any nomination.
6. SamBC(talk) 22:04, 4 October 2007 (UTC)
7. Because of the enemies I made because of my opposition of SIHULM, I couldn't become an admin if I wanted because SPAs would flood the RfA with baseless opposes. -Jéské ^{(v^_^v Kacheek!)} 22:33, 4 October 2007 (UTC)
8. RfA can be an extremely harsh process, and it seems to have become worse over the last few months. Acalamari 23:04, 4 October 2007 (UTC)
9. Adrian M. H. 23:19, 4 October 2007 (UTC)
10. WODUP 05:24, 5 October 2007 (UTC)
11. I know this was the case for me. — madman bum and angel 13:15, 5 October 2007 (UTC)
12. Absolutely. IvoShandor 05:11, 6 October 2007 (UTC)
13. By George, I think you've hit one of the many nails on the head! This is ever so similar to my rationale - I've given up commenting on tricky XfDs/RfAs/FACs as I might vote the wrong way or provide a view which I feel is correct, but no one else does. I know I have the skills, but it's just not worth the repercussions if I get one wrong. I've seen so many nit pickers at RfA pick up on very obscure details that I'm surprised editors who've failed previous RfAs even come back for a second round! You've got everything completely correct Z-man. Spawn Man 09:48, 6 October 2007 (UTC)
14. We need to make adminship "not a big deal" again (well at least not a very big deal). Gizza^{Discuss ©} 11:42, 6 October 2007 (UTC)
15. Well said, adminship is way too big a deal now. Not that being an admin is not a position of great responsibility, because it is, but it is a huge deal and people should not be leaving the project over such a small thing as a failed RfA. Neranei (talk) 19:27, 6 October 2007 (UTC)
16. Carlosguitar 08:06, 8 October 2007 (UTC)
17. I agree - I would never have wanted to be nominated the way things are now. On the other hand, there are just one or two people who need to be put off nominating themselves. Deb 18:01, 6 November 2007 (UTC)

Comment on this view by jc37

While I have to agree that the current RfA process is daunting, to say the very least, I'm forced to admit that perhaps that's not as incredibly a bad thing as one may at first think it is. Wiki-stress and Wiki-fatigue are becoming more and more common among our editors, and particularly the admins. Can any of us imagine how many times someone combatting vandalism or just closing an XfD discussion, was told that they were abusing the tools, had repeated postings to WP:AN/I about them, or even just the petty name calling, and so on? I would guess that that may grate on even the most positive of personalities. Not to mention dealing with the incessant turntable of controversy. No, I think RfA may need to continue to be as troubling and difficult as it is, just to help prepare the cantidate for what's coming, and to help alert the community about possible issues in dealing with confrontation and/or stress. Because if the cantidate can't handle the negativity from a single discussion, how will they respond down the road in their 5000th discussion? Or when they're the one determining consensus and closing discussions? So yes, the current situation can be daunting, and at times has really gone out-of-bounds of WP:CIVIL, much less, WP:EQ. But I think in most cases it may be a good thing.

Users that endorse this view

1. - jc37 07:12, 5 October 2007 (UTC)
2. Agree. I'm one of the least active admins in terms of using admin functions (20 blocks, 5 page protects and around 200 deletions, almost all uncontroversial speedies), and still have a talkpage full of abuse for it. Someone who'll be traumatised by "Oppose, doesn't need the tools" or "I'm not confident you'll understand policy" is probably someone who'll either snap back or leave the project after their user page is replaced by "you are a fat pedophile fag" three times — iridescent (talk to me!) 16:07, 5 October 2007 (UTC)
3. A bit of immature vandalism does not really compare to the harsh light of the interrogation process. When someone calls you a "poo head", you can't fail to come out of it looking the more mature of the two parties. Adrian M. H. 18:05, 5 October 2007 (UTC)
4. It's much more stressful being told that you're abusing the tools than that you don't deserve them. · AndonicO ^Talk 00:18, 6 October 2007 (UTC)
5. Admins will need to be hard headed to work properly. -- _{Escape Artist Swyer} ^{Talk to me} _{Articles touched by my noodly appendage} 16:58, 7 October 2007 (UTC)
6. Right! A candidate who is too thin-skinned for the process can't be a good admin. Gray62 13:53, 14 October 2007 (UTC)

View by Nihiltres

Adminship is one of the most contentious things about Wikipedia as a community - our administrators are our leaders and guides, regardless of the otherwise relatively flat hierarchy we have. It is this that makes our process for granting adminship frustrating - who is to say who among us deserves the mop for any particular purpose? While a cultural change about the process is definitely needed, I doubt that anything short of changing the nature of adminship will lower the high wall it is perceived that one must climb with the process. For one, granting adminship more freely but removing it as quickly would lower the wall - people who misuse the privileges won't get through any process a second time. I'd support the privilege of admin right removal being given to bureaucrats for this idea. Further, a system of single admin tools might be useful - if the tools were unbundled, users who might work only in deletions could specialize and gain a useful tool and do useful work even if they were known to be bad with the blocking right. This would be another method of lowering the barriers to adminship, which is what I think is really the issue in RfA - while adminship should be no big deal, it's often made a big deal of because of the different natures involved in it. This shouldn't replace our current adminship, but supplement it. If the community can choose to give people some rights knowing that those can easily be taken away, and they are taken away when misused, adminship might become less of a major issue.

Users that endorse this summary

1. Nihiltres^(t.l) 20:34, 4 October 2007 (UTC)
2. --Phoenix 15 20:45, 4 October 2007 (UTC)
3. The ability to desysop admins who turn out to be less good will probably allow users to be more likely to support users who they don't know. Od Mishehu 21:21, 4 October 2007 (UTC)
4. Good idea. -Jéské ^{(v^_^v Kacheek!)} 22:35, 4 October 2007 (UTC)
5. Support. I still think that maybe having more than one "level" of adminship might be good, as it almost certainly would increase the number of probably reliable people involved. Perhaps if we maybe gave a few, limited admin powers to "admin trainees", who could be chosen on the basis of apparent trustworthiness, and have the powers revoked by any reasonable objection to their continuing to have them? Those who are successfull and demonstrate trustworthiness could then be considered for full adminship later, with people generally knowing rather more about what they would be likely to do with the tools than they often do today. John Carter 23:56, 4 October 2007 (UTC)

Comment on this by jc37

In reading your proposal, it occurs to me that we could give the bureaucrats the ability to "block" admin tools, rather than the ability to remove them. Similar to the admin ability to block. Admins can't remove userrights from a user (such as the right to edit, etc), but they can block them from at least some of those rights by the use of "block". What if Bureaucrats could "block" an admin from using the admin tools for "x" length of time? - jc37 00:27, 11 October 2007 (UTC)

That's a pretty good idea - if we block editing privileges for editing abuse, blocking admin privileges for admin abuse sounds equivalently logically sound. I'd especially support this with the idea that individual tools could be blocked. Nihiltres^(t.l) 14:47, 11 October 2007 (UTC)

View on the Removal Process by User:J-stan

While the process is decidedly flawed, standards for RfA candidates are acceptable. People seem to want a very easy removal process. This will lower standards, and users who are unqualified will get promoted, however quickly it might get taken away. I propose that RfA standards stay the same, and users who are promoted go through a probationary period for their first two or three months. If editors have a legitimate cause for complaint, they can make their case known. If there is enough reason to desysop, they will get desysoped. I think that the users who deserve to keep the tools will make it through the probationary period, but those who make mistakes won't.

Users that endorse this summary

1. Even if there was a quick and easy removal process, adminship will become more like a toy if standards drop dramatically. Gizza^{Discuss ©} 11:45, 6 October 2007 (UTC)

View by Jéské Couriano

Because of all the things, minor or no, that could sink a nomination - irrelevant views expressed by the candidate on outside websites, a clique of editors who loathe the nominee, or a bad edit made months or even years ago - I see that the problem isn't that the nominee would be bad, rather it's that someone brings these up in an RfA (case in point: the CharlotteWebb RfA/Arbitration) and the process turns from a public forum to a Salem witchcraft trial. Support isn't much better, mostly just copping "I like him/her". I propose that we require all support and opposition to include a specific reason why they (mis)trust the user with the mop and bucket, and that off-Wikipedia behavior not be considered. -Jéské ^{(v^_^v Kacheek!)} 22:50, 4 October 2007 (UTC)

Addendum in response to Iridescent

I was referring to those whose views would end up getting them overwhelmingly shot down by a group of editors, not those that ran attack sites. Those that run attack sites aren't likely to be doing much more than vandalism or POV edits. -Jéské ^{(v^_^v Kacheek!)} 20:30, 5 October 2007 (UTC)

You'd be surprised - hop over to ED and see how many familiar names you spot. Even ruling the "pure" attack sites out, surely (to take the example I'd guess you're thinking of) Kelly Martin's blog was at least potentially relevant to her RFA? — iridescent (talk to me!) 20:43, 5 October 2007 (UTC)

Yeowch. That was nothing but venom, but, frankly, she was looking to make a point. That, and, I was referring to CW, as I stated above ("Case in point: CharlotteWebb...")-Jéské ^{(v^_^v Kacheek!)} 20:55, 5 October 2007 (UTC)

Users that endorse this summary

1. Jéské ^{(v^_^v Kacheek!)} 22:50, 4 October 2007 (UTC)

Comment on this view by Iridescent

I strongly oppose the "off-Wikipedia behavior not be considered". Yes, it discriminates against people who've made their off-wiki identities known, but I see no reasons why someone who is known to run an off-wiki attack site shouldn't have the fact taken into consideration. — iridescent (talk to me!) 16:12, 5 October 2007 (UTC)

Users who endorse this comment

1. Rocksanddirt 19:26, 5 October 2007 (UTC)
2. Right! Off-Wiki attack sides sure are an issue, and afaik exactly this situation is covered somewhere in a guideline on RfA. Gray62 21:53, 15 October 2007 (UTC)

View by Cla68

I'm going to repeat my idea that I posted on the RfA talk page a few weeks ago ([1]) and that is that an RFA be processed this way:

Once a new RfA opens, a bot or other program will randomly select a group of 20-30 editors and notify them to vote in the opened RfA. Anyone can comment in the comments section in the RfA, but only those randomly selected editors can actually vote. If any of the selected editors don't vote within a certain time period, the bot will replace them with other randomly selected editors until enough have voted to show consensus. This should help eliminate the current weakness in the process that allows a small group of biased or agenda-driven editors to unfairly influence the outcome of an RfA.

Users that endorse this summary

• Cla68 00:36, 5 October 2007 (UTC)
• Novel. A bit like jury service... LessHeard vanU 12:29, 5 October 2007 (UTC)
• There would have to be selection criteria for the voters, but it would be interesting. Maybe like the electoral college. Alternately, maybe we could vote on those who would be given the task of selecting admins for a given period of time. John Carter 14:17, 5 October 2007 (UTC)
• Comment - this has been suggested in the past. The selection criteria and implementing it technically are the big hurdles to overcome. All that would need to be seet up and tested before even considering this. Carcharoth 22:42, 6 October 2007 (UTC)
• A jury service is the most plausible option. — Dorftrottel⁠, was: aldebaer 10:33, 7 October 2007 (UTC)
• May work, but I think that we need a list of users who has interest to participate in RfA. Carlosguitar 08:24, 8 October 2007 (UTC)

Comment on this view by Moonriddengirl

My bit of a background with US law suggests that there might be some difficulties with this process. The jury trial system functions because of the support personnel in the court—prosecutors and defense attorneys who are well-versed in the processes and know what arguments are and are not relevant and judges who stand by to ensure that they do not misuse the process. Jurors are not entirely chosen at random—a large pool is called in and narrowed down by these support personnel to ensure that they have not a bias that will interfere with their performing their duties or that they are not in some other way unfit or unprepared to serve. (To prevent attorneys gaming the system, of course, their selection is controlled by policies and overseen by judges.) The ones who are selected are exhaustively instructed in how to weigh arguments, and the system has a built-in appeals process because in spite of the support personnel and the many oversight procedures abuses still happen. Granted, adminship is not the big deal that life in prison is. :) But I fear some of these same issues would be concerns. Who would monitor these randomly chosen voters for bias? Is there a threshold for time involvement in the project to ensure that they adequately understand what they're being asked to judge? Such a system might work if the randomly chosen "voters" were selected from a narrower pool than "all Wikipedians"—perhaps a large volunteer committee who had themselves been juried to avoid bias—but I think that the difficulty of implementing and overseeing the system could be prohibitive. --Moonriddengirl 13:23, 7 October 2007 (UTC)

Request by Neil

Do we have any examples of RFAs that failed when they should have passed, or passed when they should have failed, as evidence to back up this fairly big assertion that RFA is "broken"? Neil ム 09:26, 5 October 2007 (UTC)

I'd suggest maybe User:Gracenotes should have passed - Wikipedia:Requests for adminship/Gracenotes. A major problem comes when the 'crats apply their discretion and give some comments less weight - every 'crat does that differently and has different standards. There's also been a few which the community probably don't think should have passed such as Danny and Ryulong's. My point is that how can support of 70% show a consensus not to promote? That seems to be going against consensus in my eyes. Ryan Postlethwaite 09:30, 5 October 2007 (UTC)

At the risk of being flippant - you would say that, as you nominated Gracenotes for RFA (note, I did support GN in that RFA). But therein lies the problem; of all 3 of the examples you provide, the flaw is a simple numerical one - one was failed despite being over 70%, two were passed despite being under 70%. The system is only to blame for that situation in that it gives bureaucrats this nebulous "discretion"; we either accept they have discression and not bitch when they apply it in a way we do not agree with, or we get rid of any discretion (in which case, we could have a bot judge whether an RFA has passed). Neil ム 09:45, 5 October 2007 (UTC)

The problem I highlight is that the 'crats use their discretion in different ways, which has led to very inconsistant sets of results. Ryan Postlethwaite 09:49, 5 October 2007 (UTC)

Could you explain what you mean by "used their discretion in different ways"? They used their discretion, which leads to results that appear inconsistent when you don't know exactly what the bureaucrats were thinking. Or to results that are inconsistent depending on which bureaucrat closes an RFA. That is a feature of discretion. If you think it is a problem, then you should argue to remove discretion. Kusma (talk) 10:35, 5 October 2007 (UTC)

• • Danny and Ryulong should definitely have failed - those cases were just instances of the bureaucrats ignoring the community and going their own way. As to those which "should have passed", I don't think there's ever been one where the bureaucrats incorrectly determined community consensus in failing to promote; they're quite right not to promote controversial admins. However, there have been several (like Dihydrogen Monoxide's) where I wish the community had decided to give them the tools; that isn't a problem with the bureaucrats, but of the community (see my outside view below). Walton^One 09:52, 5 October 2007 (UTC)

If 3 RFAs (maybe 4 if you count Carnildo's) are all the examples of the RFA process failing there are, I would suggest that 4 out of the around 3000 RFAs that have taken place being "inconsistent" is not a bad percentage at all (0.13%). Even if we say (ridiculously pessimistically) that 50 RFAs have been closed in a wrong result, that would still only be 1.7% of all RFAs being closed badly. I would submit we are placing far too much emphasis on a very small proportion of adminship requests. Ones that fail because you, personally, disagree with the result (passed when you said "oppose", failed when you said "support") are neither a problem of the RFA process, nor of the bureaucrats, nor of the community (like Walton's example of Dihydrogen Monoxide's RFA). They are not a problem at all. Neil ム 09:57, 5 October 2007 (UTC)

Yes, fair enough - in mentioning Dihydrogen Monoxide, I was trying to get across the main problem with RfA; because admins are so hard to remove, the community is reluctant to sysop people who might well be competent admins. If we introduced a "votes for desysopping" process, the requirements for promotion would hopefully be less sky-high. But I agree that no one did anything wrong with Dihydrogen's RfA; the bureaucrats were right not to promote, since there was no consensus for promotion. Walton^One 10:14, 5 October 2007 (UTC)

To run with these examples, have the sysoppings of Ryulong and Danny demonstrably harmed Wikipedia? By this I mean definite disruptions and failings after the sysopping that you can point to, not nebulous hand-waving about long-term harm. Carcharoth 11:38, 5 October 2007 (UTC)

Ryulong's use of the tools has been the subject of much concern. Friday (talk) 15:43, 5 October 2007 (UTC)

There are a few admins I once supported, who, if I had any idea how they would conduct themselves after becoming the admins, I would have strongly opposed (I vote in the RfAs of only people I know reasonably well, and this is despite that). I won't name them, obviously, but I am sure most of the RfAites would have had the same experience. Tintin 15:39, 5 October 2007 (UTC)

• (Reply to Carcharoth) I don't know about Danny, but when Ryulong blocked this good-faith user for 24h for absolutely no reason, the user left the project (he hasn't edited since April). This is an example of what happens when bureaucrats rely on their own "judgment" rather than the will of the community. Walton^One 15:52, 5 October 2007 (UTC)
  • Heh. I'm looking at "this good faith user"'s block log, and reading 9 blocks, by 5 different admins, with reasons from (admitted creation of vandalbot; physical threat) to (disruption of wikipedia (trolling, hate jokes, ignoring warnings)). The last block, the one you're objecting to, seems to be because he nominated Wikipedia:Miscellany for deletion for deletion at Miscellany for deletion. "for absolutely no reason" seems to be not quite correct. For a one time disruption like that, I would probably argue for a warning, but with a record like that, I could see someone blocking indefinitely, much less for 24 hours. If this is the worst we can find about Ryulong, he deserves praise, not criticism. --AnonEMouse ^(squeak) 15:20, 6 October 2007 (UTC)
    • The trouble with block logs is that they give the blocking admin's opinion, not the whole story. The alleged "hate joke", IIRC, was when he replaced the <!--End of MfD message... with <!--End of "nazi" message... on an MfD notice on his own userpage, after Ryulong had nominated almost all of his userpages for deletion (again, for no apparent reason). If blocking users for making one edit to their own userpage is considered proper admin conduct, then evidently I'm far too lenient to be an admin. :-) To be fair, though, this was the only time I'd encountered Ryulong, so I won't assume this to be a typical example of his use of the tools. But his treatment of Walter Humala was overly trigger-happy, IMO. Walton^One 19:44, 8 October 2007 (UTC)
      • Ryulong trigger-happy? Say it isn't so! Videmus Omnia ^Talk 20:14, 8 October 2007 (UTC)

  While I do trust the discernment of our bureaucrats, perhaps a way to resolve the above would be to require at least 2 bureaucrats (3?) to deem an AfD nom "successful". Since, atm, it can be a rather long drawn out process to de-admin someone, and otherwise the tools are "for life", having at least a second opinion might not be a bad thing? - jc37 16:04, 5 October 2007 (UTC)

  AfD ... RfA? I think the bureaucrats do have a "bureaucrat chat" when it's close. There's certainly no need for multiple people to say "yes, it's a pass" if an RFA has 100% (or near to 100%) support. Neil ム 19:20, 5 October 2007 (UTC)

  "Bureaucrat chats" are always a bad idea, for the reasons I outlined above. The community is the only body with the legitimate power to select admins. The job of the bureaucrats is solely to implement the will of the community. On the few occasions where bureaucrats have made a collective decision, they've too often got it wrong (viz. Danny and Ryulong); hence why they should close by the numbers. Ultimately, a controversial candidate who is opposed by 30-40% out of hundreds of participants is never going to make a good admin. The bureaucrats should bow to the community's judgment. Walton^One 10:05, 6 October 2007 (UTC)

  (Yes, I meant RfA) - And in hindsight, I think that a "bureaucrat quorum" (which is somewhat what I was suggesting above), for every RfA would probably be a bad idea, since it could become a second RfA. However, I think it's a great idea on a case-by-case basis. And I think any bureaucrat should be able to ask for one in any case (which would place the RfA on "hold").

  But besides that, I have to disagree with you on "too often got it wrong". But I suppose that's a subjective debate from either side of the fence. - jc37 15:34, 6 October 2007 (UTC)

Wikipedia:Requests for adminship/WikipedianProlific should have passed too.Rlevse 16:35, 9 October 2007 (UTC)

View by Sjakkalle

I am unsure what people mean when they say "RFA is failing". What is the measure of success?

The role of RFA is to get new administrators who are competent and up to the job. It is also a process to avoid granting unsuited candidates adminship. In order to determine whether someone is competent, we defer to the community to make that decision. So there are two fundamental ways in which RFA can fail.

• Competent candidates are denied adminship.
• Incompetent candidates are granted adminship.
• People are opposing for ridiculous reasons.
• People are blindly supporting.
• Bureaucrats are ignoring consensus.
• Bureaucrats are following consensus (=mob rule).

There will, in a large community, nearly always be someone who dissents. They may have good reason, and they may have stupid reasons. But take heart, usually stupid reasons don't get much support, and if they do, the reason is probanly not that stupid. When there is disagreement there will be people upset when the decision goes against them.

I think the level of anger is greater when there is perceived unfairness or double standards, as was the case in the Carnildo, Danny, and Ryulong RFAs, hence I think bureaucrats should usually follow the consensus (in other words, the "mob rule"), but I know there is disagreement and controversy over this in general, and those RFAs in particular, so I'll let that issue rest for now.

If we want to continue having the community decide who becomes admins, and who does not, has anyone yet proposed a system which obviously will work better than the current system? A few systems were tried out some months ago, with supports and opposes collected together to facilitate discussion, and another "RFC format" where different opinions were endorsed, but neither gained significant support over the current system. Until we have system which is better at determining community consensus than the current model, we cannot abandon the current model without abandoning community control.

On the issue of people's criteria for supporting, we should be aware that, consensus aside, there are no policies in place which will demand that someone must be promoted. This is different from AFD, we have policies saying that unverifiable content cannot stay around forever, and the community cannot overrule that. We have no such policies in RFA. Some people think it is ridiculous to oppose someone because they haven't contributed enough to mainspace becuase adminship is janitorial work, and writing great articles has nothing to do with adminship. Others argue that article writing is what is needed for a candidate to understand the core policies. This is a difference in opinion, and it is impossible to say that one or the other is right, objectively speaking. We can debate for a long time over issues like that, but we are unlikely to ever have a full agreement there. Before leaving the issue of "ridiculous" reasons, I want to say that reasons considered silly by most normal people do not topple a candidacies today. I have seen no examples of an RFA failing because "self-nominations are prima facie evidence of pwer hunger". Trolling reasons, which are not reasons or opinions at all, but simply vandalism/disruption ("Candidate is a s***hole") are already ignored. The fact that we have no agreement over what criteria or qualities we should look for in the admin candidates is not a problem with RFA, but a concern which has to do with the community.

Judging on the standard that we get the right people promoted, RFA is definitely failing and I can prove it: I supported the candidacies of CharlotteWebb and R and they failed. I opposed Khaosworks and Butseriouslyfolks and they succeeded. OK, I'll mention that I am using ridiculous criteria on RFA candidates, because the two above who I opposed turned out to be good admins. Any community based process will be no better than the community who participate in it

I am fully aware that Wikipedia is not a democracy, but I will finish off with Churchill's opinion of democracy: ""Democracy is the worst form of government, except for all those other forms that have been tried from time to time."

Sjakkalle (Check!) 09:35, 5 October 2007 (UTC)

Users who endorse this summary

1. Strong endorse. The community has the right to select admins. Sometimes the community makes mistakes, but it's still better than having Wikipedia run by a self-selecting elite. Walton^One 09:54, 5 October 2007 (UTC)
2. Orderinchaos 13:30, 5 October 2007 (UTC)
3. endorse this summary, Gnangarra 13:56, 5 October 2007 (UTC)
4. Mailer Diablo 05:14, 6 October 2007 (UTC)
5. Endorse. It's not that RfA is incapable of improvement, but editors need to have a realistic assessment of what proposed changes would mean - the negative consequences as well. Sam Blacketer 10:16, 6 October 2007 (UTC)
6. Endorse. Far too many people, when they say RfA is broken, really mean that the results don't always agree with what they think the result should be. -Amarkov moo! 22:49, 6 October 2007 (UTC)
7. Nail, head, bang. —Cryptic 01:50, 7 October 2007 (UTC)
8. Very well put. --Quiddity 07:44, 8 October 2007 (UTC)
9. Strong endorse. Sjakkalle is a genius. Xoloz 12:10, 12 October 2007 (UTC)
10. Totally agreed, there is no consensus over which criteria will make good admins and bad. Carlosguitar ^{(ready and willing)} 03:53, 29 November 2007 (UTC)

View by Walton

There is absolutely nothing wrong with the RfA process itself, or with the voting process. Ultimately, sovereignty on Wikipedia belongs to the community as a whole, and it is the community who have the inviolable right to select administrators. Each established user in good standing should weigh up the arguments and decide whether to vote Support or Oppose. This is why the discussion - challenging other people's comments - is also an essential part of RfA, because it allows the arguments and evidence to be fully explored, and allows strong arguments to influence others.

The only problem with RfA is that, as Friday points out, admins are very hard to remove - hence why we have a lot of semi-abusive admins, who don't do anything bad enough to get sent to ArbCom, but are habitually incivil and uncommunicative. This, in turn, makes RfA harder to pass, as people are reluctant to give the tools out unless they trust a candidate 100%; hence why good editors like Dihydrogen Monoxide and R have been consistently failing to get adminship, even though they would both make competent admins. If we had a "votes for desysopping" process, RfA would become easier to pass, because people would be more willing to give out the tools knowing that they would be given out easily.

Admins are, fundamentally, the exercisers of power and judgment; and like all such power-holders, they must be held to account by the community. An admin who is not supported by the community loses their legitimacy. Despite the much-overused maxim that "Wikipedia is not a democracy", Wikipedia belongs (morally) to its community; it's the community who create and maintain it, and the community alone who should decide how it will be run. The community has an inalienable right to select admins through a fair, equal voting process, and should also have a right to remove them at will. Walton^One 09:52, 5 October 2007 (UTC)

Users who endorse this summary

1. Carbon Monoxide 11:42, 5 October 2007 (UTC)
2. As I said before, with minor reservations. I believe every admin makes mistakes, and should be given the opportunity to correct them, especially if they were made in good faith such as a lack of specific knowledge or a quick, innocent misjudgment. However, Walton summarizes a key principle to any organized community: accountability. The community should have the right to hold an admin accountable for his/her overall performance, and not only for one negligent action (as is the usual, but not the only case in ArbCom proceedings). - Mtmelendez ^{(Talk|UB|Home)} 13:28, 5 October 2007 (UTC)
3. Rocksanddirt 15:25, 5 October 2007 (UTC) - heartily endorse this postition on the need for a better deadmin process.
4. Nihiltres^(t.l) 15:37, 5 October 2007 (UTC)
5. Strong endorse — iridescent (talk to me!) 16:15, 5 October 2007 (UTC)
6. ♠TomasBat 21:08, 5 October 2007 (UTC)
7. Mailer Diablo 05:13, 6 October 2007 (UTC)
8. Generally concur, but removal should be made more exacting than giving as arbitary removal of the mop will discourage application by some editors - and may result in having sysops not wishing to make difficult actions because of the possible repercussions... so that would be a weak endorse! LessHeard vanU 11:14, 6 October 2007 (UTC)
9. Camaron1 | Chris 13:49, 6 October 2007 (UTC)
10. Agree that a more community-accessible venue than ArbCom is needed for making the will of the community known. shoy 15:24, 6 October 2007 (UTC)
11. Chris B • talk 15:45, 6 October 2007 (UTC)
12. ≈ jossi ≈ (talk) 23:06, 6 October 2007 (UTC)
13. a.z. 00:42, 7 October 2007 (UTC)
14. Gizza^{Discuss ©} 22:51, 7 October 2007 (UTC)
15. нмŵוτнτ 03:13, 10 October 2007 (UTC)
16. I have some minor disagreements with Walton's largely democratic approach in other areas; but, for RfA, he is absolutely right. Trust can only be conferred by the community -- it cannot be bestowed from the top. (Of course, there can be singular exceptions to this general rule -- as with the adminning of WM's legal counsel -- but the guiding principle should be firmly acknowledged as beyond dispute.) Xoloz 12:14, 12 October 2007 (UTC)

View by Carcharoth

RfA is a community process that decides who should be given the administrative tools. But the fundamental question, in my view, is this: Are those who pass RfA suitable for adminship? If the answer is "sometimes, yes, sometimes, no, sometimes we won't know until they start using their tools", then what we need is another process to assess admins and decide whether they are using their tools properly. Let RfA continue the way it operates at the moment, but make clear that all it is is a community process to decide who should get the admin tools, but not to decide who should keep them. To counter-balance it, start up a new process that is designed to decide whether those who have passed RfA and are using their admin tools, are using them well. This would not be a desyopping process, but more an assessment process. Much like we have "editor review", this would be "admin review". Almost like an RfC, but focused on admin conduct. Again, the problem is how to avoid drama, and to avoid discouraging admins from working in controversial areas, but I don't think that problem can ever be overcome. If an admin's conduct ever gets to Arbcom level, then these previous reviews and RfC-type reviews would be a useful step in the process. Carcharoth 11:52, 5 October 2007 (UTC)

Additional comment - Another point is that people shouldn't think that RfA is the only point at which to stop unsuitable people becoming admins, or to restrain unsuitable behaviour from admins. RfA should probably be seen as the first barrier to stop those who are unsuitable. But the community should not relax after this, and there should be a robust and searching review process that works to restrain and retrain borderline admins, while not supporting trolls and POV-warriors. That way there would be something in-between RfA and Arbcom to question admin actions. The periodic threads at WP:AN (either "admin abuse" or "please review my action") sometimes serve this purpose, but one alternative would be if the RFC process was simplified to allow review of specific admin actions in a "how could this have been handled better" type reviews. And then the results actually noted somewhere so successive generations of admins actually learn something from all this. Carcharoth 22:37, 6 October 2007 (UTC)

Users who endorse this summary

1. Proposer. Carcharoth 11:52, 5 October 2007 (UTC)
2. John Carter 14:15, 5 October 2007 (UTC)
3. This could make room for a slightly less critical and nitpicking approach in RFA. I would support this, provided that it does not add too much to the workload to be practical. Adrian M. H. 15:12, 5 October 2007 (UTC)
4. Not a bad idea. Walton^One 15:54, 5 October 2007 (UTC)
5. I like the idea, I can imagine that this may lead to less worried and unpleasant attitudes at RfA, once it gets going. SamBC(talk) 16:59, 5 October 2007 (UTC)
6. I like this idea quite a bit. The reason we are clamoring for a desysop process but can't/won't come up with examples of who should be desysopped is because most of the "abuse" is at the border of acceptable content, and no one wants to lose face and make enemies by campaigning for the desysopping of someone only to have the community back the admin. Framed as a discussion of how an admin has used the tools rather than a disciplinary measure, this would be much more effective in identifying and stopping problems and keeping clear the standards of acceptable behavior (which change as the community changes).--ragesoss 17:11, 6 October 2007 (UTC)
7. An "admin review" is a good idea. Admin hopefuls would have a clearer idea of what is expected of them. Gizza^{Discuss ©} 22:53, 7 October 2007 (UTC)
8. Endorse, per Adrian M. H., ragesoss, and Gizza. --Quiddity 07:51, 8 October 2007 (UTC)
9. Support, but the workload problem is real. Editor review & admin coaching are both excellent related schemes that in practice suffer from a lack of experienced editors. Espresso Addict 19:39, 10 October 2007 (UTC)

Comment by user who is neutral to this summary

1. Similar to ragesoss's point, given the current climate of nearly unconditional support of admins for other admins in disputes with peonsordinary users, naming names is not an action to be taken lightly. When users get blocked for compiling evidence for RfCs, accusing abusive admins is a recipe for drawing the abuse on oneself. Argyriou (talk) 19:59, 15 October 2007 (UTC)

View by Spenny

I think there is a schizophrenia in adminship that is at the root of the discontent. There are two different tasks: the routine maintenance of the database and the first line of arbitration and enforcement of the policies of Wikipedia.

In some responses we see adminship as some tools to be given and some tools to be taken away. That is an uncontroversial decision and should be no big deal: has the person demonstrated the appropriate behaviour that would make the tools a useful addition to their contribution? Abuse the tools - de-sysop.

The controversial element is the oversight. There are plenty of people, admins included, who have had run-ins with admins. There are some admins who rule the roost and have taken it as their personal duty to dictate the rules, often under the guise of this being for the good of Wikipedia. There are others who perform this role brilliantly. Aside from page protection, they don't need tools.

Perhaps the issue is that we need moderators and we need admins. The two tasks are different and neither the selection process, nor the deselection process acknowledges this. Do we too often see the foibles of a diligent admin forgiven when they screw up on the moderation - I think we do. We need people to work to their strengths and allow weaknesses to be recognised.

So I think this is asking the wrong question and is simply dealing with one of the symptoms of a deeper issue. Spenny 15:46, 5 October 2007 (UTC) (oops!)

Users who endorse this summary

1. Carcharoth 13:57, 5 October 2007 (UTC)
2. John Carter 14:14, 5 October 2007 (UTC)
3. Rocksanddirt 15:28, 5 October 2007 (UTC) - excellent point.
4. I am pleased to see that someone has highlighted the distinction between the two roles. Adrian M. H. 16:42, 5 October 2007 (UTC)
5. This does make sense, although I'm not sure that the "moderation" role needs any formal recognition. What may be better is clarifying, as loudly and clearly as possible, that admins aren't moderators in general. SamBC(talk) 17:35, 5 October 2007 (UTC)
6. Endorse, plus the 4 comments currently below. --Quiddity 08:57, 8 October 2007 (UTC)
7. endorse, with the caveat that removing the power to punish users is far too hard right now. Argyriou (talk) 20:02, 15 October 2007 (UTC)

Comments

I think formally recognising the moderator role would be helpful. The admins who enjoy pottering in the background are then shielded from the politics, and there is clarity in what the role is for those brave souls who put themselves on the line aiding the less co-operative of our editors. As I suggested, I see one problem as there are some really hard workers who need recognition for that, but probably only get noticed when they get involved in dealing with users, where their skills are not so good. Other admins are then unwilling to take them to task because they don't want to offend someone who does a lot for the project. Yet we have some really good moderator admins who can diffuse situations, and they should be recognised for that rare skill. Spenny 18:35, 5 October 2007 (UTC)

Absolutely. Though as a note of caution I would point out that there are some admins who get involved in moderation roles and who don't start off or even remain objective. That needs to be addressed as well (ie. stopped). Truly uninvolved moderators are needed in some places. This also happens on policy and guideline pages as well, with entrenched camps staking out their positions and an uneasy stalemate ensuing. Carcharoth 11:50, 6 October 2007 (UTC)

Also, any editor can be a moderator. I've tried moderating some disputes, with not much success. There also seems to be a mentality among some editors involved in a content dispute to ask "an admin" to resolve the dispute, when they should realise that it is the community of editors involved in that page that need to come to a consensus. The perception that admins, because they can block, are best suited to resolving disputes, needs to be changed. Blocking should only happen when incivility or disruptive editing (reverting) is disrupting constructive discussions and editing. Carcharoth 11:53, 6 October 2007 (UTC)

The trouble is, though, that a really effective moderator has to have some degree of authority. Experienced users realise that admins have comparatively little wiki-political power (though we do have some), but the newer users tend to assume that we're the only people with the "authority" to sort out their disputes. I quite like the idea of creating a separate "moderator" position with formal authority, although I'm not sure it will gain consensus. Walton^One 23:03, 6 October 2007 (UTC)

Moderation sounds awfully close to binding mediation. Is this the intent? - jc37 00:27, 11 October 2007 (UTC)

Not at all, a moderator is simply, if you like, a facilitator of discussions with an ability to give a slap on the wrist - in other words, exactly as happens now in the best cases. I'm not sure where you made the leap to binding mediation apart from syllables and the letter M :) Spenny 14:15, 12 October 2007 (UTC)

View by Laleena

Note: I made changes to my summary 12:07, 7 October 2007 (UTC)

I think that it is correct to reform the process. However, I am asking that everyone be made to apply on the current page, but only allow certain people that have been drawn randomly (a sample of the community whole) to vote and be counted. This is probably not a great idea, but it seems like we could then make community-run. Oh, and by the way, there must be strict regulations on this, drawing for each.

Users who endorse this summary

1. Laleena 12:49, 5 October 2007 (UTC)

Response to this view

• I strongly disagree with this. "The most important people in the Wikimedia Foundation" have neither the right, nor the good judgment, to take power away from the community. The Wikimedia Foundation may own the servers, and, yes, they have a legal duty to intervene in Wikipedia's internal affairs in certain rare cases. However, Wikipedia belongs to its community, and the community is sovereign. We don't want "the great and the good" making decisions; we want the whole community (the people who actually write and maintain the articles) to make binding decisions. For an example of where "the most important people" get things totally wrong: Kelly Martin was appointed to the first ArbCom by edict of Jimbo Wales. If you would care to peruse her recent failed RfA, you will see numerous diffs which demonstrated her total unsuitability for any position of authority or trust. The community is more reliable than the Foundation, therefore, in determining right from wrong in the appointment of admins, and also has a moral right to make these decisions IMO. Walton^One 15:58, 5 October 2007 (UTC)
  • I tend to agree with Walton. The disadvantage of giving the power of appointment exclusively to the Foundation is that those individuals, for all their good qualities, are less involved in the day-to-day functioning of wikipedia than many of the editors are, and on that basis are almost certain to rely upon others for some of their information. We have no way of knowing if these sources would themselves be necessarily aware of the individual strengths or weaknesses of individual editors, and this presents a situation in which very serious problems could, and almost certainly would, arise. I agree with the idea in theory, to a degree, but see major difficulties if were to ever be put into practice. John Carter 16:03, 5 October 2007 (UTC)
    • Walton hits the nail on the head. Giving the Foundation this kind of power (quite aside from the fact that they don't have the time and I suspect not the inclination to exercise it) converts the whole project from one of the most successful community-run organisations in the world to a glorified blog for Jimbo's friends. (I suspect ^demon, who proposed this RFC, can think of a rather more recent example of one of the "most important people" taking a very dubious decision.) — iridescent (talk to me!) 16:22, 5 October 2007 (UTC)

View by OhanaUnited

There are flaws in the RFA (some are major, some are minor). I'm not here to against the removal of RfA, nor promoting to retain it. I'm here to explain the flaws and provide ideas so that we can correct this system. Instead of writing an essay, I decided to put them into point form:

• (major flaw) Way too many people snowball votes. You can see this very often when people say "per XYZ editor" and without any additional comments.
• (major flaw) Oppose for the sack of oppose. (Scenario: You opposed me in my RFA, I'll oppose yours when you apply for adminship.) How does that help Wikipedia?
• (minor flaw) Humans tend to speak up more often when they want to oppose something. If you took statistics courses, your teacher/prof probably mentioned about this kind of bias.
• (somewhat major flaw) Oppose because you are not active in (fill in the blank). If you are not active in voting for RfA, you shouldn't be opposed because of that. What you should say is vote neutral and say something like "I am concerned because you're not active in XYZ department."
• (major flaw) Oppose because you're self-nom (or nominated by XXX editor). The system allows us to do self-nom, so there's no way you should vote oppose because of who nominated you. Yet these votes are often counted for the %.
• (super major flaw, ^demon's idea) RfA claims to be working based on consensus, in accordance to Wikipedia's guidelines, but the reality is not the same way as it says. What is the definition of consensus? 50%? 70%?
• (minor flaw, not occured in English Wikipedia at the moment, but I believe it will occur sooner or later, could turn from minor flaw into major flaw) Support the candidate and ask for something in return. I was in Hong Kong this summer and went to several WikiMeet as well as the formation of WikiMedia Hong Kong division. I exchanged the status of English Wikipedia to some users and admins using Chinese Wikipedia. They said once in a while, this kind of "group" support pops out in RfA. Let me give you an example:
  • Users (A, B, C) decided to nominate a user (D) for adminship. Person A, B, and C voted support for D.
  • In return for obtaining the mop, person D becomes a protector for A, B, C.
  • Whenever those 3 players want to push POV, D can jump in and "shield" the fire directly at those 3.
• (minor flaw) After obtaining the mop, some admins went on extended wikibreak. We need something so that admins actually do their job after obtaining the mop. This will also discourage from obtaining the mop simply because they want to be admin, but not serving the community.
• (major flaw) Oppose because of edit count, activeness, how long the nom joined Wikipedia. We should be judging if we trust the nom, not because how much work he did.

OhanaUnited^{Talk page} 17:54, 5 October 2007 (UTC)

Response to this view

• This is substantially why we have WP:AAAD, and why discussion of RfA comments is not only allowed, but encouraged. Arguments such as "Doesn't have 2000 projectspace edits" or "only been active for 3 months" are, as a rule, challenged; the editors who base their votes on these factors are entitled to their opinion, but it tends not to influence anyone else. I also don't agree that all the things you brought up are necessarily flaws; for instance, admins are as entitled as anyone else to go on extended wikibreak. Inactive admins don't do any harm to Wikipedia, and since there's no upper limit on the number of people who can be made admins, there's no harm in promoting admins who won't necessarily use the tools every day. Walton^One 09:26, 6 October 2007 (UTC)

View by Sxeptomaniac

My one thought is that we should consider doing away with automated edit counters. I feel they are often a crutch for editors who poorly research qualifications in RfAs. This would not stop all of the editcountitis problems, but it would force editors to go to the person's contributions page to find out approximately how many edits an applicant as, encouraging research into what the content of those edits is. (Perhaps an edit counter that purely gives percentages might be a reasonable exception, though.)

Users who endorse this summary

1. Sχeptomaniac^{χαιρετε} 19:51, 5 October 2007 (UTC)

View by Eiler7

This is more of a meta-comment. It is hard to think about this issue and evaluate possible solutions without having clearly defined the problem. Perhaps we can try to define the problem first and see if there is over 50% agreement that it is important. A clear definition would be something like "RFA is failing because it takes to long to promote editors" or "RFA is failing because backlog X is not being handled". Without knowing whether a definition with clear support exists, it is hard to think concretely about a solution.

Users who endorse this summary

1. Eiler7 21:03, 5 October 2007 (UTC)
2. John Carter 18:00, 6 October 2007 (UTC)

View by Anonymous Dissident

RFA is the process involving the making of new administrators, the users who have shown their dedication to and their interest in the project, and their knowledge of the policy which is integral to the project itself, and their experience as an editor and as a user. Such a process is never going to be simple, because what can result from it is certainly not simple - a new administrator, who can block people, delete pages, and protect pages. The process is going to have flaws. Different people are going to dislike different aspects of it, because it is a complex procedure, involving sometimes hundreds of people. It isn't broken. It is still functioning. The page is still there, and people are still submitting candidacies, and people are still being promoted, and some are not. The page is working generally as it was supposed to, but it has some flaws that truly don't go too far beyond the surface; flaws that should be repaired through improvement, rather than the tearing down of the process. We could end up diving headlong into a process much worse, when we could have worked on the improvement of the current process, that already has solid and substantial roots, and that has been proven to actually work. Reform is needed. Change is needed. But tearing down a functional, but admittedly flawed, process is not the answer, when a better idea has not really been found.

Views on WT:RFA

We first need to sort out the discussion page for this process. WT:RFA is a complete mess, and any attempts to sort out the problem(s) are dismissed quickly, even by those who complain the loudest about RFAs issues. Consensus on what to do needs to be achieved through polite discussion on WT:RFA before any proper progress toward the elimination of the problem can be made. Currently, we really are stuck in a rut on the process' talk page, with ideas constantly recurring each month, only to be gunned down. Maybe trial and error is the key, because discussion and disagreement certainly isn't doing anything. -- Anonymous Dissident^Talk 01:16, 6 October 2007 (UTC)

Users that endorse this summary

1. Me, I suppose. -- Anonymous Dissident^Talk 01:16, 6 October 2007 (UTC)
2. · AndonicO ^Talk 01:18, 6 October 2007 (UTC)
3. SamBC(talk) 11:00, 6 October 2007 (UTC)
4. Camaron1 | Chris 13:51, 6 October 2007 (UTC)
5. Chris B • talk 15:46, 6 October 2007 (UTC)
6. The current process is not perfect but it can never become perfect. Only small refinements can be made. Gizza^{Discuss ©} 22:58, 7 October 2007 (UTC)
7. I also endorse DaGizza's comment, above. нмŵוτнτ 03:15, 10 October 2007 (UTC)
8. - jc37 00:27, 11 October 2007 (UTC)
9. Nihiltres^(t.l) 14:50, 11 October 2007 (UTC)

Suggestion by Cacycle

Admins get only a few additional buttons and functions that make site maintenance easier (the "mop"). Because these functions can be abused, access has to be restricted. Just like in real life - you do not hire janitors or security guards and give them the keys if you do not trust them. At the same time, adminship is seen as a higher management position with representative function and Requests for adminship discussions often resemble more a presidential election than a simple job interview.

On the one hand, we have many trusted users that do a great job as vandal fighters who could clearly benefit from the admin buttons. On the other hand, we have many users that have grown into higher positions ("chief editors", "human resource managers", "politicians", "chief technical officers") without having the mop of adminship. I think that most problems with the current system arise from intermingling these two different concepts. Both jobs, janitor and manager, do actually not have much in common. One way out of the current misery might therefore be a clear and explicit distinction between both duties.

Just as an example: We could have "janitors" which are chosen solely by their trustedness and which can have their rights revoked upon misuse in a simple process - in the spirit of the no big deal doctrine. Independently, we could elect people to carry an official title of honor (such as "manager", "distinguished editor", or, of course, "administrator") in a process that resembles the current Requests for adminship discussions. Such a status would then be the user equivalent of a featured article :-)

No major changes would be needed: "Janitors" and "administrators" would be granted the system's "sysop" status and we would use the established RfA procedure to nominate "administrators". We only had to establish a "Request for Janitorship" (and a "dejanitorizing") procedure.

Users who endorse this summary

1. Cacycle 04:28, 6 October 2007 (UTC)
2. Interesting idea - maybe we should call it Featured Editor status. :-) Walton^One 09:27, 6 October 2007 (UTC)
3. I had thoguht about this earlier; it's actually quite a good idea. Editors which predominantly edit articles could get only the revert tools, whilst more deletion involved editors could get others to help them. It'd save people nit-picking one certain editing aspect (Such as edits the mainspace too much compared to Xfds), as editors would only run for one type of job. Could work. Spawn Man 09:38, 6 October 2007 (UTC)
4. Support. I believe that the admins currently are asked to solely perform a few too many duties. I'm not sure about "Featured Editor", though. Maybe instead retain the "admin" title for the upper rank, and give a title like "security and maintenance officer" to the lower rank. Qualifications for the lower status would probably be less frequently contested, and would allow the admins more time to perform the more difficult tasks. John Carter 15:41, 6 October 2007 (UTC)
5. Kind-of-support. I think that this is an idea that should be more thoroughly explored. I do know, however, that there's been a lot of resistance to the idea of more tiers of users before now, and there would need to be a lot of thrashing to define the two different roles. I'm not sure I'd view one as higher than the other, however, more seperate-but-equal. One might be harder to get, but that's just because it would have more potential trouble and fewer people would be suitable; it wouldn't make it more senior. I think it would be best if the jobs are to be split that neither be a prerequisite for the other, nor have a superset of the tools. SamBC(talk) 19:19, 6 October 2007 (UTC)
6. This was similar to an idea I mooted earlier about a sub-admin level without the block/unblock buttons and a much simpler process to obtain. I think there's some merit, but the Wikimedia developers might strongly disagree with me and for good reason. However I'll still endorse. Orderinchaos 20:47, 7 October 2007 (UTC) (Please see my clarification for required changes above - Cacycle 02:41, 8 October 2007 (UTC))

View by Spawn Man

Support votes are given waay too much weight when it comes to consensus - I don't really think a support vote of "Support - Per nom - User:Supporter" cuts the cheese, when oppose voters give so much more thought and time into delivering their rationale. I've always been a big believer that there should in fact be no supporting votes allowed - admins simply see a tally with 70% supporting and discard the 30% of possibly good objections. However, if supporting voters could come up with some way to provide a better rationale such as "This person should be an admin because they've done this here etc etc etc", it'd seem much more credible than "Per nom" - although this could be better, I think scrapping the whole supporting vote side of things is much better. The underlying problem is that people don't trust admins & 'crats - if they did, we wouldn't be questioning now about consensus. Unless Jimbo himself is promoting users (And even then there'd be some complainers), we won't be able to solve the problem and frankly, I don't think we ever will. Everyone has their own opinion which is very different from everyone else's. Spawn Man 09:33, 6 October 2007 (UTC)

Users who endorse this summary

1. Spawn Man 09:33, 6 October 2007 (UTC)

Response to view

• I don't really agree with this. We always need more admins, so in most RfAs the default should be to Support. An unadorned Support vote, or a "per nom", doesn't imply that the user is just piling on; it implies that they've read the nomination statement, investigated the candidate, read any opposes, and decided to agree with the nominator. Opposers, on the other hand, really need to clearly demonstrate why the candidate should not be an admin, and need to provide diffs to back up their argument (as I did in Kelly Martin's recent RfA, one of the few I have ever strongly opposed). Certainly, 30% of users making good objections should not be ignored, and usually someone with only 70% support will not pass (when such people have been promoted, it's usually been a total disaster). As to your second point: I don't trust the 'crats in general, because they have a track record of poor decision-making in controversial cases (e.g. Danny and Ryulong). And I trust Jimbo even less than I trust the 'crats; the fact that he appointed Kelly Martin to the ArbCom makes me question his judgment. The only people who should make decisions about promotions are the community as a whole, through a fair and consistent voting process. Walton^One 10:12, 6 October 2007 (UTC)

  I agree with Walton, generally. Sometimes the preamble, answers to questions and other discussion covers everything that a supporter wanted to clarify. Since there is no further reason to support (and support needs demonstrating) then "per X" is appropriate. LessHeard vanU 11:27, 6 October 2007 (UTC)

I agree, but if this is so, why do people constantly complain about the supporting votes and their lack of depth? As I said, I don't this issue will ever get resolved as everyone has their own views... We may have to scrap the whole thing and restart all over again or just get rid of RfA all together; the only real tool that's given that normal users don't have access to is the block button. Spawn Man 00:05, 7 October 2007 (UTC)

Just look at who it is that complain about "lack of depth of support votes". — Dorftrottel⁠, was: aldebaer 14:43, 7 October 2007 (UTC)

View by Mailer Diablo

RfA doesn't break by itself; People do.

You can tear down the building and rebuild a different one, but so long the occupants and the management treat it with disdain it's only a matter of time the cycle begins all over.

Users who endorse this summary

1. Mailer Diablo 05:09, 6 October 2007 (UTC)
2. Walton^One 09:29, 6 October 2007 (UTC)
3. Pedro :  Chat  14:51, 6 October 2007 (UTC)
4. Iridescent15:01, 6 October 2007 (UTC)
5. Chris B • talk 15:44, 6 October 2007 (UTC)
6. A brilliant point--Phoenix 15 18:27, 6 October 2007 (UTC)
7. Yes, as long as we treat adminship as something very special this will occur. Humans love status. The only way to fix this is to make adminship less special. - cohesion 18:39, 6 October 2007 (UTC)
8. Captain panda 13:23, 7 October 2007 (UTC)
9. Very important point indeed. — Dorftrottel⁠, was: aldebaer 14:41, 7 October 2007 (UTC)
10. Orderinchaos 20:45, 7 October 2007 (UTC)
11. Indeed Gizza^{Discuss ©} 23:02, 7 October 2007 (UTC)
12. Sjakkalle (Check!) 06:43, 8 October 2007 (UTC)
13. Mailer reached the core of the problem. @pple complain 09:46, 8 October 2007 (UTC)
14. Sχeptomaniac^{χαιρετε} 15:20, 8 October 2007 (UTC)
15. This is certainly true. нмŵוτнτ 03:21, 10 October 2007 (UTC)
16. Phgao 15:49, 10 October 2007 (UTC)
17. Indeed. O₂ (息 • 吹) 04:00, 15 October 2007 (GMT)
18. Was there ever a wiser undeliverable error message than Mailer diablo? - Che Nuevara 19:47, 15 October 2007 (UTC)
19. Longhair\^talk 08:36, 16 December 2007 (UTC)

An analogy / view by jc37

Imagine someone standing in front of a tree. They have a spoon, and are attempting to cut a branch off the tree.

Well it's moderately sucessful, but after teaching this to others, the others begin to complain that the spoon is "broken", because it's not as effective removing braches as they might like.

Someone else comes up with a plan to modify the spoon so that it can cut similar to a knife, but is sent away by naysayers who claim changing the tool isn't the problem, the people complaining are. Calls of: They aren't using the tool effectively. This is the way we've always done it, and it seems to work just fine. If we change now, the new version may not work, and then we'd be back where we started, etc.

Are such views "correct? Perhaps. But then that would leave us in the Stone age, and we'd never make it to the Bronze Age, much less the Iron age, if we follow that line of thinking. Tools must be tested continually. What may have worked well in the past may not be the best choice of tool for the future. I doubt anyone on this page surfs the web with a 286 computer running DOS, for example. People will claim whatever they want regardless of what the tool actually in use is, they are the constant in this equation. So let's work on what we can, and take a look at the tool and see if there are ways to improve it.

Users who endorse this comment

1. - jc37 18:44, 6 October 2007 (UTC)
2. Carcharoth 22:39, 6 October 2007 (UTC)
3. Good analogy. Adrian M. H. 22:50, 6 October 2007 (UTC)

Response to view

• To modify your analogy, I would say that it is more like people who, upon being handed a saw, attempt to cut with the back of the blade, or perhaps use the correct edge, but in a way that causes it to cut poorly or bind. The saw works well enough in the hands of those who know how to use it properly, but there still may be ways to modify the saw to make it more effective, easier to use, or safer. However, ill-considered modifications could only serve to make it even less useful for all. Sχeptomaniac^{χαιρετε} 16:15, 8 October 2007 (UTC)

View by Cohesion

This is a suggestion for a different RFA process. I admit I don't know much about the inner workings of RFA, I nominated myself, passed, and very rarely go there because it seems rather ugly to me. So, here's the process:

1. People self request, saying why they need the tools. (there is no need to nominate others, if people want access to the tools they can ask. The word "nomination" sends the wrong message, it's not an award. It's certainly not an award that people should give others.)
2. A bureaucrat looks at the requests and the person's history, assures themselves they are not dangerous, and flips the switch. Perhaps at this step people can sign up to watch the new admin and help.
3. Some time period later there is a retrospective discussion. Did the person cause any problems with the tools? Was this a good decision? If not then de-admin. (Obviously if there is an intervening serious issue the bureaucrat can de-admin immediately)

This system will make adminship less of a big deal. It will also focus the discussion on how the person is using the adminship tools, rather than how many featured articles they wrote, or if a project supports them, or any number of other unrelated metrics.

Users who endorse this summary

1. cohesion 18:51, 6 October 2007 (UTC)
2. Could it really be this simple? :-) Carcharoth 22:41, 6 October 2007 (UTC)
3. Not a million miles away from my take on Warlordjohncarter's proposal. With a bit of tuning and toning down (we don't want it to be too simple because that could make a net that has very large holes), I believe that this might just work. Adrian M. H. 22:47, 6 October 2007 (UTC)
4. Davewild 11:23, 7 October 2007 (UTC)

Comment on this view

1. Why a bcrat will have a better sense than a community of editors? ≈ jossi ≈ (talk) 23:05, 6 October 2007 (UTC)
   1. Well, step 2 isn't the final outcome, it's just a b-crat b/c they are the ones that can make admins. The retrospective would still be by the whole community. The "step two" is not much of a decision, just see if they are dangerous or not. Not whether they would be a good admin or not. - cohesion 01:09, 7 October 2007 (UTC)
2. Although I think adminship should be less of a big deal, I don't think this proposal would work, unless there was an easy process for de-adminship (which carries with it its own problems). Anyone with an agenda to push could make themselves look squeaky clean for their "trial period" before the retrospective, then go on to manipulate whatever they want. rspeer / ɹəədsɹ 23:43, 6 October 2007 (UTC)
   1. This is a real problem, maybe you could have a follow-up a year later or something. You wouldn't want the initial trial period to be too long because if they are bad you would want to shorten that. On the other hand, making it too short means it is trivially easy to "fool the process". Maybe have an initial 2 week trial, followed by a one year follow-up. Where at both points de-adminning is possible. - cohesion 01:09, 7 October 2007 (UTC)
3. I strongly agree with the first point, of self-noms only. Then the other would-be-noms could point out the requestee's good attributes/history in the "support" section, per the suggestions far above that 'just a support vote' isn't really substantial enough. I'm not sure about the other points. --Quiddity 08:38, 8 October 2007 (UTC)

View by Amarkov

So, we want to replace a process that squelches some valid concerns, overaccounts for others, and sometimes arrives at the wrong result. That's fine, by itself. But the only other community process we have squelches some valid concerns, overaccounts for others, has no clear outcome, and still sometimes arrives at the wrong result. And our non-community based process squelches any valid concerns not held by people who have been here for sufficiently long, overaccounts for any held by the in group, is decided by the voting that we're trying to get rid of in the first place, and still sometimes arrives at the wrong result. Why exactly are the latter two preferable?

Users who endorse this summary:

1. Amarkov moo! 22:01, 6 October 2007 (UTC)
2. Hell yeah. Walton^One 22:45, 6 October 2007 (UTC)
3. Very yes. Improve what we have instead of replacing with as bad alternatives. LessHeard vanU 23:10, 6 October 2007 (UTC)
4. Indeed. Xoloz 12:18, 12 October 2007 (UTC)

View by Arknascar44

See also: Wikipedia:Requests for comment/Wikipedia:Requests for adminship § View by Jonathan

RfA is not broken. It does not need to be replaced by another process, but it does need to be reformed. Therefore, I have a proposal that, hopefully, doesn't sound too radical

• When a user is nominated (or self nominates) for adminship, the RfA template displays with no oppose, support, or neutral sections. Instead, there is just one header, entitled "Discussion". For a period of 2 days (more or less, this can be changed) users participate in the RfA by listing any concerns they have about the user, or any significant contributions the user has made, thus creating a back and forth discussion.
• After this period of time, "Support", "Neutral", and "Oppose" sections are added to the RfA subpage. Then, users can support, oppose, or remain neutral based on the discussion and the nomination statement.
• A 'crat then closes the RfA by judging consensus; in other words, if 500 editors oppose the RfA without giving a reason, but 1 supports with a well-written, well thought out statement, the RfA passes. After all, RfA should not be a vote. Sorry, Rspeer :) Of course they do this already *user slaps themself in the face*.

This system allows for users to engage in discussion before voting, to ensure that when they do vote, they will have very good reasons for doing so. Users who didn't participate in the discussion should read through it carefully before voting.

Users who endorse this view

1. ( arky ) 01:41, 7 October 2007 (UTC)
2. I like the idea, and had been wishing for something along those lines. It gives people time to actually examine the candidate before the pile-on. But I'd say your 500 vs. 1 example of consensus is ludicrous and may cause people to be wary of your proposal; I think bureaucrats do well at judging consensus now, so your proposal doesn't need to include a recommendation for how they should do it anyway. Keep it simple. rspeer / ɹəədsɹ 06:42, 7 October 2007 (UTC)
3. The discussion period might be worth considering, as the current format often discourages in-depth discussion/disagreement. Sχeptomaniac^{χαιρετε} 17:09, 8 October 2007 (UTC)
4. That makes a lot of sense; people do not like it when their comments are challenged, this may lead to some more constructive discussion. Neranei (talk) 17:37, 8 October 2007 (UTC)
5. The idea of adding 2 days of discussion to the beginning of an RfA seems to be about the best idea for reform I have heard yet. ((1 == 2) ? (('Stop') : ('Go')) 17:41, 8 October 2007 (UTC)
6. What's this? A proposal that makes sense? shoy 18:53, 9 October 2007 (UTC)
7. Some form of this was discussed awhile back. I like the idea of the first 2-3 days being discussion only. Stops "drive-by" "voting", for one thing. - jc37 00:27, 11 October 2007 (UTC)

View by Durova

At Wikipedia:Admin_coaching/Requests so many requests have gone unfulfilled that the page now divides into sections. 150 would-be sysops are seeking guidance there. Category:Wikipedia administrator hopefuls has 963 participants. There is no shortage of people willing to push Wikipedia's mops. All that many of them lack is a small amount of mentorship. It saddens me to see so many people waiting at the gates while we debate whether to repaint the key. Want more sysops? Go coach some good editors. Several of my coaching students recently got their mops so I've got a couple of open slots right now. Interested Wikipedians are welcome to drop me a line. Durova^Charge! 08:14, 7 October 2007 (UTC)

Does the mentoring continue after people pass RfA. Would you be concerned if a candidate became independent and reckless after passing RfA (thinking they can now push the boundaries), and if so, what would you do about it? Carcharoth 13:56, 7 October 2007 (UTC)

In my case it does. I'll continue to advise them after they're sysopped, usually by e-mail when they seek it but occasionally on the administrative boards when a potential problem comes up. In one instance an admin I'd coached got into a disagreement and came close to the WP:3RR limit. My advice was to step back before things heat up and try an article content WP:RFC to bring in fresh perspectives.

Generally I work with people for a while before nominating them. My foremost concerns are that these people want the mop for the right reasons and demonstrate a similar mix of traits to what I've seen in successful administrators. I seek people whose interests resemble my own because my volunteer work concentrates in a few undermanned areas. Durova^Charge! 22:56, 7 October 2007 (UTC)

Possibly irrelevant comment by User:Moonriddengirl

I agree with Durova that mentorship would be invaluable and have looked at that list and plan to offer my help once I've been in the role a little longer. I think mentorship would be even more valuable for new administrators. The New Administrator's School is great, but there have been many times in my first month of service when I've wished that I had somebody specific helping me figure out the tools and how to deal with some of the trickier individual issues. (For example, I'm working to expand my grasp of image policies because that area of CSD has a perpetual backlog.) I've pestered a couple of more experienced administrators who have so far all willingly pitched in, but I hate to keep asking. :) If new administrators were paired with experienced ones, it might eliminate some of the unintentional abuses of the system and, since there would be another pair of eyes reviewing the new admins work, allow for easier correction of accidents. It might also alleviate some of the concerns people face in supporting admin candidates. Obviously, this isn't going to be of use where clear concerns of abuse of the tools exist, but slight fears of inexperience may be relieved. --Moonriddengirl 13:53, 7 October 2007 (UTC)

View by LessHeard vanU

Perhaps too much attention is given to the process of gaining the buttons, and not enough to the post promotion activity. It is only after being given the mop that it will be determined if the admin is able to use the it appropriately, in part or general. RfA is a method by which people attempt to predict the future consequences of an editor acquiring certain powers; it therefore doesn't matter if this procedure is maintained, is "improved" or is replaced - it is only when the editor is invested can it truly be found if they are what was required. Desysopping (be it easy or hard a process) will only remove those who grossly violate the trust placed in them, but there is no apparatus to monitor and mentor admins who are not fulfilling their roles in the manner expected of them.

Whatever is decided upon in promoting admins (and in removing the worst) there will still be admins who will not reflect the trust the commmunity invested in them. LessHeard vanU 10:47, 7 October 2007 (UTC)

Users who endorse this view

1. Indeed, and that is why admins should be accountable to the community and why the community should be able to express its distrust in an admin. Melsaran (talk) 13:40, 7 October 2007 (UTC)
2. Agreed. My thoughts on partially addressing this right above, at Wikipedia:Requests_for_comment/Wikipedia:Requests_for_adminship#Possibly_irrelevant_comment_by_User:Moonriddengirl :) --Moonriddengirl 13:55, 7 October 2007 (UTC)

   That would be the comment approximately 100-125mm (depending on monitor) above this one, yes? Just checking! ;~) LessHeard vanU 16:52, 7 October 2007 (UTC)

   That's the one! :D --Moonriddengirl 17:23, 7 October 2007 (UTC)

View by Jonathan

I don’t think the page should be deleted, but I do think the process should be changed in a major way, as the following:

1. The user nominates a user (or self-noms) as always.
2. Then, there is a 4-day discussion only period.
3. Next, the RfA can have votes for 3 days to finish off the week-long RfA period.
4. Then, the bureaucrat judges by consensus.

It is generally what Arky said, but I have tweaked the days and made it a little shorter. jonathan (talk — contribs) 17:08, 7 October 2007 (UTC)

I strongly oppose this - a three day window is much too short — iridescent (talk to me!) 19:23, 7 October 2007 (UTC)

I was actually debating whether or not to switch the voting and discussion times while I was writing this. jonathan (talk — contribs) 23:38, 7 October 2007 (UTC)

Users who endorse this view

1. I'd support this too, except I'd prefer to switch the days: 3 discussion, 4 voting. rspeer / ɹəədsɹ 17:46, 7 October 2007 (UTC)
2. Oops. I forgot to endorse this myself...jonathan (talk — contribs) 18:24, 7 October 2007 (UTC)
3. 2 or 3 days discussion is fine, 4 too long, and 1 not enough. If pressed, I think I prefer 2 and 5, since 5 is the current length of XfD discussions (and if it's good enough for content discussions...) - jc37 00:27, 11 October 2007 (UTC)

View by Eagle 101

How do you have a discussion with 60+ people? There is a limit to where we lose the ability to have a pure discussion, and are forced to tally and abstract folks views. Generally the abstractions on RFAs result in those supporting, and those opposing because of XXX reason. (generally opposes have reasons, or say per someone else). (note this format is really just me plopping my idea down and seeing how many people agree with it, not really "discussing" the matter)

Lets face it, no 'crat is going to promote someone with 1/4th of the community supporting, and the other three quarter opposing. Same for the other way around, if everybody supports someone... is a 'crat really going to not promote. Generally the system that we have now is very effective in the obvious cases. The part of the community that participates on RFA is fairly good at identifying problems with potential admins, and it really is a rare event that any people who are promoted are ever de-sysopped. The only cases where there is much dispute is when a request for adminship falls into that zone where a significent portion of the community opposes and a significant proportion supports. Generally this zone is around ~66% to ~75% (Please do note I'm casting a wide net here, these numbers could be different, I just chose those between 2/3 and 3/4). The 'crats promote most of the ones near 80% and reject most of those near 66%, but there are some edge cases both ways. Alas this is the crux of the matter.

I propose that we recognize that the system in place now is largely never going to change, and with its success rate at weeding out bad candidates, really does not need significant change. No matter what system we make use of there will always be arguments over the "middle zone", those candidates between about 66% and 75%. Could we improve the system now? Probably, but the system as far as results is not fundamentally broken.

As an aside, I would be interested in cases where this system has clearly failed. Please note I'm talking about sysops that fail their first RFA (preferably a "close" failure), but get promoted in their 2nd one (and are still a sysop today), or sysops that pass their RFA and later get desysopped. (either those that should have been promoted and those that should not have been).

Failure Cases

I know this is out of jive with the RFC format, but we need some way to collect where this system has obviously failed

• Please submit failure cases here, please provide enough links and explanation to make this useful.
• To help out, anything listed at WP:DESYSOP#Cases (not the proposal, but the research done there) is a possible failure case. Of real interest is anyone with more then say 3/4th of the RFA watching community supporting them that got desysopped, but a listing of them all with notes on the RFA's would be interesting as well. —— Eagle101^{Need help?} 21:18, 7 October 2007 (UTC)

Users who endorse this view

• Obviously I do ;) —— Eagle101^{Need help?} 20:59, 7 October 2007 (UTC)
• Yes. Any method of organizing input from any large number of people will resemble voting, and when the choice is binary (as is the case here), it will be barely indistinguishable. Once we accept this, we can decide what reforms may be needed. -Amarkov moo! 21:02, 7 October 2007 (UTC)
• A "discussion" with 60+ people without keeping any tally would create mayhem. Gizza^{Discuss ©} 23:06, 7 October 2007 (UTC)
• I think Eagle has noted some very important things here. -- Anonymous Dissident^Talk 23:44, 7 October 2007 (UTC)
• Ditto with the current system not changing, owing to the diversity of everyone's opinions. O₂ (息 • 吹) 23:49, 07 October 2007 (GMT)
• Endorse. I do not notice consensus on this discussion even on the simple question of whether RfA promotes too many and therefore lets undesirable users get access to admin tools, or promotes too few and therefore denies the tools to people who would help the encyclopaedia. Sam Blacketer 08:35, 8 October 2007 (UTC)
• Endorse. I think the closing RfA comments in successful cases should be a lot like that when you pass a driving test: "you have reached the standard required to continue learning by yourself without having an instructor in the car with you". Carcharoth 10:07, 8 October 2007 (UTC)
• Endorse, well stated. Haven't seen a meaningful discussion with more than 60 people participating. Kusma (talk) 12:55, 8 October 2007 (UTC)
• Well said, the current system does compromise between a difficult-to-work-discussion, and a simple vote quite well. Camaron1 | Chris 20:14, 8 October 2007 (UTC)
• Endorse as per my previous comments. ≈ jossi ≈ (talk) 21:00, 8 October 2007 (UTC)

Comment on above view

There are only four general points of discussion in a RfA; the two prime arguments are "I trust this editor with the tools" and "I don't trust this editor with the tools", the latter usually with more examples than the former, and the two sub arguments are those opposing one of the above (again with a preponderance of examples given by one side). Everything that does not devolve to one of the four arguments is just so much noise. LessHeard vanU 19:38, 8 October 2007 (UTC)

View by Zero1328

Now, I've been out of the loop for some time, so I'm not saying whether it's failing or not, but I think some people feel it's failing because they don't understand it. It's not very clear on what comments are right or wrong, what shouldn't be used, etc. Some people learn by copying what others do, not by reading the instructions. Go to any RFA, and try counting how many people you think have read WP:AAAD. This essay demonstrates an ideal RFA discussion. All the other related articles show ideal discussions in other areas, too. Why do you think people haven't read it? Do you know what it is? Have you even heard of it? One possible solution would be to clarify and emphasise this essay to everyone. Rather than changing the process, change how it's presented to newcomers. If it's not easily understandable, people may either be turned away from Wikipedia's processes or try other forms of learning, like, perhaps, copying how other Wikipedians work.. It's a vicious cycle, and it doesn't apply to just RFA. - Zero1328 _Talk? 02:00, 9 October 2007 (UTC)

Proposal by Lara_bran

I propose minor change for current version of RFA, since it is already working well. In "support" and "oppose" sections, it should be divided into votes by users who are already admins and non-admin users. That is, those who are already admins should vote (with or without comments) in one section and other users should vote in separate section, so that votecount will be separate. Same weightage will be given though. This gives better idea of support, and would help recognize candidate's support. This is because lately i noticed, users who participate in XfDs get more admin votes, and who edited more in article space get more user votes. Thanks. Lara_bran 10:43, 9 October 2007 (UTC)

Though admins are minority when compared to registered users, others hardly vote in a RFA. This separate section will encourage them to participate in RFAs. Thanks. Lara_bran 10:47, 9 October 2007 (UTC)

I don't agree with "others hardly vote" at all, and don't know where you're getting that from. Looking at AA's RFA, as a current one with an average level of participation, of the first 20 votes (discounting one from a now-deleted account) four are from admins and 16 from non-admins. That's hardly "non-participation". — iridescent (talk to me!) 15:54, 9 October 2007 (UTC)

Admins aren't consistently intelligent. Eight admins supported the RfA for Captain panda (who listed "voting in rfa" as one of his best contributions?????)

The lower the bar, the more naive editors will become admins, and the more naively they then collectively vote; it's a vicious circle, which lowers the average admin respectability. I think the standards should be Higher, and rfa should be tougher, if anything is going to change. 24.68.83.117 16:57, 9 October 2007 (UTC)

It has been heavily argued in the past we need more admins, not less; and more that standards have got higher, not lower. I am not sure how supporting CaptinPanda's RFA makes you unintelligent - there are more important things when looking at an RFA candidate than there answers to (optional) questions. Camaron1 | Chris 17:33, 9 October 2007 (UTC)

So why should we separate admins and non-admins? An admin's vote/comment is not more valid than a non-admin and vice versa. They have equal weighting and only their reasoning might change that. I personally think that a large percentage of the voters would be admin hopefuls. I remember stepping up my participation in RfA before my own RfA because many voters like to see participation in RfA. Gizza^{Discuss ©} 22:17, 9 October 2007 (UTC)

I agree admin-aspirants vote, but there needs more canvassing for RfA. If you consider me, a non-admin, sometimes my known users apply for RfA but i miss them. Separate section for non-admins would certainly an encouragement for non-admins to vote, also note that it has no disadvantages. Also posting/canvassing RfAs in some common places such as community portal or RfC list(policy) would result better participation. Lara_bran 04:51, 10 October 2007 (UTC)

View by User:Dustihowe

I feel that no matter what we do, not everyone is going be happy with what the changes are. We cannot keep chaging our policies because someone is unhappy with them. I feel that we should keep our policy on this the way that is unless someone has an extravagent idea that everyone agrees with. Then, we might change the Rfa policy. Dustihowe 17:42, 9 October 2007 (UTC)

Users who endorse this view

• LessHeard vanU 20:21, 10 October 2007 (UTC)
• User:Dustihowe I guess I better put my name here too!!!!

View by User:Eye of the Mind

As much as I believe that the RFA system has to be changed, in however large or small a way as you like, I feel this is not the way to go about it. There are currently something in the range of 40 "views" on this page. It is impossible as far as I can tell to come to any sort of a consensus in this way. This page needs to be archived and then, on a new page a "discussion," needs to be started. Without everybody jumping to give there, opinion. Things need to be proposed and quickly rejected or supported. No one wants to have to look through a 40 plus section page. --Eye of the Mind 03:43, 10 October 2007 (UTC)

Users that endorse this summary

1. Eye of the Mind 03:44, 10 October 2007 (UTC)
2. Banana 04:09, 10 October 2007 (UTC)
3. LessHeard vanU 20:22, 10 October 2007 (UTC)
4. User:Dustihowe AMEN!!!!!! Dustihowe 16:09, 11 October 2007 (UTC)
5. · AndonicO ^Talk 02:18, 12 October 2007 (UTC)

Comments

Fair comment. Probably the next step is to re-factor those views that have had useful support into a discussion document. I think there are a few points of consensus: some form of de-admining, possibly some differentiation of roles, some streamlining of the voting system to remove some systematic biases. Spenny 16:19, 10 October 2007 (UTC)

Seems like a good idea to me. --Rocksanddirt 18:09, 10 October 2007 (UTC)

View by PrestonH

I feel that requests for adminship standards are getting higher each time a fellow admin slips up or goes rogue (which in a long term run, may slow down the rate of producing and promoting admins). Compare the standards right now and back in 2005. As you can see, the standards were a lot different and there were less conflicts then. More people keep complaining about the request for adminship either because of the people's comment, how it works, unreasonable standards etc. I believe this RfA process is getting to the point where wiki-drama and conflicts occurred. Tensions tend to rise and on occasion, good contributors leave as a result of this mess. We need to ask these two questions ourselves before submitting our opinion. Do I trust this user with the tools? Will he/she misuse the tools? I believe that all it takes to submit an opinion. We don't need to go to this unnecessary rambling conflicts.--PrestonH 03:27, 11 October 2007 (UTC)

Users who endorse this view

• Exactly. All this article writing stuff is ridiculous. Cheers,JetLover (Report a mistake) 03:37, 11 October 2007 (UTC)
• --Mark (Mschel) 17:12, 13 October 2007 (UTC)

View from editor Thomas H. Larsen

In my opinion, RfA is bent, if not broken. The "Wikipedia is not a democracy" policy statement exists for a reason. I am seriously concerned over the method that RfA currently uses to decide the community's administrators.

As a gift culture, Wikipedia (should) function on a method of rough consensus, not voting and polling. Is there any reason that RfA could not be implemented as a simple discussion? Look at RfAs now. Virtually no discussion takes place with the candidate before participants vote, which means that some bad candidates get through and some good candidates do not. I suggest that a more wiki-like rough-consensus-and-discussion approach be made to RfA.

— Thomas H. Larsen 05:07, 11 October 2007 (UTC)

Users who endorse this summary:

• — Thomas H. Larsen 05:07, 11 October 2007 (UTC)

View by User:Tony Sidaway

A simple proposal: Just about all experienced, regular editors of English Wikipedia should be given the sysop bit on request unless there is a good reason for them not to have it.

Comments on this simple proposal

• Simple proposal is simple. It's a shame it won't work this way :-( --Deskana (talk) 16:25, 12 October 2007 (UTC)
• It works in theory, in practice, you would need a sizeable number of people running around sysopping and desysopping people who abuse the bit, so it just shifts the divisive problem somewhere else. Maybe not quite as many, but then that makes the ruling elite all the more "elite". Nick 16:33, 12 October 2007 (UTC)
• Endorse: all admins should have +sysop and +desysop, adminship should be given even to some who don't ask for it, and we should massively expand the admin corps. This would nicely get rid of perceptions of "elitism", which are probably justified given the stupidly small number of admins. Moreschi ^Talk 16:36, 12 October 2007 (UTC)
• Simple is good. But this puts a lot of power in the hands of whoever is deciding what a "good reason" is. Besides, quite a few people pass RfA even though one or more of those opposing has given a good reason why they shouldn't - just the supporters disagree or feel that bad point is mitigated by the candidate's other qualities. This could end up making it more difficult for people to become admins, which I think is the opposite of what Tony was intending... WjBscribe 16:39, 12 October 2007 (UTC)
  • Well with all the time they'd save in (not-)counting (non-)votes, the bureaucrats would have to find something to fill their empty days. But seriously what I'm considering at present is an almost across-the-board sysopping of thousands of good editors who still edit regularly and have an unblemished record. "Hasn't any edits in Wikipedia space" and "Has expressed suspect views on copyright policy" wouldn't be good objections. "Has abused his editing powers" would be a very strong contra-indicator, however, as someone likely to abuse other powers. --Tony Sidaway 17:01, 12 October 2007 (UTC)
    • Find a crat willing to do it, and off we go. Of course, I suspect many crats may be extremely reluctant to do this unless they had a way to undo it when needed. Mistakes are inevitable. Friday (talk) 17:05, 12 October 2007 (UTC)
      • Why not abolish crat status and give all admins +sysop and +desysop? Make life a lot easier. Moreschi ^Talk 17:06, 12 October 2007 (UTC)
        • I already suggested that at WT:RFA. Sadly, it got shot down in flames (unsurprisingly, it was strongly opposed by the existing bureaucrats). Walton^One 17:35, 12 October 2007 (UTC)
• I simple cannot support this idea. I simple don't think that enough of the people given the bit will use it responsibly. I think it would result in a lot of de-sysoping and a lot of damage along the way. I don't mean click click its fixed damage, I mean bad blocks and disruption to NPOV in the articles, people leaving over these thing. Lowering the standards for adminship is not the way to go. 1 != 2 17:07, 12 October 2007 (UTC)

  To take one of your suggested abuses of sysop power: if an editor hasn't used his editing abilities to push a point of view, why would he be likely to use his sysop abilities to do so? --Tony Sidaway 17:10, 12 October 2007 (UTC)

  Because before he had no advantage. 1 != 2 17:11, 12 October 2007 (UTC)

  Think it through: if there are now many thousands of admins, he still has no advantage. --Tony Sidaway 17:42, 12 October 2007 (UTC)

  What about new users who have not become admins yet? Or are new users going to be admins automatically? I think NPOV will suffer. 1 != 2 17:46, 13 October 2007 (UTC)

  I think the qualifications "experienced" and "regular" in my original proposal take care of that objection. --Tony Sidaway 19:25, 25 October 2007 (UTC)

• If the only worry was admin tool abuse this might work. However, they'd be huge potential for misuse of admin tools when inexperienced users suddenly have access to blocking, deletion, protection, etc. In an ideal world, I'd love to see this happen, but Wikipedia isn't that world. Chaz ^Beckett 17:17, 12 October 2007 (UTC)

  Note that I did see that the proposal mentioned giving the tools only to "experienced" users, but without some sort of review, how to determine this? That puts us right back in the position of having an RfA process. Chaz ^Beckett 17:21, 12 October 2007 (UTC)

  I would think a proposal for experience would be something really vague and low. Three months and greater than 250 total edits or something. Or something very high (5,000 edits and one full year). —Preceding unsigned comment added by Rocksanddirt (talk • contribs)

  The problem is that neither edit count nor editing duration really correlates with whether an editor knows what the heck they're doing. That's pretty much the goal of RfA and it's just not possible to boil it down into X edits and Y months editing = +sysop. Chaz ^Beckett 17:29, 12 October 2007 (UTC)

  I'm trying to get away from elitism, not raise it to new levels. --Tony Sidaway 17:42, 12 October 2007 (UTC)

  Yeah, I get that. Just to be clear, the comment above mine wasn't written by me. I was replying to it. Chaz ^Beckett 18:30, 12 October 2007 (UTC)

  A proposal I made a while back went like this: chap wants adminship, asks bureaucrat. Bureaucrat satisfies himself that this is a bona fide request by an experienced, regular article editor. A few days discussion. Bureaucrat looks and see if any serious problems have been raised in discussion (blocks, serious misconduct). Makes decision, promotes if okay. --Tony Sidaway 18:40, 12 October 2007 (UTC)

• I like it - How many of the current admins use the admin tools very often? Some do admin stuff all the time, some do not. This would really cut down on the number of requests for admin actions, and then the AN and AN/I boards can officially become what they sort of are which is the 'complaints department'. Maybe only give it to folks who haven't had some sort of sanction? but let others continue with an RfA process to have a discussion of reliability/trustability? --Rocksanddirt 17:22, 12 October 2007 (UTC)
• Imagine a Wikipedia in which an RfA is going to the bureaucrat noticeboard and requesting the rights, getting your contribs reviewed, and after a bit of chatting between crats, you either get it or you don't. Wow. We should probably make sure the candidates are trained, through coaching or what else. We might need some more crats if we see a surge in requests. J-ſtan^Talk_Contribs 18:44, 12 October 2007 (UTC)

  I would actually think that maybe not even asking for it...for the first round of this, just let the 'crat's run down a list and assign the features/tools. You log in next and there you are. --Rocksanddirt 19:17, 12 October 2007 (UTC)

• This proposal goes into the same direction as my proposal above for a low threshold to give 'sysop' status to "janitors". At the same time I propose to keep and to split off the "higher management" chores of current administrators. I would appreciate your comments on this above. Cacycle 20:26, 12 October 2007 (UTC)
• I agree with the principle as stated, but the devil is in the details. There are some excellent experienced editors who regularly get rather heated in discussions, and I would fear they might extend their heat to blocking, deletion, or both. There are a number of editors who have created multiple featured articles about whom I am regularly thankful that they do not have these admin tools. --AnonEMouse ^(squeak) 20:33, 12 October 2007 (UTC)

  I strongly agree with this, and LessHeard vanU below. --Quiddity 17:43, 13 October 2007 (UTC)

• Yes! This will put an end to editcountitis and other assorted silliness once and for all. szyslak 21:04, 12 October 2007 (UTC)
• Vacuously true. Agreeing on whether a given reason is good is the problem. —Cryptic 21:20, 12 October 2007 (UTC)
• This is what happened originally, all registered users were the then equivalent of sysops. It was found that a method of determining who should have those responsibilities and "powers" was needed as the then system was found to be open to abuse. That was back when WP was small and people knew each other. Wikipedia is now much larger, as are the consequences of universal suffrage regarding access to the buttons. It is a wonderful ideal, but not practical. LessHeard vanU 21:46, 12 October 2007 (UTC)
• I like this idea, but I'd prefer if only bureaucrats had the +desysop tool. Maxim_{(talk) (contributions)} 17:45, 13 October 2007 (UTC)
• I'd argue that this is actually the system we currently have. It's just that people disagree over what constitutes "experience," "regular," and "good reason not to" and RFA is the forum where that plays out. You can delete the RFA page and people will continue to disagree over those fundamental issues, and so you'll have the exact same arguments and exact same problems. --JayHenry 22:41, 13 October 2007 (UTC)

  Oh the current system is that nobody gets the bit unless there is extremely strong, provable consensus that he should get it. That is very different from my proposal, which says nothing about provable consensus, strong or otherwise. --Tony Sidaway 21:12, 16 October 2007 (UTC)

• I'm absolutely certain this would push edit wars to the new level of admin wars. Horrible idea. Gray62 14:50, 14 October 2007 (UTC)
• This is one of those ideas like moving the U.S. to a purely libertarian system. In the long run, it may be stable, and the overall harm caused may be less than the existing system, but the transition will be a killer - there will be months (weeks, at least) of really awful wheel wars. Imagine giving the admin bit to the experienced balkan-nationalist pov warriors. Argyriou (talk) 20:06, 15 October 2007 (UTC)

  Obviously no edit warrior would get the bit under a system like this. --Tony Sidaway 21:09, 16 October 2007 (UTC)

• I agree with the basic principle; all experienced editors with a record of good conduct should have a chance to be administrators, and to help out with the various admin backlogs which need tackling. However, I think that the only way this will happen is if we implement an effective community desysopping process. People will be much more eager to give out the admin tools if we know they can be easily revoked. Under such a system, good candidates like Dihydrogen Monoxide and Agueybana might have passed RfA. Walton^One 13:25, 17 October 2007 (UTC)

  As I've remarked in the past (and Friday, who has also proposed a desysopping mechanism for the same reason, now appears to agree with me) when it's necessary to desysop someone it happens within ten minutes. For this reason I believe that the desysopping proposals are misguided. We've put admins on a pedestal: what better way to remove the pedestal than to give the bit to just about everybody who asks for it? --Tony Sidaway 19:05, 22 October 2007 (UTC)

• I think this is a great idea. It would also reduce the number of people who's only goal seems to be becoming an admin. If you don't get glory from it, then the only reason for requesting it is to be able to contribute more effectively. I do think though that it would be necessary to be able to remove adminship very quickly. --Mark (Mschel) 19:48, 22 October 2007 (UTC)

  It is very easy indeed. You just pop into the wikimedia stewards channel on freenode IRC (irc://irc.freenode.net/wikimedia-stewards), point them at the problem (the username and wiki is sufficient because then they can look at the damage being caused by examining the admin log) and then they perform an emergency desysop. --Tony Sidaway 00:20, 24 October 2007 (UTC)

  There are two kinds of problem admins. Those who need to be desysopped quickly - admins on a rampage, wheel-warriors, that sort of thing. For that kind of problem, immediate desysopping by the stewards is appropriate, and sufficient. However, there is the other sort of admin that should be desysopped - one who doesn't create a lot of destruction quickly, but who pushes the limits just a little too far a little too often. The kind of admin who blocks new users who are being disruptive, but the admin is in a content dispute with the user. Or who tells users that he doesn't have to explain his actions when he blocks people or protects pages. For that kind of abuse of admin powers, calling in the stewards is inappropriate. Some sort of community discussion is necessary, to determine if the admin in question really is overstepping the bounds, and if desysopping or some lesser sanction is appropriate.

  I think that overall, admin behavior has gotten better in the past few months, and at the very least, the idea that possibly controversial blocks should be discussed on AN/I has helped a lot. But there are still cases - probably one or two a week - of admins blocking in really inappropriate cases. If there are to be many more admins, there will be more such behavior, and some of the offenders will just not get it and need to be desysopped. But unless we create a lot more bright-line rules, the stewards won't be able to intervene.Argyriou (talk) 00:43, 24 October 2007 (UTC)

  This conversation repeats itself over and over again. Somebody says "we need a quick way of desysopping admins." I say "is ten minutes quick enough?" and then the discussion moves on to saying something like "no, not really dangerous admins, but less urgent cases." And I say "well we've got dispute resolution for those cases" and I also usually point out that admin actions can be reversed and administrators can be blocked. Dispute resolution at its simplest is community discussion of a problem. But I'm sure this will crop up again, even though we obviously already have all the tools we need to deal with any problems we encounter. --Tony Sidaway 01:16, 24 October 2007 (UTC)

  Community dispute resolution does not work with respect to administrators. An admin who abuses users to the point where they complain can throw up a smokescreen about other users' behavior, and count on support from a few friendly admins to overwhelm the users' complaints. If the complaints are exceptionally well-grounded, the admin can promise to reform, lay low for a couple of weeks, then return to his previous behavior patterns; if new complaints arise, the old complaints are dismissed out of hand by supportive admins. Only if an aggreived user is willing to drag things out to arbcom, and spend an enormous amount of time compiling evidence, will anything happen. Even compiling evidence is dangerous - admins routinely delete evidence-gathering pages as "attack pages", and verbally abuse users who question the decision. Argyriou (talk) 01:29, 24 October 2007 (UTC)

  Could you give an example of a case involving a problem admin, where serious problems have been downplayed at RFC by other admins, and evidence-gathering pages have then been deleted by admins? That would be a really extreme case of abuse and I don't think it would take more than a few hours to convince the arbitration committee to investigate. --Tony Sidaway 01:35, 24 October 2007 (UTC)

  User:Sir Nicholas de Mimsy-Porpington. He displayed fairly severe incivility on numerous occasions. There have been at least 2 RfCs regarding his conduct; his incivility is on display there. When I was preparing the first RfC, User:Pilotguy deleted my evidence page with no notice or warning to me or anyone else. Pilotguy has since given up being an admin; Nearly Headless Nick (as Sir Nicholas de Mimsy-Porpington signs himself) remains an admin, despite repeatedly showing himself to not have the temperament to do so, nor the ability to reasonably judge consensus on AfDs. If you want to take him to ArbCom, feel free. Given the amount of abuse I took from admins in my tangles with him, my contribution to an arbcom case will be limited. Argyriou (talk) 17:39, 2 November 2007 (UTC)

  Thanks for your response. I note that this involves the following RFCs:

  Wikipedia:Requests for comment/Sir Nicholas de Mimsy-Porpington

  Wikipedia:Requests for comment/Sir Nicholas de Mimsy-Porpington 2

• Of these, I note from the first one the heavily endorsed outside views of Samir and Samuel Blanning.

• Samir's statement addressed only the blocking of Kuntan and included the statement: The username block is entirely apt and I strongly question the integrity of any RfC that is asking for an administrator to comment on the blocking of a clearcut troll. Exonerating Sir Nicholas of blame in this particular matter, it was endorsed by 42 editors. Samuel Blanning's view examined only the disputed deletions, and included the statement that Sir Nick ...appears to have a much better grasp of AfD closing than the RfC requester, who refers to AfD here as a vote, which it is not, and claims in all all but one of the AfDs listed that Nicholas was at fault because he closed against the numbers. It was endorsed by 26 editors. A motion supportive of Sir Nick, asserting that fD is not a vote. Admins are expected to close debates using their discretion and based on the balance of evidence. had 18 endorsements.

• Of views critical of Sir Nick's conduct, a motion that he had "enacted a block in a way inconsistent with policy" had support of 12 editors, including Sir Nick himself, who remarked that he had apologised and did not intend to repeat the error.

• No other motions or views on the RFC came close to this level of support.

• The second RFC seems to have had a far lower level of patronage. It started as a request for arbitration and was then moved to an RFC. Few editors seem to have paid it any attention. Radiant!, whose comments on such matters are seldom less than apposite, correctly remarked that the RFC/arbitration request had been made while deletion review was still considering the disputed deletions. Later, GRBerry commented as "the primary closer of deletion reviews", remarking that only 2 of the 6 were overturned and that according to analysis of earlier deletion reviews, "a 30% to 33% overturn rate is fairly typical for contentious cases at deletion reivew. We thus learn from this sample that the rate of overturning Nick's actions is fairly typical of the administrative corps as a whole." After considerable analysis of the circumstances pertaining, he continued, "Nick's actions are of similar quality to other administrators, so there is nothing more to do here." I concur.

• In short, the suggestion that Sir Nick "[does] not have the temperament to do so, nor the ability to reasonably judge consensus on AfDs" seems to be poorly founded, if this is the evidence upon which it is based. --Tony Sidaway 18:33, 2 November 2007 (UTC)

View by Irpen

All problems with the current system can be reduced to the following.

1. Some of the candidates that would make good admins do not pass.
2. The percentage of bad admins among the promoted candidates is higher than it should be (ideally zero, but that's impossible).
3. If a bad admin is promoted, undoing the mistake is so hard that it is almost never done.

All these problems would be addressed if the workable procedure for deadminning of bad admins by community, safe from trolls but still effective, would be introduced. People would be less hesitant to vote support at RfA's if the mistakes are correctable and more candidates would pass. On the other hand bad admins would assuredly loose their bits while now it requires a torturous ArbCom case and the ArbCom is barely functioning anyway.

Whatever reform is taken, if a good community deadminning process is developed, the problem would be largely solved. The community should concentrate on hammering out the deadminning process as the first and most necessary step.

Users that endorse this summary

1. Irpen 23:17, 13 October 2007 (UTC)
2. I agree. A more streamlined process to remove 'bad' admins would considerably lower the hurdles towards becoming admin. It's my opinion, too, that many editors are overly critical in the RfA process, because they fear of becoming stuck with an abusive admin that is very difficult to remove in the ArbCon procedure. That's only self-defense. It seems to me that ArbCon decisions often don't meet the expectations of the broad majority of editors. This begs the question, why such a complicated 'trial' at all? If adminship is "no big deal", removing the tools shouldn't be a "big deal, either. So, I'm totally with Irpen in demanding that the process for removing admins has to change. Why not simply copy the RfA procedure for a new RfR (Request for Revocation) process? Gray62 13:35, 14 October 2007 (UTC)
3. Walton^One 17:22, 14 October 2007 (UTC)
4. As much as it may suprise some, I am fully endorsing this summary.-- _{Piotr Konieczny aka Prokonsul Piotrus | talk}  03:58, 15 October 2007 (UTC)
5. A point like this also needs to be raised at the WT:RFA straw poll, where a lot of people are claiming that de-adminning is "a solution waiting for a problem". shoy 18:55, 15 October 2007 (UTC)
6. Quite a fair way of looking at the situation in my opinion. Camaron1 | Chris 20:09, 15 October 2007 (UTC)
7. A potential benefit would be that bad intentioned applicant might not go through the process (including building fake cv's over time) if they were aware that misapplication of powers would result in deadminning. LessHeard vanU 21:28, 15 October 2007 (UTC)

   Good point, LHvU! Yup, this might result in adminship becoming less desirable for editors who want to misuse it for pushing POVs. Besides, it also could result in admins becoming more careful regarding WP:Civil :-) Gray62 22:04, 15 October 2007 (UTC)

• Comments
• Such opinions have been stated over and over again. Consensus for change on this subject will not materialize until those urging it actually come up with evidence to support the handwaving. Here Irpen claims that it's hard to remove an admin (I've cited much evidence, in the form of summarily dismissed admins, that this just isn't the case) and that there are more bad admins than we should have (but he conveniently fails to quantify how many bad admins). It is a systemic problem of such claims (of which there are quite a few on this page) that the evidence of bad admins being desysopped is ignored while the claim that there are significant numbers of bad admins still operating is never supported. --Tony Sidaway 00:18, 1 November 2007 (UTC)

View by Argyriou

Normally, this should be handled somewhere other than user conduct RfCs, but it appears that this is really a complaint about the conduct of various users *at* RfC. While it's often possible to take a single user and show him the error of his ways, it's a lot harder to convince 200 or 300 users, of which many are existing admins, that they're doing it wrong. The problem is not that users do inappropriate things at RfA, but that the existing structure of RfA encourages poor behavior.

^demon says: people vote because of their own arbitrary standards, rather than judging the candidate on his own merits and abilities pertinent to adminship. This is largely false. Some people apply their own standards rather rigidly - anyone who says "come back when you have 2001 edits, 1999 isn't enough" is being ridiculous, but by and large, people use standards like mainspace edit count, edit summary count, involvement in talk pages, involvement in project pages, and the like, as proxies for the candidate's "merits and abilities pertinent to adminship". I've opposed candidates for beign too focused on the WP project space, because, in my opinion, mainspace contributions are pertinent to adminship. Now, that's my particular bugbear. Other people may care much more about participation at XfD, or recent changes patrol, or vandalism fighting, etc. Those may be "arbitrary", but they, too, are indicators of a candidate's "merits and abilities".

So really, there is no (significant) problem with the conduct of users at RfA, there's a structural problem with adminship. I personally don't see any real need to change how admins are created, only how they are overseen. That said, I'd be ok with switching to a straight vote, and might support several other reform proposals made above. But if there remains no effective way to sanction abusive admins, I'll just keep (!)voting no on lots of RfAs. Argyriou (talk) 20:34, 15 October 2007 (UTC)

Users who support this comment

1. I mostly agree, though I don't vote no on many rfa's. I also don't vote yes on a lot. --Rocksanddirt 22:05, 15 October 2007 (UTC)
2. Agree. And it seems to me, too, that ^demon's remark about "judging the candidate on his own merits" is besides the point: That's what we do at RfA! Well, I guess ^demon doesn't like the way we do it. But the problem is, how are merits defined, and what amount of merits is necessary for an admin? Who's gonna decide this? ^demon? No, thx. Let's stay with the good ole consensus, it ain't broken. Gray62 22:16, 15 October 2007 (UTC)
3. I agree with Gray62. Basically this gets to the heart of the problem. I myself am a fairly generous voter at RfAs, but I can understand that people are reluctant to promote those who might become abusive admins, knowing how hard it is to get rid of them. Walton^One 08:05, 17 October 2007 (UTC)

View by Gray62

RfA isn't broken. The statistics (from 8/2007) show that about 40% of candidates become admin. Regarding the fact that there are many snowball noms, where candidates simply haven't 'the merits' yet (hardly surprising after two weeks, an actual example), and that there are candidates who have issues with WP:Civil or POV, this isn't a bad result. And many editors are successful in their second run, after working on their 'weaknesses' pointed out in the first try. One such candidate is just on the way into adminship with an impressive show of positive consensus. So, everything A-OK with RfA.

The real problem is, we don't get enough candidates. The RfA process could handle more, but too few editors volunteer for adminship. All those critics that want to get rid of the procedure should better use their time to recruit some good candidates. Regarding the masses of commited editors, this can't be too hard. Gray62 22:31, 15 October 2007 (UTC)

• I agree that the problem is that we get too few applicants. I suspect that this is because no sane person wants to put himself through the rather distasteful process that RFA has been for the past two years. All committed, experienced, trustworthy editors should be admins. That's probably a large proportion of all regular editors, if not a majority. Yet the process has given many of these same editors the impression that an administrator is something special. It wasn't always an obvious result of the process, but it has become more evident over time. --Tony Sidaway 21:18, 16 October 2007 (UTC)

  Is there some evidence to support the assertion that RFA has become distasteful lately? That sounds suspiciously like being "broken" to me. I haven't been convinced that it's broken yet, but I'm willing to be persuaded by evidence. Friday (talk) 21:24, 16 October 2007 (UTC)

  Yes, there is overwhelming evidence. For a good example of RFA at its absolute worst, I'd raise the Danny RFA where there was much evidence of editors voting on concerns quite unrelated to adminship. Although a great effort was made by the bureaucrats to remedy this and the decision was accepted, the discussion should not have been allowed to go that way. But the paucity of candidates is the strongest evidence for RFA's negative effect on the public perception of adminship. The many thousands of good potential candidates are staying away. The right process should not have that effect. --Tony Sidaway 21:33, 16 October 2007 (UTC)

  I'm apparently not seeing what you're seeing. I see new candidates coming along all the time. Saying "yes but there should be way more" isn't really evidence, it's just an opinion. As for Danny, well.. some RFAs are controversial. Some people find it distasteful to see that many people disagree with them. I don't see how this is any kind of indication of a problem with RFA. Friday (talk) 21:43, 16 October 2007 (UTC)

  You see "new candidates coming along all the time." What you do not see is many candidates, nor many administrators. Ignore it or not, that is evidence, not opinion. --Tony Sidaway 22:26, 16 October 2007 (UTC)e

  Yes, and the only way we will get more administrators, and allow all experienced editors to participate in routine admin tasks, is by implementing a community desysopping process; thus standards at RfA will become less ridiculously elevated, because admins who don't meet the community's expectations once sysopped can be desysopped at will. Walton^One 08:03, 17 October 2007 (UTC)

  Yep. The attitude of admin-evaluators should be "we'll give you a shot" not "you must be perfect". And the attitude of (some) admins should be "I better stop being a dick" not "Hah, I can game this thing all I like and no one can stop me."--Father Goose 08:26, 17 October 2007 (UTC)

  It's not enough to say "the only way we will get more administrators, and allow all experienced editors to participate in routine admin tasks, is by implementing a community desysopping process."

• Firstly one must present evidence that there are many poor existing administrators whose behavior is not adequately curbed by dispute resolution. Nobody has done this.

• Secondly one must present evidence and reasoning to enable us to predict that the existence of a procedure in addition to our existing community procedure for desysopping (namely, dispute resolution) would persuade administrator candidates to come forward in greater numbers. Nobody has done this. --Tony Sidaway 04:21, 24 October 2007 (UTC)

• It has been commented (at a RfA) that the vigorous atmosphere of some applications give the new appointee some taste of the less attractive consequences of acting as an admin. I cannot foresee vandals, trolls and some genuinely aggrieved editors reigning back their criticisms, so making the process less adversarial may permit those who are less capable of suffering the slings and arrows of outrageous personal attacks into an environment they are not prepared for. Obviously, not every sysop is going to enter the particularly nasty areas (and nor should they if their talents lead them elsewhere) but admins as a group to tend to be lumped together. LessHeard vanU 21:41, 16 October 2007 (UTC)

  Trolls can be ignored most of the time; all editors have to deal with the attentions of trolls so this is a given. The false legitimacy afforded by RFA evidently causes much more damage to Wikipedia than the odd troll, by limiting the number of candidates to those who are prepared to undergo a beauty contest. You're correct to state that "not every amdinistrator is going to enter the particularly nasty areas." That is precisely the point. Most administrative actions are humdrum and routine, and all experienced editors should be given the power to perform them. --Tony Sidaway 22:37, 16 October 2007 (UTC)

  And not everyone is going to have a crappy RFA. If RFA is to be a test of a candidate's patience, it should be stated as such and every candidate should get the same treatment. Mr.Z-man 04:09, 17 October 2007 (UTC)

  RfA is generally a worse experience than receiving abuse from trolls or vandals. This is because, at RfA, one is subjected to harsh criticism from established editors in good standing; in my experience, critical comments are far more hurtful if they have some degree of validity, and come from a person that one respects. Most editors will take a comment like "Oppose, editor is immature and uncivil (diffs)", from an experienced editor, far more to heart than they would "Ha ha ha ur a wanker wiv a small dick" from a drive-by troll. So I don't accept the argument that RfA is a "training ground" for the abuse one receives as an admin. Walton^One 08:01, 17 October 2007 (UTC)

  Maybe people are seeing different crappy RFAs than I'm seeing. When I see them, often it's the candidate himself contributing to the mess. People who get all bent out of shape due to RFA opposition are only demonstrating that they are poor candidates. In this regard, the system works well. Being able to handle disagreement without whining, slinging insults, or stomping off to sulk in a corner is a necessary skill for an admin. Friday (talk) 14:15, 17 October 2007 (UTC)

  And that might show they will respond poorly to criticism as an admin. If they get no significant opposition though, there is no way of knowing. On the other hand, I always though it was odd that people oppose because the candidate tries to counter the oppose comments. Why is that so bad? Why is defending yourself (if you can manage to stay civil and not become a drama queen) a bad thing? And what of the opposition comment mini-wars? Situations where someone opposes and a debate breaks out (that may or may not involve the candidate) about the legitimacy of the reason for opposition. The issues with userboxes and WikiProject endorsement come to mind as well as an opposition I've seen at least once (maybe twice, I try not to get too involved in RFA) where a person opposes a candidate for not having a full year of experience. (Though if standards continue to raise, that will probably be the norm by next year.) Last I checked, Wikipedia is not a battleground. If there are so many battleground situations elsewhere on Wikipedia that people suggest RFA should be a "test" battleground, we should really look at how well we enforce WP:BATTLE in other ares instead of just raising expectations for admins to make up for failure to adequately enforce policy. Mr.Z-man 17:19, 17 October 2007 (UTC)

• The reasoning that justifies the appalling ordeal that RFA has become by saying that it's a test of the candidate's mettle is exactly the problem that has driven good admin candidates away and will continue to keep them away. We've got ourselves into the absolutely ridiculous situation where a little bit of extra power has been fetishized to the extent that only exceptional editors with a quite frightening dedication to becoming administrators are likely to even consider asking for it. And here many editors are seriously proposing that the problem is caused by the fact that we cannot desysop poor administrators. This is just further evidence that the admin bit is being represented as something that it is not. At the same time we've got records showing that the number of admins who have behaved so poorly as to need desysopping is minuscule. The problem of bad administrators is greatly exaggerated, indeed almost imaginary. --Tony Sidaway 01:26, 24 October 2007 (UTC)

Those are records of admins that actually got desysopped. The number that would get desysopped if they were subject to recall/relection of some sort would be far higher, I predict. Then again, that might only be an increase from 2% (ish?) to 5%. But the principle of ongoing accountability is still important. Imagine how careful you would be if, when hiring someone for a position, you knew you could never ever fire them, no matter how badly they behaved, as long as they didn't do something illegal?

I'm all for adding more admins so as to diminish the "fetishizing" of admin powers -- this is something that absolutely should be done. But adding a more responsive means of desysopping should absolutely be done at the same time. Having adminhood be a "life membership" is what fetishizes it more than anything.--Father Goose 02:59, 24 October 2007 (UTC)

I've already said, and presumably you agree, that poor admins are already dealt with--sometimes in a matter of minutes. You ask for accountability but you must already be aware that every single editor, including every administrator, is fully accountable for every single action he carries out on Wikipedia. That's what dispute resolution is for. In one of the first arbitration cases in which I was active as a clerk, the case began with five administrators summarily desysopped. One could hardly say that we don't hold sysops accountable for their use of the bit.

What concerns me here is that it looks to me very much as if some people are proposing, in the face of these facts, that we reward the nasty, vicious snakepit culture that has grown up around RFA by extending these appalling activities to cover not just prospective administrators, but acting administrators, too. The problem is the snakepit culture itself. It's quite fatuous to say that the culture exists because admins aren't accountable, when they clearly are. Let's not make any excuses for the nastiness. If extended to enable the hounding of admins (even further than they are already pursued by assorted pov-pushers, trolls and the like) it could only reduce their effectiveness. Moreover prospective administrators would be unlikely to come forward if they believed that they'd have to face the sort of ugliness that we've, sadly, accepted for far too long on RFA, every day of their time on Wikipedia as long as they were administrators. --Tony Sidaway 04:04, 24 October 2007 (UTC)

You've repeatedly failed to provide any evidence that RFA is broken, or a snakepit, or whatever it is you're saying it is. Friday (talk) 13:57, 24 October 2007 (UTC)

On the contrary, I've pointed out what I view as evidence for the problems of RFA. The most obvious is that we're just not getting enough people coming forward to give us a reasonable proportion of administrators. You may differ with me on whether that's evident, but that's a different matter.

On my "snakepit" comment, that's a matter of opinion. I've already given the Danny RFA as an example of a case where many comments unrelated to adminship were made, though in this case the bureaucrats skilfully handled the problem. The low application rate also suggests that my opinion is shared by many potential administrators. --Tony Sidaway 19:00, 25 October 2007 (UTC)

Can you give some examples of RfAs that were shot down by pov pushers and/or trolls? Mind you, we're in muddy waters with that language to begin with -- what I'm asking for is some where the "objections" could not be construed as reasonable concerns.

Separately, it's not that the admins aren't accountable, it's that they're only accountable in the most extreme of cases. There are plenty of POV-pushing admins (not very many trolls, thank goodness), but you can't desysop them for pushing a POV until they do something overtly "illegal". Even then, they often get away with big no-nos for a long time, like protecting pages where they have a stake in the dispute, or banning users who push a POV opposite to theirs.

In theory, admins should not have any special power over content. In practice, they have a lot. We want admins that can conduct themselves in an admirably neutral manner, and we want the power to defrock those who don't. Arbitration is usually possible only in the most extreme of cases. Short of that, admins can abuse their powers quite often. Only a few actually do, but there's no reason why we should be unable to curtail such behavior.

Basically: if becoming an admin should be no big deal, being de-adminned should be no big deal as well.--Father Goose 18:51, 24 October 2007 (UTC)

I'm not sure whether your request "Can you give some examples of RfAs that were shot down by pov pushers and/or trolls?" is addressed to me. I have made no such suggestion. I have said that the problem lies with the culture of RFA itself, not the occasional troll.

Again for intrusion of unreasonable concerns into RFA I'd have to cite the classic case: Danny. It's not that there aren't many other such cases, rather I'd prefer to cite the most widely known, widely acknowledged case.

I agree that "there is no reason why we should not be able to curtail" bad behavior by admins. We are already able to do so, and have sometimes done so in a matter of minutes. --Tony Sidaway 19:08, 25 October 2007 (UTC)

I'm an admin, and I often debate where I have a strong personal opinion on the merits of the articles; I never close these debates, though I am fully aware I could simply refrain from debating and close the ones that are borderline and get away with it--presumably until I got over-confident and did one that was noticeably wrong. I am often asked offline to intervene on topics where I have a personal view. I always turn down requests to use admin powers in such cases. The most I will do is comment as an ordinary editor. I expect every admin to be as reticent as I, and it is obvious that some of them don't. It seems clear to me admins need another level of supervision. De-adminning may be overkill though, just as a sense of disapproval at AN/I may be too weak. DGG (talk) 18:31, 25 October 2007 (UTC)

Enforce and reinforce Wikipedia policy at RfA

Such as Wikipedia:No personal attacks. Since noms are often chastized for defending themselves against accusations, claims made against a nom should be substantiated. The Transhumanist    02:15, 17 October 2007 (UTC)

The challenge there is that my unsubstantiated personal attack, is your reasonable question on appropriate use of wikipedia policy backed up with multiple diffs to a discussion over several days on an article talk page. --Rocksanddirt 16:17, 17 October 2007 (UTC)

View by Phoenix-wiki

Very simple solution; Oppose an RFA only if you have a very, very good reason--Phoenix-wiki (talk · contribs) 21:35, 27 October 2007 (UTC)

View by User:Husond

No system is flawless. RfA may therefore have a few flaws, but it's far from being a failed procedure. In fact, it's so far from being a failed procedure, that years of discussions about reforms and improvements for it are yet to provide the slightest community-approved modification. Here's an abridged version of the most hated flaws and why they still make the best system we came up with so far:

• A candidate should be approved by consensus, not percentages: Face it, on Wikipedia any non-random, non-abstract, non-unrecognized form of consensus will always require a number or percentage. The closing crat will always need his or her priorly established number, over which would lie consensus.
• Discussion should be encouraged instead of a vote: Discussion should by all means be encouraged, so to expose the candidate's positive or negative traits/qualities/conduct. These will ultimately influence the number of supports, opposes and neutrals. Thus RfA is never a vote, it's just a vote-look-alike outcome of a regular discussion.
• The percentage for promotion is too high: Well, it should probably be. After all, promotion is simple, but demotion is a lengthy, slow, painstaking and insanely boring process. Everybody knows that and so there has been reasonable consensus for maintaining the traditional 70% promotion line.
• Participants often support without providing any particular reason or adding anything to the discussion: By simply signing on the supporters section, users effectively endorse an RfA candidate. They need not provide any further explanation really, their endorsement already implies that they consider the candidate suitable for the job, be it for the nomination, personal experience with the user, scrutiny of his/her contributions, etc. No further complication required.
• Participants often oppose in bad-faith: RfA is hardly responsible for an unfortunate trait of the human nature. Nonetheless, bad-faith usually stands alone on RfAs and rarely influences the outcome.
• Popular, unproductive users will be promoted, while unknown, hard-working users will not: This simply does not happen. "Popular" users have unsuccessful RfAs just like everybody else when their mistakes or lack of productivity are pinpointed and exposed. Unknown users will usually have less participation because, well, they've had less interaction. But their hard work will have them attain a successful RfA, just like everybody else as well.
• The right candidate failed an RfA: Yes, it could be the great vandalfighter who does nothing but fight vandals or the incredible editor who's written hundreds of articles. They have failed their RfAs. Many thought they were the right candidates, but the community decided otherwise. The supporters should just get over it, and the candidates should focus on the reasons that caused their RfAs to fail, in order to achieve a successful one next time around.

I support this system and its flaws. Húsönd 00:34, 19 October 2007 (UTC)

Users who endorse this view

1. J-ſtan^Talk_Contribs 01:31, 19 October 2007 (UTC)
2. Right on the money. Keegan^talk 02:38, 19 October 2007 (UTC)
3. Yes, I think your right on in your view of the "flaws" of the current system. The current system is a distance from been broken. Camaron1 | Chris 11:10, 20 October 2007 (UTC)
4. I think that Husond has made a very good assessment of RfA here. Captain panda 02:47, 26 October 2007 (UTC)
5. Well put. Sjakkalle (Check!) 09:37, 1 November 2007 (UTC)
6. I strongly endoser this view. Husond states is well, i accept this system and its flaws. If you failed an RFA (i did my first one), go back and work on imporiving yourself, not blaming it on the wiki. Chrislk02 (Chris Kreider) 14:53, 12 November 2007 (UTC)
7. Yes, this sums up my take on things after reading through all the different views here. It isn't perfect, which (many)others have mentioned - but neither their solutions or mine improves the situation without adding other, more problematic things into the equation. I think the system as it stands is most likely to identify active, civil folks. Sausage != sausage making. Cheers, Ryo 18:35, 15 November 2007 (UTC)

Views by CO

This RFC isn't going to solve anything, it's a waste of time. A majority would have to want a change, and from what I see, this isn't going to happen. Carbon Monoxide 03:16, 20 October 2007 (UTC)

With respect, I disagree. Any system that has come under sustained critism from experienced editors for the last year is not working in a way that is satisfactory to this encyclopedia's community. Should this RfC fail to produce any results, I urge that this matter be taken to the Arbitration Committee. — Thomas H. Larsen 03:28, 20 October 2007 (UTC)

What on earth could ArbCom do? They have no power to change policy.--Father Goose 05:51, 20 October 2007 (UTC)

Actually, I recall that it was taken to ArbCom a few months ago and rejected. I can't find the link, though. Captain panda 20:11, 22 October 2007 (UTC)

I respectfully disagree as well. When this RFC closes, there will not be any change in the system immediately as the result of the closure. This is a request for comment in the simplest sense; it has provided a forum for users to pitch and debate their ideas. The result of this discussion will be the community coming to an understanding of just what the flaws are that need to be modified. Many of these comments are leaning towards a consensus as to where the discontent lies. This has provided a centralized forum for discussion and will inevitable result in viable proposals for alteration. And I agree that ArbCom has no sanction of RfA. Keegan^talk 04:42, 21 October 2007 (UTC)

It's quite evident that there is no consensus in this RFC.

You say that the arbitration committee "has no sanction" in this matte. That's not strictly correct. The arbitration committee has in the past made rulings covering the conduct of RFAs, the bureaucrats, and everything else, and is quite likely to do so again if ever this becomes necessary. However it's unlikely to become involved in this instance, which is a policy discussion, not a conduct dispute. --Tony Sidaway 20:21, 22 October 2007 (UTC)

As I've said before, despite the constant claims to the contrary, a majority do want change. However, there is not agreement on any specific change, and many are mutually exclusive. -Amarkov moo! 22:26, 25 October 2007 (UTC)

And that is the exact problem. There is consensus that there should be a change, but nobody really knows what that change should encompass. Stifle (talk) 10:59, 27 October 2007 (UTC)

Proposal for temporary status of provisional admin

I have read comments: “we need more admins”; “the bar is set too high”; “adminship is no bid deal|”; “reluctantly oppose … not enough experience …”; “not comfortable with him becoming an admin at this time”.

The problem here is that the promotion is an all or nothing process. “Administrators open to recall” was an attempt at compromise, but the negative action of invoking a recall is distasteful.

Can I suggest the compromise of “provisional admin”. A provisional admin has all of the buttons of an admin, but is on a limited contract (one year?). At the end of the contract, the provisional admin is reconsidered, only this time there is a year of admin actions to judge. This should mean that the act of promoting to provisional admin is based upon trust, not proven ability, and subsequently the act of promoting to full admin is based upon evidence. --SmokeyJoe 02:05, 29 October 2007 (UTC)

It's more or less what Friday proposed above. It's a de-adminship process, only that it's automatic. I would support it. A.Z. 02:45, 29 October 2007 (UTC)

The automatic de-admining of provisional admins is a big difference in that nobody needs to be the bad guy who makes allegations. An iffy applicant who proceeds to be an iffy admin can be simply left to lapse. If he re-applies, it is up to him to make his case. --SmokeyJoe 05:18, 29 October 2007 (UTC)

I support this proposal, provided that the status was granted for a month rather than a year. — Thomas H. Larsen 04:42, 29 October 2007 (UTC)

I’d suggest longer than a month, reasoning that a month of careful behaviour is too short to be permanently habit forming for a candidate who is suspected to be of questionable suitability. Candidates of quality as are currently promoted to Full Administrator should continue to be promoted straight to Full Administrator. If the provisional period is a year, there would be nothing to stop a Provisional Administrator from returning to WP:RFA well before the completion of a year. --SmokeyJoe 05:18, 29 October 2007 (UTC)

In many organisations, the concept of a probationary period is quite normal. In this case it would seem to be that a three month period ought to be enough to see the true abilities. A year is a long time, and although a long period has the advantage of being able to call a halt any time earlier, a three month review seems to create a sensible balance to me. That review could be as simple as demonstrating that there have been no complaints, or that complaints raised were unreasonable or not of sufficient concern to merit de-sysoping. Spenny 11:47, 29 October 2007 (UTC)

for the problem admins I have seen, one month would not have been enough, but three months would. I think this is an sensible reform that could well attract support. I'd suggest that the fairest way to do it would be simply to apply it to all admins, regardless of how excellent they appear to be. Some of the apparent problem admins did not appear to be such until they received the tools. I would not, however, lower our initial standards, in the hope of removing our mistakes. I think they are quite low enough already. I will mention that an admin would likely not feel comfortable acting on difficult cases until full status. DGG (talk) 08:51, 30 October 2007 (UTC)

Okay, here I go again: which particular problem admins are currently active and should be desysopped? Again and again I see claims that there are such administrators, who should be desysopped, and again and again I ask for specifics. I haven't had a reply yet. --Tony Sidaway 00:12, 1 November 2007 (UTC)

You are asking for dirt. Trying to take a more positive path: Have there been any candidates who failed RFA but might have succeeded to the temporary position of Provisional Admin? --SmokeyJoe 00:26, 1 November 2007 (UTC)

I'm not asking for "dirt". I'm asking those who espouse this view to substantiate it or stop making vague accusations about the existing admin corps. --Tony Sidaway 00:33, 1 November 2007 (UTC)

All admins, on the whole, in various styles, do a fantastic job. The Provisional Admin proposal supposes that the bar is too high, out of fear of letting a substandard candidate through. The proposal is of an error-correction mechanism, to catch the mistakes that would occur if RfA were allowed to become more efficient, with a higher flow. The initial assumptions are (1) more good admins are wanted; (2) potential good admins exist, but they fail, would fail, or fear failing the current RfA process. The Provisional Admin proposal would do nothing in respect to the existing admin corps, and it has nothing to do with the deadminning of existing admins. --SmokeyJoe 01:10, 1 November 2007 (UTC)

The devils will be in the details. What process would we have for the admin's review? One in which trolls and blocked users come to have their pound of flesh in a public hanging? Or one in which fellow admins evaluate his/her performance from an adminiship viewpoint? Another process? ≈ jossi ≈ (talk) 00:18, 1 November 2007 (UTC)

The process for review of a Provisional Admin would be the current RfA process. Nothing would be different except that actual admin actions could be scrutinized. The assumption is that, as in DNA replication, the addition of the error-correction, or proof-reading step would lead to increased efficiency, higher throughput, without sacrificing quality. --SmokeyJoe 01:10, 1 November 2007 (UTC)

You can put me down as an enthusiastic supporter of the probationary adminship proposal. The period isn't important; anything from a couple of months to a year should be fine. I further propose, as an optional extra, that a provisional admin may post, on his own talk page, a statement of intention to become a permanent admin as his period closes, and if there are no objections on that page when the period expires a bureaucrat can simply announce the posting to be permanent. This would prevent RFA clogging up with pile-on supports for provisional admins who have proven exemplary. --Tony Sidaway 02:14, 1 November 2007 (UTC)

The details at this point are obscure, and I would not support until these are clarified. A two-pronged RfA one at the outset and one after a probatory stage, will only mean useless drama. If an admins misbehaves, there are other ways to address desysoping.≈ jossi ≈ (talk) 02:33, 1 November 2007 (UTC)

Which/whose point is obscure? “will only mean useless drama” No. It will mean more benefit-of-the-doubt can be allowed on the first round. Provisional approval is less critical. This means less drama. nb. I prefer "provisional" which implies temporary, whereas "probation" is associated with convicted offenders. --SmokeyJoe 03:32, 1 November 2007 (UTC)

View by Szyslak

RFA is a mean place.

By that, I don't mean to say those who frequent RFA are mean people. RFA is a mean place because the experience of going through an RFA is unpleasant. And needlessly so.

Imagine this scenario: You've been a Wikipedia editor for many months, or perhaps many years. You've devoted hours upon hours to improving this encyclopedia, to the extent that your free time and your individual editing talents permit. You've gained the trust of the Wikipedia community, without so much as a smudge on your block log. You've run into many situations where you've felt admin tools would benefit you.

One late night, you're dealing with a particularly obnoxious vandal. At the moment, no admins are on duty at WP:AIV. The vandal runs amok for a few minutes, until an admin blocks him/her. Said admin notices your valiant vandal-fighting, and offers to nominate you for RFA. You figure why not; maybe it's about time.

The RFA goes live. One support come in. Then another. And another. It seems that the community feels you're ready for the tools. So you go to bed, confident that your RFA will succeed.

You wake up the next morning. Curious, you check your RFA page. In the past six hours, twenty more editors have supported -- not bad! But, just below that exuberant show of support, there lurks an "oppose" comment:

1. Oppose You seem like a fine user, but you clearly lack the wide-ranging expertise needed for adminship. For example, I noticed you have only 27 edits in the category namespace. Therefore, you haven't shown to my satisfaction that you understand categories, an important part of the encyclopedia. In addition, that three-day wikibreak you took last month shows a lack of commitment to the encyclopedia. ToughRfaGuy 11:28, 30 October 2007 (UTC)

You say to yourself, Of course I understand categories. I have two featured articles I built from scratch! Do you think an article can reach featured status if its categorization is nonstandard? And who cares about the three-day wikibreak. This isn't a job where you get fired if you don't show up for three days. It's a volunteer project!

The RFA may succeed. Or it may not. That one oppose won't sink your RFA. Still, some damage was done. That oppose comment was mean - it didn't assume good faith, it made hasty assumptions based on a quick reading of namespace balance, and it went by outrageous standards with no basis in reality -- i.e. you have to have hundreds of category-space edits and you're not allowed to take wikibreaks. (I'm sure there are excellent admins who have a handful of category edits and take long, frequent wikibreaks.)

Whether RFA is "broken" is debatable. But two simple changes would bring us a long, long way:

• Assume good faith
• Drop the silly oppositions

Users who endorse this summary

1. szyslak 11:28, 30 October 2007 (UTC)
2. Adrian M. H. 01:32, 1 November 2007 (UTC)
3. K. Scott Bailey 14:24, 1 November 2007 (UTC)
4. Agree. Cirt (talk) 15:48, 12 December 2007 (UTC).
5. AGF is our most important principle. --Aqwis (talk – contributions) 12:26, 3 January 2008 (UTC)

Comments

Anyone who thinks that hypothetical oppose comment is "mean" is probably too thin-skinned to be an admin. Argyriou (talk) 17:41, 2 November 2007 (UTC)

Maybe someone who takes such a comment personally is a little thin-skinned. But that doesn't distract from the fact that such RFA comments don't assume good faith. Besides, why should RFA be a trial by fire? szyslak 20:08, 2 November 2007 (UTC)

People seem to very frequently completely misunderstand what "assume good faith" means. I don't understand this- the words are well known and quite clear. Yet, here we see it again. There's nothing in the hypothetical oppose that fails to assume good faith. Friday (talk) 20:13, 2 November 2007 (UTC)

It doesn't assume bad faith. szyslak 20:33, 2 November 2007 (UTC)

The definition of a "silly oppose" is relative. Whereas you think opposing for category edits is silly, others in Wikipedia think it is something that administrators should be involved in. Also, that example oppose wasn't assuming bad faith. Opposing in an RfA does not mean that you hate the candidate. It simply means that at this time, you do not think the candidate should be an admin. Captain panda 13:22, 3 November 2007 (UTC)

"why should RFA be a trial by fire?" It shouldn't if being an administrator meant only dealing with civil users. Unfortunately, being an administrator means being able to deal with uncivil users. And not only vandals - some of our best editors can get quite heated in an argument. An administrator needs to be able to stay cool under criticism, even unreasonable criticism. And while I would never oppose just to raise the level of heat in an RFA, if an administrator candidate can't coolly react to mild criticism, even unreasonable criticism, I would certainly oppose due to that. --AnonEMouse ^(squeak) 15:45, 3 January 2008 (UTC)

Szyslak's invocation of WP:AGF

About my invocation of WP:AGF: I, like a number of other users, choose to apply AGF more broadly than "required" in the guideline page. For me, the spirit of AGF goes beyond simply not assuming bad faith. One way to put it is, don't make any negative assumptions or hasty generalizations about another editor without solid evidence. (To be clear, this is about my own practice of AGF, not about how I'd "enforce" that principle or what I expect of those I interact with.)

So, is 27 category space edits really solid evidence that an editor misunderstands categorization? You can add several hundred categories to various articles while making just a handful of edits to the actual category pages. While the hypothetical example above doesn't assume the candidate is a bad user, it does jump to a rather negative conclusion based on a frivolous namespace count. (Of course, not all namespace or edit counting is frivolous; for instance, most of us agree an editor with a couple hundred edits shouldn't be an admin.) Remember that the opposer is also assuming a "lack of commitment to the encyclopedia" based on a single three-day wikibreak. Opposing an otherwise qualified user for adminship based solely on category-space edits and a three-day wikibreak is unnecessary and simply ridiculous. While most editors are not personally offended (I certainly wouldn't be), such silliness helps poison the atmosphere, along with general rudeness and a big helping of wounded pride.

My point: RFA is an unnecessarily hostile place. And silly oppositions based on harsh, arbitrary criteria aren't helping. szyslak 21:52, 3 November 2007 (UTC)

I think I understand you better now, but: I still think you're making yourself less clear by using the words "assume good faith" to mean a bunch of stuff other than "assume good faith". If you mean "don't oppose for stupid reasons" I think it'd be clearer if you said that. Friday (talk) 18:32, 6 November 2007 (UTC)

"Don't oppose for stupid reasons" is part of it, but so is "Don't assume negative things about someone with flimsy evidence". Here, a "small" number of category edits is taken as evidence of (a) unfamiliarity with categorization, which leads into (b) a lack of "the wide-ranging experience needed to be an administrator". Similarly, a wikibreak is used as flimsy evidence that the candidate lacks commitment to Wikipedia. szyslak 20:25, 6 November 2007 (UTC)

View by Hiding

WP:RFA doesn't work, except when it does. It's a hard one to explain; we all know what the flaws are, and yet we can't seem to agree on an alternative, because they all have flaws too. Let us remember the words of Winston Churchill: "It has been said that democracy is the worst form of government except all the others that have been tried", as well as the adage "Better the devil you know". Of course, the obvious rebuttal is "evolve or die" or the other Churchill quote on democracy, "The best argument against democracy is a five minute conversation with the average voter."

To sum up, WP:RFA is the worst process for choosing admins, except all the others we've thought of, and the best argument against WP:RFA is any random individual request. If we can't find a commonly agreeable solution, are we sure we have a problem? Hiding Talk 17:53, 3 November 2007 (UTC)

Users who endorse this summary

1. -- Tawker 06:02, 11 November 2007 (UTC)

View by TomStar81

It may be true that our RFA process isn't perfect, but we must all recognize that the only time something will function perfectly is in the pages of our favorite fictional universe. The question before us then is not an issue of whether or not this process has failed, but how much succsess we have had with the system we are currently using. Some say that this system is too flawed to be saved and thus should be scrapped, others say it works but could work better. Having had no real experince with rfa and its associated points, I do not consider my self to be in a position to judge whether or not it works.

My recommendations for changes to the process are as follows:

• We need a blind nomination system. I wholeheartedly believe we need to start by informing those who would be admins that they need to place themselves in a category listing Users who wish to become admins. Once a week a bot can randaomly select a set number of users for adminship and send out notices to the perspective admins informing them of thier nomination and instructing them to fill out the rfa pages if the accept. In this manner, we can eliminate paybacks, politics, and so forth from the nomination process, and we can reach out to those who do not wish to self nominate but would still like to be admins.
• Start with a Q&A session. I think that the nomination process should begin with about a weeks worth of Q&A for the perspective admin so that the community can get a feel for where the admins stand on issues related to hadling the mops here. In my minds eye I see this as a process similar to FAR, where people bring up their objects without any support, oppose, or neutral tabs and allow the potential admin to vioce his or her feelings, beliefs, etc for others to read and evalutate. If after a week there is a generally good review of the candidate, we move on to the actually "consensus" part (which we have all established as being a vote).
• Keep new admins on training wheels. If an admin passes the adminship process we should insist that they handle only simple tasks, like clearing out speedy delete requests and closing non controversial afds and such (by non-controversial, I mean those where the consensus was 75% or better for keep or delete). In this manner our newer admins can particiapte in areas where more experinced admins are in operation, and in the process our newer admins can gain some experince in hadling new adminship buttons under the watchful eyes of others. This also frees up our more experinced admins to handle tougher admin-realted duties.
• Require recertification. Once every year I would like to see all of our admins go through a mandatory adminship review; in which the community reviews the contrinutions of the admin taking into account that admins contributions with the extra buttons and there handling of any controversial situtations. If after a week the commonity consensus is that the admin in question is not using their tools for the betterment of the project the review becomes a request for adminship removal.

In reading this it needs to be stated that I am not an admin (yet), and that I (as previously noted) have little occasion to edit on the rfa page. These are merely my suggestions after reading through some of the proceeding comments. TomStar81 (Talk) 08:00, 11 November 2007 (UTC)

Users who endorse this summary

• Endorse the third and fourth suggestions. I think #2 has definite benefits, but I'm afraid it would be too much of a bureaucratic slowdown. I don't really like the idea of randomizing the nominations process, but perhaps we could use anonymous nominations if it would help succumb unwanted wikipolitics. Not sure if I'm really convinced of the benefits, but it's an interesting idea. — xDanielx ^T/_C\^R 07:21, 23 November 2007 (UTC)

Comments

I think that this is a great idea, but I, as always, have my small quibbles :)

1. I am fine with self-nominations and the like, as well as normal ones. However, I think nominating oneself or another for adminship requires human intervention and thought. Thus, I'm not sure if I can agree to a bot nominating users placed in a category for their sysophood. Plus, from a logistical perspective, Category:Wikipedia administrator hopefuls has just south of 800 members, and RfA might get a bit clogged with requests.
2. As for the recertification process, I would limit it to include only administrators that have had issues with their mops in the year prior to their recertification, as it wouldn't be fair to put longstanding and trusted admins through the process, as their would be little chance of them being desysopped. Thus, only problematic administrators should need to be reconfirmed (although "problematic" is a broad term and probably would lead to debate =D)

Other than that, I would agree with this proposal. Happy editing! ( arky ) 19:04, 11 November 2007 (UTC)

Respectively:

• Remember that their would be human intervention in the process anyway because we would still need people out there to comment on the proposal, and of course the persepctive admin would still need to answer the three questions asked of everyone in the begining. In my minds eye I see the bot as sending out that template that says "___ would like to nominate you for adminship!" and creating the initial page (nomination justification would be "listed in the category of admin hopefulls"), but would leave it to the nominee to accept of decline the offer, and leave it to community to find conensus. That last part is particularly important since building a consensus requires human intervention and thought. Having said that though I will concede a point in your favor; god forbid the bot would find and nominate a sock puppet or abusive editting account or something along those lines (the horror, the horror...).
• Remember that this process is composed of two parts: a section for constructive critiscm and a removal section. Requiring recertification is my way of allowing all admins to get a feel for what they are doing. Those admins in good standing would likely not end up being removed, like FAR/FARC the admin review would only go down if the community had a number of unadressed grievences with the user in question. Having said that, we could reserve this process to newer admins only by introducing a concept of tenure for our admins in good standing; tenur (assigned by a beauract) would exempt an admin from the review process provided he or she is in good standing the community as a whole (although it ought to be noted if we do this that it is not a form of immunity, and that long time admins in good standing can still be recalled or dysysoped if their is consensus to do so). This was brought up originally to allow everyone in the community to comment on an admin's work over the past year and thus allow an admin to get a little constructuve criticism. It would be daunting task though, requireing 1500 admins to be recertified would be a very big undertaking :) TomStar81 (Talk) 23:29, 11 November 2007 (UTC)

View by Epbr123

I think each successful RFA candidate should only remain an admin for a year; after which, they need to re-apply or be renominated. Epbr123 (talk) 19:24, 24 November 2007 (UTC)

Seems like that would cause a ridiculous backlog at RFA. Stifle (talk) 09:52, 4 December 2007 (UTC)

It would create about an extra 30 RfAs a week, which is still far less than the backlogs at places such as WP:FAC and WP:GAN. Epbr123 16:55, 4 December 2007 (UTC)

I agree with the principle, but instead of having a re-nomination or re-application, how about something like Editor Review? A year after a succesfull RfA, the admin would go up for Admin Review, where members of the community will review his/her general performance as an Admin. If the general consensus is that they are not doing a very good job, then they will have to re-apply/be re-nominated. --θnce θn this island ^Speak! 00:32, 5 December 2007 (UTC)

There seems to be an aversion amongst the community against desysopping admins. I think a poor admin would still have a good chance of passing an Admin Review, but wouldn't have as much chance of passing a repeat RfA if mandatorily desysopped at the end of the year. The community seems more willing to prevent potentially-poor admins from becoming admins, than to desysop admins who have proven to be poor. Epbr123 (talk) 01:48, 5 December 2007 (UTC)

Absolutely agree with your observation. Perhaps we should we require that all administrators voluntarily opt to add their names to the noncompulsory Category:Administrators open to recall. :-) — xDanielx ^T/_C\^R 06:48, 10 December 2007 (UTC)

View by Onceonthisisland

There are definitely some problems with RfA, and I think one of the main problems is editcountitis (opposing/supporting based on quantity of edits). Above, Sxeptomaniac suggested doing away with edit counters. Edit counters aren't the main source of the problem, it's more that people judge based on edit-count, and getting rid of edit counters wouldn't solve that. People would still go to contributions page's and approximate the edit number. I would suggest instead ruling that there can be no votes along the lines of:

1. Support, 10,000+ edits

OR

1. Oppose, less than 2,000 edits

This is my main view of this, and I can't wait to see what the eventual decision will be. --θnce θn this island ^Speak! 02:08, 28 November 2007 (UTC)

Users who endorse this Summary

1. As nom. --θnce θn this island ^Speak! 02:09, 28 November 2007 (UTC)

Comments

1. Agree with the principle to an extent, but I wouldn't support a blanket prohibition of comments of that nature. As I see it, edit counting isn't a great proxy reflection of a candidate's overall experience and consistency, but it's much better than blind guesswork. There are other ways of gauging these qualities (Q&A for example), but I think those should be regarded as compliments rather than substitutes. Also, I don't think adopting a policy of this sort would have a substantial impact on how editors vote -- I expect they would just express their reasoning in the form of "too little experience" instead of "not enough edits." And this is a bit tangential, but as the English Wikipedia we don't really have discretion over the toolserver where the edit counters are hosted (and if we did, the counters could be moved to private servers) -- the most we could do is remove links on the RfA page. — xDanielx ^T/_C\^R 06:27, 10 December 2007 (UTC)
   1. People could just say "Inexperienced." without ever mentioning an edit count. Malinaccier (talk) 00:07, 15 December 2007 (UTC)

Yes, but low edit count is not always an indication of being inexperienced. --θnce θn this island ^Speak! 16:52, 16 December 2007 (UTC)