Wikipedia talk:WikiProject Administrator/Archive 1

Archive 1

Archive 2

Archive 3

Teamwork

Most emphatically not pointing fingers or recalling any particular instance, it is my contention that the amount and quality of teamwork among admins has failed considerably. While we might be concerned with the downsides of high group cohesion, administrators generally are supposed to be oriented towards working together to improve and protect the encyclopedia. I feel in this aspect, we admins have collectively failed to work with each other. and there has been real damage done. Am I the only one who thinks so? If not, what, if anything can we do about it? How do non admins feel about administrators - are we working with you or against you? --Tznkai (talk) 05:20, 5 October 2009 (UTC)

I have some opinions as a non-admin. would this be a good venue to share them? Perhaps we can organize a survey for Wikipedians to take. bahamut0013^words_deeds 05:54, 5 October 2009 (UTC)

I would be in favor of a broad survey, but i'm also interested in kick starting some discussion.--Tznkai (talk) 06:02, 5 October 2009 (UTC)

We're all supposed to be on the same team, admin or not. The core issue the wiki is facing is it's own size; that changes everything. On smaller projects, it is still possible to have a discussion that does not run off on dozens of tangents. The last few days here have seen a lot of what we call drama. What was it about? It's easy to identify a specific few things that triggered events, but things blossomed, became about other things, too. Since every damn incident involves many of the same players and unresolved issues, we tag them with the same five-letter word. Since ancient Athens, people have known that discussion by everyone does not scale well. This is why some look to the AC to sort things; there are many problems that need sorting and mostly things just get smoothed over and the issues run for years. I've just been over on jv:wp and ms:wp trying to sort one of en:wp's problems: the Bambifan vandal, whom we have exported to smaller wikis. This shit is *our* problem yet he's running amok on the the project in the language that the word 'amok' is from. It is not just a question of people *here* working together, it about all of the many hundreds of projects working together. This wiki is the toxic wiki, the one that is so contentious. The other projects look to this one and cringe; they want to avoid this fate.

What do we need to do? Just discussing things is not the solution; any discussion with too many participants and high stakes fails, they run to a megabyte or more and amount to no consensus. On the high level issues, the consensus seeking approach can break down. Earlier this year, the AC sought to start an advisory committee. I don't recall the exact title off hand. A host of people tore it down. Did it have issues? Sure. But some big shift in how we do things is needed. The wiki explodes with great regularity and I feel that it's because core issues are not solved; we just get by incidents. Please think big, ok? I'll check back tomorrow; I gotta go. Cheers, Jack Merridew 06:48, 5 October 2009 (UTC)

I have much to agree with Jack about. We, as a community, love to talk everything to death. The sheer size of the community guarantees that there will be a diverse range of opinions on every single issue. However, I've noticed that centralized discussion in the last few years tends to go one of two ways: everyone squabbles for many days and teh whole thing goes unresolved because there is no consensus, or the whole thing goes ignored except for those who have an axe to grind. Sooner or later, we will be crushed by our own girth, and the whole decision-making process will grind to a halt, or close to it. The only things that can continue to move will be those things spelled out in policy in black and white, but anything grey or new will be congested by insane amounts of divergent opinions. Heh... special interest Wikipedians, perhaps?

Regarding teamwork... I'd say that Wikipedia is much like a cross-section of any other community. Some people are passionate about some aspects, and others couldn't care less. Small groups and cliques form and dissolve as time goes on. Some people get along with others, and some bicker and do everything they can to politic and interfere. For example: Operation Majestic Titan is a fairly small group that has done amazing work in the last few months, some of it solo but most of it coordinated, on articles relating to battleships. In contrast.... just go to WP:ANI or Wikipedia:Village pump (proposals); where agreement is not a common thing.

As far as how admins factor in is just as random and uneven. I've been lucky in that most of my requests for admin assistance have usually found a sympathetic ear within short order. But some admins have been downright toxic, even so far as one threatening to block me for "personal attacks" in a discussion where I opposed his point of view without ad hominem. Do I think there is teamwork between admins and non-admins? for teh most part, but it depends on whom is involved. Is there teamwork between the admins? From what I've seen, not as much as even the rest of the community, and yes, it has done some damage. bahamut0013^words_deeds 09:36, 5 October 2009 (UTC)

From what Jack says above: We're all supposed to be on the same team, admin or not. - spot on! — Ched :  ?  16:38, 5 October 2009 (UTC)

Indeed - it's often forgotten that admins are simply editors with tools. Orderinchaos 02:14, 6 October 2009 (UTC)

Some background

Though the focus was on RFA, we did look at the traditional role of an administrator at the RFA Review. Statistics can be found at Wikipedia:RfA Review/Reflect#The role of an administrator. Bear in mind that these are from over a year ago, and only reflect 194 editors (admin and non-admin alike), but they give some good insights for all that. UltraExactZZ ^Claims ~ Evidence 14:15, 5 October 2009 (UTC)

• Good link - I added it to the historic section. Some good stuff there that I had never known about, thanks ;) — Ched :  ?  16:35, 5 October 2009 (UTC)

I created Wikipedia:WikiProject Administrator/survey. ChildofMidnight (talk) 17:49, 5 October 2009 (UTC)

Making some modifications.--Tznkai (talk) 18:25, 5 October 2009 (UTC)

I was a little disappointed with this edit [1], but otherwise the changes look good. :) How will the survey be taken or distributed?ChildofMidnight (talk) 18:54, 5 October 2009 (UTC)

I used to work in market research for about three years, including pre- and post-analysis, and I can tell you from experience that open-ended questions like this are hell to tabulate, blistering hell to find actual trends in (as opposed to finding the trends the client--us--wants to find), and not terribly useful for finding focused answers that can provide avenues of action. A better approach would be to have a series of statements and require users to agree or disagree using a 1-5 or 1-10 scale. Those answers can be tabulated much more easily and lead to the more nuanced open-ended questions. → ROUX ₪ 19:00, 5 October 2009 (UTC)

Yes, I see that some of us have a preference for open ended questions and the solicitation of comments, while others like numerical results that can be tabulated, graphed, and analyzed. I think a combination will work. Feedback that can't be tabulated can still be useful in getting input and ideas that numerical responses miss. ChildofMidnight (talk) 19:20, 5 October 2009 (UTC)

I agree. Suggestions especially should be open ended. Consolidate discussion here please.--Tznkai (talk) 19:39, 5 October 2009 (UTC)

(edit conflict) One of the major problems in both Wikipedia and in market research is what colour to paint the bike shed. The more open-ended the questions are, the easier it is for there to be varying interpretation of the results and concomitant argument about which thing said by whom in response to question 7 is more important. This is a chronic problem on Wikipedia. If we first assess numerically how many people feel how strongly about a given statement, we will then have a clear path towards gathering more nuanced data in a much more focused way. Examples:

Q: What do you think about the current page protection policy?

A: I hate it because someone protected a page I was trying to fix with real facts

A: It's good except when admins protect the pages I want to work on

A: Great policy, works fine

A: It's inconsistently applied

A: All articles should be semiprotected

...and so on. How do you untangle those answers to find a clear path? It's very, very difficult, and people are paid significant sums of money to make up a trend analyse them. Trust me--I was one of them. Now, contrast:

Statement: The current page protection policy is:

	Disagree      Agree
Too restrictive	1 2 3 4 5 6 7 8 9 10
Inconsistent	1 2 3 4 5 6 7 8 9 10
Effective	1 2 3 4 5 6 7 8 9 10

So let's say a plurality of people answer '8' for 'Inconsistent'. We now know that many people find the protection policy inconsistently applied. At that time we can start drilling down with open-ended questions to get more granularity in the data, because we are then asking specific questions about how the application of policy is inconsistent. It suddenly becomes much easier to identify trends in the data--which we could theoretically then convert to another set of agree/disagree scales in order to come up with specific actions and proposals for community approval. Quantitative data is essential for identifying actually problems, followed by qualitative questions which expand on the numerical answers given. In the context of an in-person or telephone interview, these things are handled programatically by flowcharts or computer software--e.g., someone answers '8' to question 7, which will trigger a a skip pattern to then drill down for a qualitative/open-ended answer or clarification. That would be somewhat unreasonable to do here, so two rounds of polling will be necessary--once to clearly identify areas of concern, and the second to get more nuanced input and, as Tznkai says, suggestions for solutions. But the actual problems need to be identified first. → ROUX ₪ 19:44, 5 October 2009 (UTC)

Survey: Wikipedia:WikiProject Administrator/survey

Several folks have said they have ideas, but nobody will listen. Let's voice them. ;)

• Tznkai mentions above the idea of a survey - great idea, what should it ask? — Ched :  ?  16:43, 5 October 2009 (UTC)

Probably something very generic, non-detailed, and lacking in specifics to build consensus and generate ideas. Writing a proposal first and then trying to get it approved doesn't work that well, it seems. --Alecmconroy (talk) 17:34, 5 October 2009 (UTC)

The survey I'm planning will be a general survey, with no real commitment to specific proposals. I'm out to gather general data, somewhat like a gallup poll (though probably less sophisticated). I think specific ideas should be spun off on different forms. I am at the "what is wrong, if anything" stage, not the "how do we fix it?"--Tznkai (talk) 18:34, 5 October 2009 (UTC)

As noted above, I think some generalized questions may be useful in soliciting input and ideas at this early stage in the process of assessing our adminship system. ChildofMidnight (talk) 19:23, 5 October 2009 (UTC)

• I think it should gauge how the community views the role of admins. We're all supposed to be equal, with equal voices, and admins are simply granted the tools; however, in practice admins are elevated in the community, usually by non-admins. The phrasing of the question would be important, since the goal of the question would be to formulate an answer to "Are admins seen as 'above' average editors, and if so, what does that mean?" -->David Shankbone 15:17, 6 October 2009 (UTC)
  • Added two more questions that hit what I think you're getting at. → ROUX ₪ 15:33, 6 October 2009 (UTC)
• Just curious .. how is this going to get done? Are we going to handle it similar to the recent CIV poll? Use some "survey" style site that calculates the info? I like this by the way ... great work all. — Ched :  ?  23:50, 6 October 2009 (UTC)

I'm going to bother some people at the WMF about it.--Tznkai (talk) 01:11, 7 October 2009 (UTC)

This would be a fine idea. A related proposal is adding a good survey extension so that people can gather this sort of data on their own without making a todo about it. +sj+ 23:48, 7 October 2009 (UTC)

Long rambling post on the problems and initial brainstorming of potential solutions

Good project, glad to see people again at work to try to improve things. Thanks to the organizers of this.

In response to a question above, the most demoralizing thing I've encountered in wikipedia career is the inequity between well-established "full-time" users and casual non-admins.

The Problem

I've noticed, as have others, that behavior policy enforcement is kind of backwards around here.

I feel like the more time you've put in and the more trust the project has placed in you, the more you should be expected to comply with the rules. I try to always bend over backwards to accept incivility from a brand new editor-- they might be a kid, they might not realize people actually read their edit, they might not know that wikipedia isn't a free-for-all forum. An established editor should know better than to be incivil. And an admin is worst of all, because they are official representatives of the project, and incivility from them will be translated directly into animosity towards Wikipedia.

I assumed that the more time you spend there, the better we expect you to behave. But instead, it's backwards-- newbies can be blocked for anything, problematic well-established users have to have a mental breakdown before their behavior is brought in check. This is a big problem.

It's my belief that the rare admin who is incivil leaves behind them a wake of bitten newbies and casual users who grew to despise wikipedia because of a single bad apple.

For an admin, using incivility is a little bit like using racial slurs to insult someone. You don't just insult one person, you insult us all. Even if the person you were attacking was deserving of rebuke, everyone who sees the incivil comments noticed that an admin behaved in a way that would have been an instablock for a non-admin. This is utter poison for morale.

I'm a case in point-- I care about Wikipedia very much and I still contribute whenever I think I have a unique contribution to make-- but I'm don't live the life of a full time wikipedian-- in part, because the community issues, minor though they are, led me to devote more of my free time to collaborative open source software development, where things aren't particularly contentious.

Why does this problem exist?

Wikipedians are good people, admins doubly so. There might be a kernel of truth in the oft-repeated allegations that "admins take care of their own"-- but if such a kernel exists at all, it's truly microscopic. Think about it-- the admins are always at each others throat. There is no shortage of drama-filled inter-admin dispute. Neither a unified cabal nor an admin "code of silence" is at work here.

After much study, I think the answer is this:

Our dispute resolution system can't solve problematic behavior by well-established users

There are a tiny few bad apples who make it through RFA but go on to exhibit problematic behavior. These individuals are astonishingly rare-- most admins spend them entire careers doing oodles of good work, making the occasional mistake, and never ever coming anywhere close to flouting policy or causing any sort of trouble.

But there are some bad apples who become admins or well-established editors, and Wikipedia has no working system for handling them.

Admins can't handle it

Our first line of defense against problematic users is the admin corps. Normally admins handle the day to day policy enforcement (and do good job of it too). But the system only works when dealing with casual users. Whenever a well-established user acts problematically, the admin corps collapses into dysfunction.

An admin who enforces policy against a well-established user is entering a world of pain. Well-established users make friends, they form committees and teams and projects, they go to meetups, they have dozens of wiki-friends on speed dial, and they collaborate in chat, email, etc. Blocks are quickly reverted, and the blocking admin will be instantly be subjected to attacks. Personal histories will be poured over to find mud to sling. A giant drama storm erupts. With the admin corps split, no further action can be taken by the admins, the first level of dispute resolution has been exhausted, and no good has come of it. Everyone watching sees the dramafest, notices that no good came of it, and they become less likely to try to enforce policies against well-established users in the future.

RFC can't handle it

For the same reasons, RFC is completely ineffective when dealing with a problematic well-established user. If an RFC could solve the problem, the admins would have already solved it. An RFC against a well-established user will always be divisive, the results are invariably split, and RFC isn't even empowered to touch the admins anyway.

RFC isn't useless-- it lists the dispute on a problematic users's "permanent record" so that patterns of behavior problem can be more easily identified in by other users in the future if the problems continue. RFCs are effective as a form of institutional memory-- but they don't result in any direct remedy.

Arbcom is reluctant to handle it

By the time a case involving an established user reaches arbcom, it is a mess. The lucky cases reach arbcom quickly and quietly, but most have stopped off at ANI or RFC, ensuring tempers have flared and everyone has had plenty of time to dig up dirt on everyone else.

By the time arbcom gets their hands on it, there are no good options, only varyingly bad ones.

• Do nothing-- and you demoralize all the parties who have spent all the time trying to get the dispute resolved.
• Admonish-- and you demoralize one group of editors who support the admonished party, while simultaneously demoralizing everyone else for only issuing a slap on the wrist after all the drama that has been caused trying to actually solve the problem.
• Desysop a well-established user-- and you demoralize lots of people who feel Wikipedia has betrayed a true loyalist, taken the user's hard work and rewarded it with harassment, drama, and insult. You will also demoralize the arbs, who will now start to be attacked themselves.

Traditionally, arbcom has tried to do nothing or to admonish whenever possible, since those routes are usually the quickest straight-shot to cooling tempers. Ultimately, though, the same people return to arbcom time and again, and I think arbcom has started to be more liberal with desysopped.

Very very tentative brainstorming for a solution

It's probably premature to talk too much specifically about potential solutions, but I can't predict if I'll have time to type this out later, so I'll do it now.

Wikipedia desperately needs a less-dramatic process for either the community or the admins to effectively intercede earlier in disputes involved well-established users. Here are some of the ideas I'm encountered:

Lower courts

Create a new body empowered to adjudicate inter-admin disputes. A mini-arbcom that has fast turnaround time, with its decisions appealable to arbcom. But fast turnaround is essential, or else we'd just add another drama-filled step to dispute resolution.

Community initiated RFAs

A sufficiently large number population of editors can cause an admin to stand for RFA. This has been suggested before, though, and it didn't take back then.

Admins-initiated RFAs

A sufficiently large number population of admins can cause an admin to stand for RFA. This would solve the problem, but it might be excessively bloody, spurring the formation of factions and counter-factions and vendettas and alliances.

Arbcom-initiated RFAs

Arbcom begins to regularly ask parties to stand for RFA. Seven days later, it votes on who does and does not still have the trust of the community.

This would be a big improvement, but I don't know how to make it come to pass. There would still be lots of drama on the way to Arbcom, but at least things would tend to get settled. On the plus side, this option is already within Arbcom's power, no new policy needed. On the downside, arbcom is a drama-filled bottleneck.

Single-Admin-initiated RFAs (aka Thunderdome)

This is kind of radical-- I almost didn't type it out, but there's no judgment in brainstorming, so here it is.

Listen all of you. This is the truth of it. Fighting leads to killing, and killing gets to warring. And that was damn near the death of us all. But we've learned-- by the dust of them all, Bartertown's learned. Now when men get to fighting, it happens here. And it finishes here!

Mad Max Beyond Thunderdome

At any point, any admin can refer their inter-admin dispute to the community. Both admins would independently stand for RFA, seven days later the crats or the arbs close it.

This would be a complete game changer. It would immediately grant admins the power to effectively police themselves. Inter-admin disputes could be solved in a week's time, the community would quickly and efficiently remove adminship from any admin exhibiting problematic behavior. No more long, drawn-out fights, no more huge seven-month-long arbcom cases, no more problem admins. It's infinitely scalable, it's would be an end to inter-admin drama as we know it.

But there is a cost-- both sides of inter-admin disputes would be likely to fail RFA, and in the worst case, we could stand to lose a good admin for every single "bad apple" we fix. However-- But good admins outnumber problematic ones by at least 100-to-1, and disputes would burn out very quickly, letting us all get back to encyclopedia writing, rather than fester for months on end.

Thunderdome requires us to truly mean it when we say that "adminship is not a trophy". Under Thunderdome, we wouldn't just say it, we would mean it-- because we would wind up having lots of good people go back to being editors instead of admins. But it would also remove the stigma from a desysop, since everyone would be able to think of lots of people who are good editors that got desysopped. We could have greater flexibility in recruitment, since problematic members could be quickly dealt with. Right now, a desysop is usually a permanent scarlet letter that guarantees future RFAs to fail. With Thunderdome, this would change. People would be disciplined more, but they also could regain the tools a lot more easily as well.

--Alecmconroy (talk) 19:42, 5 October 2009 (UTC)

• Wow, for a long rambling post, that's a very good one! I'm speaking as a plain editor who has been here a little over a year, and it seems to me that the background discussion (ie, parts 1 and 2) has it exactly right. I think we all would need to think a while about what would be the best remedy, but I really appreciate where this suggestion is going. (I know it's a perennial, but I think having a system of term limits for admins, after which one has to stand for RfA again, is something that could, somehow, be made to work.) --Tryptofish (talk) 20:00, 5 October 2009 (UTC)

I agree that the identifying and outlining the inequity problem. The divide (and cases of double standards and different treatment) also extends to regular content contributors and admins. Resolving these issues should definitely be considered in any reform proposal.

I'm less enthralled with the proposals made. Wikipedia is already very political and I suspect making it easier to initiate desysop proposals would elevate disputes when they are at their hottest instead of encouraging the resoration of calm and the use of more deliberative processes and consideration. I think having a more structured procedures to have desysops considered might be good (perhaps through Arbcom as seems to be the direction things are headed now?). My other suggestions have been:

• • For a binding content mediation board to deal with content disputes to ease the arbcom burden and provide an open, collaborative and collegial forum to work through disputes instead of the current Wikijustice system that focuses too much on behavior and enforcing sanctions.
  • I also think some type of communication and/or warning should be always be made prior to blocks of good faith contributors (ie. non-vandals). Fly by blocking on issues that other editors and admins may be aware of can be quite disruptive. Discussion and resolution by collegial means should always be attempted first.
  • I also think navigating the various forums and acronyms is too cumbersome (ANI, AN, WQA, BLPN, AIV, etc. etc.). Dealing with issues should be made easier and streamlined (or at least directed through one page for noobs?). The Arb process, for example, has oodles of pages and the process favors Wikilawyers and those familiar with dispute processes (ie. those in lots of disputes) who spend lots of time working up diffs and navigating all those pages as opposed to content contributors, newcomers, and those not interested in engaging lengthy and time consuming feuds.ChildofMidnight (talk) 20:41, 5 October 2009 (UTC)
• This has always been and will probably alway be an extremely tricky area. The current process of admin recall is utterly toothless since it is voluntary and the admins themselves dictate the criteria for their own recall. Thus, ArbCom ends up getting stuck with most of the calls for desysopping, and that bothers a lot of folks who feel the community at large should make such decisions. There probably should be a way for the community to challenge admins directly when there is a feeling that they have lost the trust needed for use of the admin tools, but we don't want to make it too easy or it will become mired in baseless cases from every person who feels their article was unfairly deleted. It should be abundantly clear that any such process is only for the very worst kinds of admin abuse such as out-of-control wheel warring or numerous unjustified blocks. Certification such as is required for a user conduct RFC should be used to filter out baseless or malicious cases from being filed. In my short tenure as an admin I have found that most of my colleagues are very careful about such things and are mortified when they do make a mistake such as an unjustified block or deletion. What we don't want is a process that would make admins feel like every action they take could land them in "WikiCourt," but it seems clear that many in the broader community would like some sort of process. I recall there was one admin a month or two ago who voluntarily re-submitted themselves at RFA due to complaints about their actions, and they were pretty much told that they were wasting everyone's time and they should get back to work. Beeblebrox (talk) 21:26, 5 October 2009 (UTC)

Beeblebrox mentions "out-of-control wheel warring or numerous unjustified blocks". I think that the original post by Alecmconroy was actually bringing out another, very useful point (see "The Problem", above): the admin (a tiny minority) who is incivil and drama-mongering. It seems to me that our existing policies do know how to deal with "bright line" misbehaviors such as wheel warring, but the problem discussed here in this talk thread is something else. --Tryptofish (talk) 22:52, 5 October 2009 (UTC)

Good point. Yes, it's worth noting that in most areas, our dispute resolution does work. Serious "bright-line, out-of-control" admin abuse does get handled, no matter how well-established the offender is. Gray-area tool misuse does get reversed, although we rarely prevent future occurrences. Content dispute resolution works. In time, we'll solve this "well-established yet problematic user" glitch too.

Tryptofish is correct-- the unsolved problem isn't people who cross the line. The unsolved problem is how we handle people who know exactly where the line is, live right on the edge, and push the envelope. We're getting better now, but there was a time we just gave them infinite warnings, over and over, until after stepping on hundreds of toes, they finally miscalculated / lost composure and stumbled across a bright line.

The problem is this: What should we do with admins who, without gross tool misuse, are nevertheless hurting the project? What should we do with people who showed promise at RFA, but ultimately turned out to just plain not have the knack for being an good admin?

I don't know what we should do, but I know the status quo isn't handling it. Right now, the only thing we can do is extend their admin contracts indefinitely, regardless of whether they turn out to be suited for the job or not. We can do better. --Alecmconroy (talk) 23:52, 5 October 2009 (UTC)

Yes, that's it exactly. And I'm very glad that you have articulated it. Thanks! --Tryptofish (talk) 00:12, 6 October 2009 (UTC)

• Good points all around. However, with regard to making admins re-submit to RFA on a timed schedule, I and many others are dead-set against such an idea. I was opposed to it before I ever even ran at RFA, as I had already seen from the outside how brutal it can be. It's a complete Wiki-rectal exam, done with little to no anesthesia or lubricant, and the idea that you might have to run for re-confirmation might scare away otherwise qualified users. The only other method I can think of is some sort of notice board for reporting admins who have not crossed the line into obvious career-ending errors but are nevertheless no longer a "net positive." The problem there, which I touched on earlier, is that we need to find a way to keep it from being an admin witch hunt through some sort of certification or vetting process before full-on drama ensues. Beeblebrox (talk) 00:38, 6 October 2009 (UTC)

Agreed about the automatically time-triggered RFA. The plus is that it would let us re-evaluate without any stigma attached, and it's logical that we would want to remove those stigmas-- but even logic must give way to physics. The population of admins is so large that intelligently re-evaluating them all on a regular basis is not only unneeded, but probably impossible. 1600 admins and only 365.25 days in a year. --Alecmconroy (talk) 06:06, 6 October 2009 (UTC)

• Thank you for this summary. I also don't like automatic time-triggered RFA renewals, as I think they will be drama magnets. The community grants the tools, and I believe the community should be able to take them away; however, for a de-RFA to proceed, I think it should require 3-5 uninvolved admins to open the process, with the community voting. I don't think average editors should be able to do so, as it opens up the process too much to problems with sockpuppets and meatpuppets. -->David Shankbone 15:22, 6 October 2009 (UTC)

• Check out WP:RFC#Request_comment_on_users. There is a certification process used there to help filter out unwarranted cases that has clearly spelled out criteria a case must meet before it goes forward, I think something like that could be used in these cases as well. I suppose what we are suggesting here is an "admin recall rfc with teeth," I think we can take some hints from what works and what doesn't with existing processes. Beeblebrox (talk) 17:23, 6 October 2009 (UTC)

• Just so we have something to look at while we discuss, I've knocked up the beginnings of such a noticeboard at Wikipedia talk:WikiProject Administrator/Admin RFC draft please feel absolutely free to edit or comment upon it. Beeblebrox (talk) 18:30, 6 October 2009 (UTC)

• I looked at Beeblebrox's draft Admin RfC, and I definitely think that this is a good idea that merits further consideration. --Tryptofish (talk) 19:10, 6 October 2009 (UTC)

I've just expanded it it quite a bit to include a rough outline of what such a process might entail, and what kinds of request would not be appropriate. Let me know what you think. Beeblebrox (talk) 20:54, 6 October 2009 (UTC)

Mutual (and community) support

One thing that I think would be very helpful for the entire community would be some sort of "subject matter expert" list. I will often find myself in a situation where I need to find out some arcane bit of information about templates, or images, or copyright; these are all areas where I know my skills aren't anywhere near that of many of my colleagues. On the other hand, we have some administrators (and editors) for whom particular areas are a passion, and they are true subject matter experts. It would be excellent to have a list of people who have these special skills and a willingness to share their knowledge with others. Some areas where this might be particularly useful:

• Assessment and procedures related to deletion discussions
  • Scripts, assessment of consensus, "Your first deletion review"
• NFCC for images and other media
  • Understanding the policy, what the different tags mean and which ones are most correct, what happens when an image goes to Commons
• Editing in the MediaWiki space
• Useful scripts for editors and administrators, what they do, how they do it, and where to get them
• Database reports and what to do with them
• Dispute resolution and arbitration enforcement
• Code writing (including subspecialties for different types of code)
• Rescuing traumatised newbies
• Template design and refinement

As I note, I wouldn't necessarily limit this "resource list" to administrators, as there are many expert non-admins around here too. Some of the points, however, are somewhat admin-specific. Thoughts? Risker (talk) 21:06, 5 October 2009 (UTC)

Personally, I am utterly unashamed of my ignorance in complicated parsing of templates and so forth, and I usually use {{helpme}} if I can't find the answer on my own, and the nice folks who answer such calls have never failed to resolve my issue for me. However, I really like the idea of having a go-to list of users who will try to help traumatized newbies. Beeblebrox (talk) 21:30, 5 October 2009 (UTC)

I enjoy the idea of a newbie rescue squad as well.--Tznkai (talk) 21:38, 5 October 2009 (UTC)

I think trying to organize our resources, human or otherwise is something that could be really helpful. Not sure how to balance it without excessive bureaucracy, but I think this is something that we should focus on.--Tznkai (talk) 19:06, 6 October 2009 (UTC)

Wikiproject:Wikipedia Experts with sections for people to sign up for expertise in various areas. Keep it a single page and just a list of names (similar to WP:HAU is generally what I'm thinking). → ROUX ₪ 19:12, 6 October 2009 (UTC) On second thought, a page for each 'discipline' (with members lists transcluded to the front page) might make sense; users can then use the individual talkpages to ping for help. → ROUX ₪ 19:13, 6 October 2009 (UTC)

I like this idea. Some additions: provide a more organized way for people to pick out useful userboxes/cats to describe themselves, their on-wiki interests, and their capacities. (oh no! another profile-building interface...) On the WP:WE page (for instance), rather than having a list of names by section, have a list of categories by section; each category being a list of names. Encourage newbies and others to check out such a list. I would find it quite useful myself... when figuring out how a quirky nested template works, or a linked animated image, or how to modify a tricky easyTimeline. +sj+ 01:02, 8 October 2009 (UTC)

There is a related piece to organizing resources: finding people who know how certain off-wiki wikiProjects work: where people are building wikireaders for offline printing, or gear for a conference booth, or making WP ads for their national news media. Who has access to the [a] T[t]oolserver, who works on press releases and announcements, who builds vandalbots, editing bots, upload bots for incorporating great databases into the projects. It would be interesting to see an en:wp-focused list of Initiatives that people are engaged in. While this moves a bit farther away from Administrator lists, these are also people with privileges not granted to everyone, because they are trusted to have access to them - whether it is a bot account, participation in a contest jury, or access to free services that a third party is providing to benefit Wikipedia. +sj+

Wikipedia:Arbitration Enforcement

We need to re-evaluate admin role in Arbitration Enforcement. FloNight♥♥♥ 21:25, 5 October 2009 (UTC)

In my opinion, Adminship became a big deal when admin started enforcing Arbitration sanctions that were given to established users. Are Admin able to do this type of work well with the current policies and training that they get? FloNight♥♥♥ 21:32, 5 October 2009 (UTC)

As I've said before in WP:AERFC, AE does not work well at all.--Tznkai (talk) 21:34, 5 October 2009 (UTC)

To expand on that slightly, what we do now is roughly equivalent to peeking into someone's school records, saying "well, (s)he seems like a decent guy/gal," then minting them judges without bothering to pesky things like law school or training. Established policies (various) and practices (warn, admonish, block) don't give you the grit or flexibility you need at AE.--Tznkai (talk) 21:37, 5 October 2009 (UTC)

If I may say, some of the problem may stem from the way that the ArbCom structures its remedies. Often, instead of dealing with a problem case directly and firmly, the remedies will say something to the effect of, "no block now, but if this happens again then there will be some blocks." Actual decisions on whether to place a block or a topic ban are left to admins who may not have as full an understanding of the issues as the ArbCom did. So one solution for the problems at WP:AE would be for the ArbCom to rely on it less.   Will Beback  talk  23:30, 5 October 2009 (UTC)

I agree very much with what Will said, very good point. I wonder whether the haranguing to which Arbs get subjected makes them hesitant to just "block now"? If so, there is a need for the community to step up and communicate our support. --Tryptofish (talk) 00:17, 6 October 2009 (UTC)

I agree with Will entirely. I've never understood why arb-com hands back the power in remedies to admins. If us admins were so good at handling that sort of power, there wouldn't be an arb-com. Hiding T 09:39, 6 October 2009 (UTC)

• I agree with a lot of what is said on this thread, but I think that ArbCom involving itself in issues that are more pedestrian, as opposed to the more vexing problems, leaves them open to a lot of the haranguing. As I said recently at ArbCom, that the Committee is handling everything from whether to block the Church of Scientology to whether a person protecting their friend destroyed the community's trust too much is pretty silly. A small town might have the police (admins) and a court (ArbCom) but we're a complex community that needs different levels of dispute resolution. Not just the 'police' and 'supreme court'. -->David Shankbone 15:27, 6 October 2009 (UTC)

Speaking as someone who got completely burned out on ArbCom Enforcement, but recently re-added it to my watchlist because SOMEONE has to do it, and more and more people are just letting it go because THEY'RE getting burned out.. let me say this.

No one really wants to do ArbCom enforcement. It's all the pains of ArbCom cases (Already heated disputes, tendentious editors, used to settle score), and you get tons of brickbats from all sides.

It's fairly settled case law that AE actions are not to be undone without discussion/consensus, but admin wars over AE actions happen more often then I would like. But here's the thing. We can only do what ArbCom says we can do, so if we get vague or unclear information, then we don't know where we've crossed that line until the line has been crossed. Remember, Garbage In, Garbage Out applies to Wikipedia as it does computer programs.

So, how to get more administrators to do AE work? Good question. I thought about suggesting a mini-election for AE, but anyone who would voluntarily campaign for such a position is obviously patently unfit for a position that requires good judgment, common sense and a steel soul. But if AE isn't going to grind to a halt soon, we need more people to step up. Right now, I can only think of 2-3 editors who want anything to deal with it, and it chews people up and spits them out.

I also thought about asking ArbCom to do their own hatchet work, but then again, ArbCom burns out at probably a faster rate then AE does. It may be that the current AE system is the worst possible system, but the only possible system which works? SirFozzie (talk) 23:39, 7 October 2009 (UTC)

One option might be to have the ArbCom or community appoint or elect "enfocers", admins or others who have demonstrated an ability to effectively resolve such matters, and have these individuals resolve the matters, possibly through consensus of their numbers. It could, in effect, be a "lower court". It would also be a good "training ground" for individuals who would be interested in ArbCom itself. I can and do see how there might be a problem in getting people want to join the group, but I personally wouldn't mind at least for the start including all active former ArbCom'ers who retired in good standing and maybe the leaders of MedCom. It would be some sort of start, anyway. John Carter (talk) 17:54, 13 October 2009 (UTC)

Duplicate project

This project seems to resuscitate Wikipedia:WikiProject on Adminship which was abandoned. Graeme Bartlett (talk) 21:30, 5 October 2009 (UTC)

My reading is that project was about how we select admins and proposing new methods. This is rather more broad. → ROUX ₪ 21:39, 5 October 2009 (UTC)

A draft article about Wikipedia administrators

Hi! I'm writing a draft article about Wikipedia administrators: User:Varks_Spira/Wikipedia_administrator I would appreciate, tremendously, any help you could offer in moving the article forward. Cheers! Varks Spira (talk) 02:42, 6 October 2009 (UTC)

The Bystander effect and Uninvolved Admins

Okay, how about this. The people best qualified to solve inter-admin disputes are uninvolved admins. (individual editors might not be sufficiently familiar with the rules, and arbs are too overworked as is).

Perhaps admins CAN police other admins, so long as they don't self-select who decides what issue. What if we created an automated process that could randomly ask specific active admins to help out with specific disputes.

When help is needed, one strategy is to shout for help to anyone within earshot. However, studies show that a significantly more effective strategy is to request help from specific individuals.

Right now, we use the "Anyone, help" strategy, for example at ANI. And just as the bystander effect predicts, despite 1600 admins, we usually see the same faces over and over again when people seek help for inter-admin disputes. It's good that those admins who help out are willing to do so, but it would be better if we could get a larger pool of admins to participate.

But at the same time, we can't have the parties themselves asking admins to adjucate specific disputes, as the parties would have a motivation to seek out admins predisposed to agree, when we need the admins who are uninvolved.

So perhaps we could create a bot that, upon a certified request, could look at what admins have edited recently, randomly choose a few, and then ask them to participate in resolving a specific dispute. The same commonsense rules for who is uninvolved would still apply, and arbcom would still be there as a last resort in case it all went horribly wrong.

Of course, recruiting uninvolved admins is only part of the puzzle. Until we develop a mechanism for making sanctions against well-established users actually stick, it's all academic. The best judges in the world can't help if any one of 1600 admins can veto the remedy. --Alecmconroy (talk) 06:33, 6 October 2009 (UTC)

• I like the way you are thinking, but I would have to consider the bot idea more closely. -->David Shankbone 15:28, 6 October 2009 (UTC)

Part of the inherent problem with Wikipedia that will not go away, and I think this is a fine thing, is the drop-in drop-out nature of it. If I stopped editing tomorrow, you have no recourse against me - you shouldn't have any recourse against me. I have no problem with having higher standards of conduct for admins, but I am very hesitant to tell them to act like this is their job. We'll start losing good low drama admins who edit at the margins of their time, and we'll only entice admins and editors who are already Way To Into Wikipedia to get worse. Its the group that takes Wikipedia too seriously that is the problem, not those who don't take it seriously enough.--Tznkai (talk) 17:04, 6 October 2009 (UTC)

I agree. It's the people who think the future of humanity rests on this website--both critics and editors--that causes so much drama-mongering and game-playing. People lose perspective because they surround themselves with other people who think and feel the same way, never realizing that the average person really could care less about all of our drama. I still think the basic philosophy that the community gives tools, and should have the right to take them away, stands. -->David Shankbone 17:34, 6 October 2009 (UTC)

There is an group (that unfortunately seems to be expanding) of both rank and file editors and admins that don't even work with articles at all, just argue endlessly with each other on ANI, RFA, ArbCom, etc. These are the users and admins who seem most likely to be at the center of raging Wikidrama. In any event, randomly selecting uninvolved admins could prove tricky. A lot of admins are inactive, and other (like myself) concentrate on doing actual admin work, protecting pages, speedy deletions, etc and prefer not to get directly involved in the drama. Perhaps a list that willing admins could add their names to: "Admins willing to mediate disputes between admins" or something. Beeblebrox (talk) 17:38, 6 October 2009 (UTC)

The largest issue I see with the idea of this thread would be the comment by Beeble, "A lot of admins are inactive", which is a whole other thread on its own. His other comment rings true, and is the reason we all hate jury duty: We don't care (unless you do). The category is a good idea, but the fact that it's volunteer seems to defeat the idea of random selection of the users; we're still going to see people who want to work in 'conflict resolution', which introduces the bias this thread is about. --Izno (talk) 18:32, 6 October 2009 (UTC)

I have heard both that we have too-few admins and too-many. What's the sweet spot? Is there an algorithm we rely upon, like 'x admins for every x editors'? That would help us recognize whether inactive admins are contributing to the inertia in moving the project forward on several fronts, and also illuminate the too-few/too-many issue. I also agree with Beeble about people who contribute little content, and I've heard that presented as a reason why the people in WMF seem out-of-touch with the site: they don't really contribute much to its articlespace. -->David Shankbone 18:38, 6 October 2009 (UTC)

That is definitely one of the big ideas and countless kb of text have been spent on it and many more have yet to be spent, which is why I made the comment that that is for another thread.
I, however, am not sure that I can agree with the noted 'causation' by "too little content" contributors (correlation? maybe). Some people's talents aren't in contributing content; mine sure isn't (yet?), and I would say that adminship isn't necessarily going to attract people who contribute to content (rather than maintenance). In other words, managers.
Some managers are good managers: they get stuff done quietly and quickly. Some aren't: they curse and yell and swear and micromanage. --Izno (talk) 18:50, 6 October 2009 (UTC)

We need more good admins, and we need less bad admins. This "content" and "non-content" distinction is more distracting then helpful. If someone's only contribution is the effective mediation of problems, I have no problem with them. If someone's sole contribution to the project is reverting vandalism, the same, and if someone writes extensively on Buffy the Vampire Slayer let them stay and be valued. There is no real need to try to say one group is more valuable than the other, and I'm not sure if any group is more inherently "in touch" than another.--Tznkai (talk) 19:02, 6 October 2009 (UTC)

Agreed. "No summonses for jury duty" is an essential part of any uninvolved admin recruiting. We could do a better job of asking uninvolved admins for help, but heeding the call has to be voluntary. If a particular admin is invited but doesn't get involved, that's 100% understandable-- no harm, no foul, no stigma. The basic point is that if we could somehow match uninvolved admins with unrelated disputes, it could only help. --Alecmconroy (talk) 23:26, 6 October 2009 (UTC)

I've suggested this before. Use some mechanism (list/category/bot) to ask a random admin (who is currently active and self-identifies as willing to help) if they could pitch in with Problem X. At least trying to draw people in, with an element of randomness, is better than relying solely on whoever happens to be lurking around. Rd232 ^talk 09:25, 9 October 2009 (UTC)

More Admins? Fewer?

In the talk thread just above this one, David Shankbone asked a good question that reminded me that I was wondering about it too: Do we have too many admins, or too few? I'm not one myself, but I too notice a lot of comments around the site on both sides of the question. I'd be very interested to hear from those of you who are administrators what you observe. Are there areas where we never seem to have enough admins to get the job done, in your experience? Should we, as a community, be doing more to try to recruit new admins to work in such areas? --Tryptofish (talk) 19:16, 6 October 2009 (UTC)

It seems to me that we don't actually have many admins with a deep technical knowledge of template markup, CSS, or JavaScript. 「ダイノガイ千？！」^{? · Talk⇒Dinoguy1000} 19:46, 6 October 2009 (UTC)

The nature of the 2nd and usually the 3rd is that they are rarely changed anyway, so I'm not sure that's a large issue. Those who know the 1st also know the 3rd (or maybe the other way around?), furthermore. --Izno (talk) 19:59, 6 October 2009 (UTC)

• I think it's an aspect to consider, but I also think it isn't very relevant to the question at hand, which is should the community be able to desysop an admin. Really, that's the fundamental question: "You appoint them to have tools, should you also be able to take them away?" -->David Shankbone 20:02, 6 October 2009 (UTC)

A few clarifications about what I was trying to ask. Dinoguy and Izno, I guess I was looking more towards whether there is something like a particular noticeboard or a particular kind of administrative action where there is often a backlog of requests, and current admins find that it would be useful to have more new admins who are able to help with those tasks. David, I agree with you entirely that the desysop issue is the more critical one, and I did not intend to imply by acknowledging you that you agreed with me about asking the question that I asked. But, while we are looking at a variety of issues related to improving administrative function, I do think this question is another useful question to ask. --Tryptofish (talk) 21:37, 6 October 2009 (UTC)

Which is how originally interpreted your request; my reply to him was me giving my opinion that those areas don't really qualify for your question. :P --Izno (talk) 22:17, 6 October 2009 (UTC)

Good, thanks! --Tryptofish (talk) 22:20, 6 October 2009 (UTC)

A little off-topic, but since we mention. In an ideal world, we want far more admins. There are 1.8 billion English speakers, so I'm going to guess there's far more than 1600 people who could use the tools responsibly.

Realistically though, it's complicated. I think RFA is far far too strict and stressful, but it has to be right now because adminship is for life (barring "high crimes or treason"). I think most of our editors could do a good job, but right now, we can't risk giving them a crack at it.

So, eventually we want more, but that can't happen in a vacuum. --Alecmconroy (talk) 23:37, 6 October 2009 (UTC)

Of course we want more. Since we should think of adminship as nothing special (WP:NBD), despite what some people do think, we should try to achieve a state where every experienced user wields the mop. Of course RFA needs to be strict to avoid giving the tools to people unfit for them (which there are plenty unfortunately) but then again RFA should only be about that. "Do I trust X to follow the rules and handle the tools responsively? Yes, okay, then they can have it". Concerns like "too young", "too little content work", "too deletionist", "too inclusionist", "too much vandal fighting", "too little experience outside Y and Z", "has a userbox I don't like"/"has a point of view I disagree with", all those things have their place somewhere but they have no place at RFA. If the candidate's contributions show that they do not know deletion policy, that they don't know when to block or when to protect, then that is a legitimate reason to deny them the tools. Everything else isn't and unfortunately, not everyone !voting in an RFA understands this. RFA should not be a popularity contest but sometimes it is and that is when the system fails. Sorry for the rant. TLDR: Yes, we need more admins. I suggest everyone to approach people you see doing a good job and simply ask them if they are willing to run for it. I think a fair number of potential admins are not aware that they should get the tools and need someone to encourage them. After all, there is no harm in asking. Regards SoWhy 10:50, 7 October 2009 (UTC)

I agree that: Ideal Number of Admins = Experience Editors - Editors Who Would Abuse Tools.

This wasn't originally my idea, but perhaps it is time to start voting for everyone "Yes" on RFA where I would support the candidate if a suitable de-admin process, with the confidence such a process will emerge when needed. Or better yet, a "share-alike" inspired license: If an experienced admin with no prior problems with agrees to abide by some specific involuntary recall process that meet the Shankbone points, vote yes. If they do not believe the community should be allowed to recall them, vote no.

Has anyone noticed any admin-open-to-recall criteria that don't rely upon the admin's own subjective judgment? --Alecmconroy (talk) 11:12, 7 October 2009 (UTC)

I'm proud of mine.. User:SirFozzie/Accountability and that I have gone through two recall attempts. It's necessary to balance accountability to the community with not making it easy for one lone editor to drive you out. But if either of them had reached the threshold, I would have resigned the tools and run for re-RFA. I do not feel that a one-size-fits all model works here. Too easy to game, and it'd be best for each administrator to go with their own conscience. However, once a request forms, you'd better abide by those terms .I see no problem with learning from the experience and tightening things up afterwards, for example, but you can't move the goalposts while the ball's in play, if you want a horrible sports analogy. SirFozzie (talk) 00:04, 8 October 2009 (UTC)

SirFozzie, I suspect you know this without my pointing it out, but the issue is with those admins (a distinct minority) who do not make themselves available to recall but should. --Tryptofish (talk) 00:29, 8 October 2009 (UTC)

(outdent) I find it interesting that no one yet has really answered the question I asked at the top of this section. Maybe we really have too many admins? :) --Tryptofish (talk) 00:29, 8 October 2009 (UTC)

Honestly, in a lot of areas, like AE, above.. we have nowhere near ENOUGH admins. As to your other point, I would say that I encourage every administrator to come up with their own idea (even if it's nothing at all), and to stick by them. But I do not think that a de-admin one size fits all process will work (call me a cynic, but I think it will be just another tool in the constant bickering that goes on.) SirFozzie (talk) 00:35, 8 October 2009 (UTC)

About "one size fits all": I don't think that describes accurately what we are trying to devise here. It's no more one-size-fits-all than the current RfA system is. --Tryptofish (talk) 17:11, 8 October 2009 (UTC)

I agree with SoWhy - Adminship is most useful when it is no big deal, a set of tools available to every respectful, experienced community member... and no big deal to lose either. If you know how to use admin tools but have used them inappropriately, I can imagine the result being like a topical edit ban: no more use of the tools for some period of time. +sj+ 03:40, 8 October 2009 (UTC)

Information please...

I think whoever is chairing this project should inform every active WP administrator that this discussion is taking place here. Group discussions which are directed at the future enactment of rather broad changes, without "full disclosure to all concerned parties'" as the case may be, can be off-putting. Information for the benefit of complete transparency in the process, if for no other reason. Comments and ideas, are of course welcome! Regards, Hamster Sandwich (talk) 19:18, 6 October 2009 (UTC)

My apologies to all that were not informed of this personally. I spammed a few folks who I had seen in recent discussions on these matters, and posted to WP:AN, but failed to notify all admins. and editors. I don't know as anyone is a particular "chair" of this, but LesHeard vanU, Tznkai, Tony1, Roux, David Shankbone, Beebs, FloNight, EVula ... geesh .. the list is so long. I'll post something at WP:CENT about it. Again, my apologies to all that I didn't interact with directly, it wasn't my intent to slight anyone. — Ched :  ?  23:35, 6 October 2009 (UTC)

Agreed- everyone wants wider participation. If people haven't heard about it yet, it's just cause the project is like a day and a half old. :) --Alecmconroy (talk) 23:40, 6 October 2009 (UTC)

Now added to WP:CENT Anyone that thinks it should be posted elsewhere, feel free please - one of the Village Pump pages maybe.. — Ched :  ?  23:46, 6 October 2009 (UTC)

Basic principles

I haven't spent a lot of time considering the different, prior proposals, but off-the-cuff I believe any new process should have these basics (and I'm open to changing my mind):

1. Any admin recall process (RfDA) should mirror the RfA process
2. X number of uninvolved admins are required to begin the process
3. !Voting by the community

In terms of 'selling' the idea to the community, I think it simplifies things to say, "We have a Request for Adminship, it's time to institute a Request for De-Adminship". Relating the new process to one people already understand might help things along? -->David Shankbone 19:20, 6 October 2009 (UTC)

See WP:RFDA and WP:CDA for two proposals which hits those points. I personally believe the two should be merged. → ROUX ₪ 19:25, 6 October 2009 (UTC)

Ah - right you are. With CDA I like that ArbCom can automatically open one, but let the community decide. I'm not an admin, and I have no desire to be one b/c I like content more than management and politics, but I do feel that allowing any Editor-In-Good-Standing (which is a low bar) opens it up too much. I think most of WP:CDA is already found in WP:RFDA, which was never put forth for a !vote. -->David Shankbone 19:49, 6 October 2009 (UTC)

I publicised it around, but there wasn't all that much interest; there's much more discussion here. → ROUX ₪ 19:55, 6 October 2009 (UTC)

Maybe we just need to work off that instead of re-inventing the wheel? It had heavy participation and looks solid (I need to go through it better before I commit). Perhaps this project should take that proposal and revise it with whatever is in CDA that seems meritorious? -->David Shankbone 19:58, 6 October 2009 (UTC)

I'm happy as long as there is a robust process for desysopping which may be initiated by the community at the end of the road. I don't feel ArbCom needs to have the option of starting one--they can desysop via motion or case. If ArbCom wants community input, they solicit it. What is needed is a way for editors to take back the trust they had given. Feel free to move the page from my userspace into a subpage here. → ROUX ₪ 20:04, 6 October 2009 (UTC)

Proposal: Adopt WP:RFDA as our starting point

Further to the discussion above, I'd like to propose we adopt WP:RFDA as our starting point for a system, as it has heavy participation, has never been proposed for a !vote and seems to cover what we all mostly want. -->David Shankbone 20:09, 6 October 2009 (UTC)

I looked at it, and I think it covers some of what we have been discussing here. On the plus side, it provides a reasonable process for desysopping, and it appears to be more practical than some of the other methods that have been discussed in this talk. On the down side, I feel that the list of reasons for filing an RFDA is really unsatisfactory in light of what we discussed above in the "long rambling" section. Almost everything on the list is the kind of bright-line misbehavior that we already can deal with through existing mechanisms. The problem, instead, is the rare administrator who loses the community's faith by repeatedly acting badly in ways that do not quite cross the bright lines (please see above). What comes closest on the RFDA list is something called "administrative bullying" which links, uninformatively, to "Administratorship is not a trophy." --Tryptofish (talk) 21:51, 6 October 2009 (UTC)

P.S. It occurs to me that the draft being prepared in parallel by Beeblebrox (see above), does a better job of defining the reasons for a recall, whereas RFDA has a better-developed process for going through with the recall. Rather than adopting RFDA, it might be better to combine it with Beeblebrox's version. --Tryptofish (talk) 21:57, 6 October 2009 (UTC)

Yeah, I think I agree - I just meant adopting it as a starting point, and we can merge in Beeblebrox's good ideas along with WP:CDA's finer points. But this one seemed to be the one with the most input from a variety of established editors. -->David Shankbone 22:14, 6 October 2009 (UTC)

• I added a proposals section to the project page, added this, the survey, and the conduct thing to that section. Glad to see so much interest joining in collaboration on this. — Ched :  ?  23:44, 6 October 2009 (UTC)
• My draft page is meant only to be an example of what this might look like, to be a sort of "sounding board" to bounce ideas off of. I have no problem with using RFDA as a starting point. I think we should, however, move slowly and carefully here and not rush this to a vote. It's not going to be easy to come up with any sort of proposal that will not be broadly opposed as WP:CREEP or any of the other canned responses to such proposals. Beeblebrox (talk) 00:11, 7 October 2009 (UTC)

Purely tactical oppose. I support whatever works, but in my experience, it's not enough to build a good policy and show it off to people, expecting they too will see how good it is. The community-building and consensus-building aspects of policy development are far more important. It's not the wheel that needs inventing, it's the consensus to even create a wheel that needs inventing.

Instead, maybe we should find the most basic principles and develop a strong consensus that "something like this would be good provided it had the right safeguards". Develop a clear mandate that an RfDA process should exist, and then incrementally build in the details. --Alecmconroy (talk) 00:44, 7 October 2009 (UTC)

Granted, I've violated this very advice about 10 times in the last 24 hours, e.g. "two 'mins enter, one 'min leaves" :) --Alecmconroy (talk) 00:44, 7 October 2009 (UTC)

Or, we could take David Shankbone's simple three points and require an arbitrarily high number for initiation. 20 admins, or 50, or 100 necessary to certify. Propose it and get it passed. Absolutely no danger that it could upset status quo, but we would at last have it established that there IS a consensus that the community can revoke adminship. But since Tznkai et. al. certainly seem to know what they're doing, probably best to view the survey effort as "having the football" at this time. --Alecmconroy (talk) 10:34, 7 October 2009 (UTC)

Basic Principles and others

David Shankbone above proposes that the broad outline of a process should be:

• Any admin recall process (RfDA) should mirror the RfA process
• X number of uninvolved admins are required to begin the process
• !Voting by the community

I strongly support these. I would also put forth:

1. Very high bar to initiation of an RfDA.
2. Not a trial
3. End date and final decision

1-- we can afford to have a very high bar to initiate RfDA. Any RfDA that isn't going to result in a deadmin probably need not be initiated. We can raise the bar very very high in order to minimize needless drama.
2-- Not a trial. Consensus that someone "just plain isn't working out" is a valid reason for deadmin.
3-- Implied by "mirror RFA" and "!voting by community" is that the process should have an end-date and a final decision.
--Alecmconroy (talk) 00:32, 7 October 2009 (UTC)

What would people think about putting up something like this:

Wikipedia:WikiProject Administrator/Requests for de-adminship (preliminary poll)

--Alecmconroy (talk) 04:29, 7 October 2009 (UTC)

I propose that you use User:Tony1's AdminReview as the board for dealing with admin conduct. He has already done the leg work to get it designed. Cla68 (talk) 07:05, 7 October 2009 (UTC)

I think AdminReview deserves a good look. I'd like to get the survey done and in the works before working on any specific processes, but its increasingly clear that we need something. At the very least, I think some combination of Risker's suggestion for increased availability of various kinds of experts (highly informal) and the AdminReview process [considerably more formal] could be made to work.--Tznkai (talk) 07:50, 7 October 2009 (UTC)

Wow! AdminReview has had a lot of good work go into it too (as have many others)! I wonder how many times, total, someone has written a proposal to address this concern. Every time I think I've seen them all, I come across a new one, and invariably, they've had great thoughts and a lot of effort. We need to convene some sort of Ecclesiastical Council where we can take lots of influential wikipedian, lock them in a room, and not let them out until they've agreed on some process for this.  :) --Alecmconroy (talk) 10:19, 7 October 2009 (UTC)

Sometimes I think we should be able to line up a bunch of proposals, one of which is The Status Quo, and whichever proposal gets the most support (by a reasonably clear margin) gets enacted. Currently the status quo wins by default way too often. Rd232 ^talk 10:42, 7 October 2009 (UTC)

Precisely. That's why I think it's important to establish some sort of consensus for change-- either via a survey, a straw poll, or a very non-controversial policy that get updated incrementally until we reach actual consensus. Picking/making a completed policy proposal and trying to pass it certainly seems logical, but that approach has been tried on multiple independent occasions and never succeeded. --Alecmconroy (talk) 10:48, 7 October 2009 (UTC)

You make a valid point, I already mentioned above that any proposal to come out of all this is likely to be opposed as process creep, but the fact that soooo many users have commented on recent "de-sysop drama" indicates that there is at least a substantial percentage of users that feels the way such things are currently handled is no good. I find myself torn between trying to "strike while the iron is hot" and waiting till we are sure we have something that's not full of holes. I think if we hold out for a clear consensus to do "something" we may end up spinning our wheels for some time and this project could loose it's momentum and wind up in the same place as every other similar attempt. Beeblebrox (talk) 17:43, 7 October 2009 (UTC)

I would rather err on the side waiting for a good proposal, then trying to jam something through now. While there is risk that everyone will get distracted by the next drama and not care anymore, it is more than balanced out by the general cooling off on both sides. Disinterest is bad for change, true, but so are rock throwing partisans.--Tznkai (talk) 19:15, 7 October 2009 (UTC)

• I would disagree at having too high a bar for initiating RfDA, it would duplicate the high level needed to process a deadminning process at ArbCom for one thing and it would discourage debate (and thus consensus) during the process if the subject were considered to have been already judged as desysop material to have been accepted. Tony's admin review, which I would also propose as the necessary first stage of the process, is only concerned whether a claim of admin malpractice has occured and needs taking further - the RfDA then looks at the general issue whether the sysop can still be trusted with the tools; this leads to three potential conclusions, yes they are, yes but they need to improve some areas, or no. The last is where the level of poor admining then decides on whether any desysop should be permanent, indefinite (with mimimum periods before submitting to a new RfA), or for a finite period. By making the bar lower to review admin conduct and actions you then have the knock on effect of lowering the bar for the granting of sysop access; people with misgivings may support if they are confident that poor admins can have the flags removed, and more people are inclined to accept/make the request as the process becomes less adversarial. That is my reasoning behind trying to get a robust and easily accessible admin review and desysop process installed. LessHeard vanU (talk) 19:26, 7 October 2009 (UTC)

• I think LessHeard makes a very good point. Basically, I think the bar should be set sufficiently high as to prevent misuse of the system to retaliate against admins who are just doing their jobs, and to prevent misuse of the system to complain about an isolated case of disagreement. But I think it is key that the criteria for desysopping should approximately parallel those for RfA. RfA consensus can rest on any number of issues falling far short of bright-line offenses, and, similarly, loss of community trust (not revenge for a justified block, not a "rectal exam" for an isolated human error) should be reflected in the reverse process. --Tryptofish (talk) 19:46, 7 October 2009 (UTC)
• I'm not convinced that a new process like Admin Review is needed when there's an existing process (RFC/U for admins) which doesn't work very well. I've made a proposal to revamp WP:RFC/U here: Wikipedia talk:Requests for comment#Revamp. Rd232 ^talk 08:35, 8 October 2009 (UTC)

Isn't the fact that existing processes don't work very well a very good reason that a new process is needed? --Tryptofish (talk) 17:07, 8 October 2009 (UTC)

Not if no serious attempt to fix them has been made. Rd232 ^talk 17:34, 8 October 2009 (UTC)

I don't think RFC can deal with what we are discussing here unless there is a fundamental change (such as is suggested in my draft page nudge nudge) giving the process some "teeth." Under the current strucure, if a bunch of users agree in an RFC that a particular admin is doing a bad job and should have the tools removed, all RFC can do is ask them to give up the tools. If they don't agree to do so, it's over and the RFC has been nothing but a circle jerk. Beeblebrox (talk) 17:49, 8 October 2009 (UTC)

That's not quite true. RFCs concluding without agreement can and should produce valuable discussion/evidence for taking forward elsewhere. Rd232 ^talk 09:21, 9 October 2009 (UTC)

RfC#Admin conduct can be another method by which concerns can be aired and then taken to RfDA. Presently an RfC needs to be certified by a third party that the problem exists, and then any and all interested parties can comment (and support others comments) including those who support the complainer and the complainee. At the conclusion an consensus may form that there are sufficient grounds for the admin to undergo RfDA. Alternatively, AdminReview can be instigated by one person with one specific complaint, and a review body will then decide if it is justified in taking forward. After then hearing both sides a decision is made regarding the matter, which may be to RfDA. These are two different approaches to reviewing whether an admin is using their enhanced privileges appropriately, and there is no reason why both cannot be used to initiate a process that may result in running a RfDA. LessHeard vanU (talk) 19:54, 8 October 2009 (UTC)

i'll work more on the survey over the weekend, but I'd like to see if there is support for a recall procedure at all, and if so if they'd like something more like a vote, more like a trial, or more like a supervisory body.--Tznkai (talk) 20:09, 8 October 2009 (UTC)

Well, I do like the simplicity of the Wikipedia:Community de-adminship proposal - perhaps because it so closely mirrors the current RfA system. I think there's a lot of the items that can and should be merged into either the specs, guidelines, or documentation from the other 4 as well (EVula's, Tony's, Roux's, and Mark's). I may regret supporting this in the future, but I think that for the good of the community, some sort of community driven system needs to be created. While it's not supposed to be, the fact is that "adminship" has become a big deal. The whole ArbCom thing is so complex that it intimidates even some of the more seasoned editors here; and frankly, since the community is what puts the "mop" in place, they should be afforded the ability to remove it as well. — Ched :  ?  18:23, 13 October 2009 (UTC)

I agree, and would add Beeblebrox's draft to that list. And, to underline what I and others have said before, it should be about more than just "bright line" offenses. --Tryptofish (talk) 18:50, 13 October 2009 (UTC)

(de-indenting). I hope everyone that's working on this realizes that the easier it is to de-adminship, the harder it will be to get an administrator to do anything but simple janitorial work. We've seen numerous cases lately of groups forming to push POV in sections of Wikipedia. Under these proposed rules, an administrator attempting to bring balance to a section of battleground articles would quickly run afoul of one or both groups and quickly be de-adminned. SirFozzie (talk) 19:07, 13 October 2009 (UTC)

In many ways, that will end up being the central issue of this project, and we need to get it right. But that does not mean making the recall process so restricted that it accomplishes little or nothing. --Tryptofish (talk) 19:13, 13 October 2009 (UTC)

Although the current drama that prompted all this has concluded (much quicker than I ever thought it would) for the moment this is the central point of this project. The community has called for a better process so many times that it seems the time has come to try and craft it. The points I think are the most important are making sure it is not easily susceptible to abusive/frivolous complaints, and making sure that somehow the process is community driven but does not drag on for a a million years like most ArbCom cases. Beeblebrox (talk) 20:11, 13 October 2009 (UTC)

Yes, that's exactly right. --Tryptofish (talk) 20:50, 13 October 2009 (UTC)

I am going over your heads...

I will be posting a link to this at Jimbo's talkpage. My idea is to start the ball rolling and build up some momentum as regards the concept of removing the flags from administrators who no longer have the confidence of the community. It may be best that any backlash is directed at my "personal" proposal, so that any proposal developed on these pages can take those views under advisement and be incorporated. I will, however, be suggesting to Jimbo (and thus his readership) that he looks over these pages also - it is linked quite a few times from my proposal anyway. So you had best wash behind your ears and manicure those nails. LessHeard vanU (talk) 21:19, 7 October 2009 (UTC)

Ha. Jimbo can't (won't) even block admins anymore after you know what, so what clout/gravitas is he supposed to be bringing to this party? If he supposedly isn't fit to be dishing out nailed on short blocks to seasoned admins in his capacity as monarch, then what possible credible 'role' does he have in shaping the new and improved P45 instructions? Huzzah indeed. MickMacNee (talk) 23:56, 7 October 2009 (UTC)

"Going over our heads" would be the exact reason this WikiProject was created... so that people didn't go over any heads.
Sigh. --Izno (talk) 00:22, 8 October 2009 (UTC)

Hmmm... It is a shame this website doesn't support the ironic font typeset. Fact is, Jimbo's talkpage is heavily trafficked and often by those who do not inhabit Admin or policy related pages, and I had opened a discussion with Jimbo back on 1st October regarding policy and admin desysopping. So rather than unilaterally saying, "look what I've gotten involved in!" (which would also be undue weight on my part) I concluded the essay I had already had in mind and linked to that, but emphasising the links to this project page and that of Tony1's. While community consensus is a difficult beast to manage, very little seems possible to achieve without it; and thus you need as much participation as possible. Placing a link to my own viewpoint on Jimbo's page will hopefully gain some barnstars wider interest which will usefully be channelled here. Finally, not everyone has heard that Jimbo is no longer the influence he was and approval from him is likely to resound strongly among the general editorship (y'know, the people who all this is supposed to help!) LessHeard vanU (talk) 12:44, 8 October 2009 (UTC)

I got the irony, and I think it's a good idea. Jimbo's talk page has become more a community bulletin board, with Jimbo overtones, than just a place to communicate with Jimbo. Additionally, we all like to talk things to death; this issue has been discussed to great lengths in the past, and there is motivation to move things forward now. We don't need to re-invent the wheel with a entirely new proposal, we just need to tweak the ones already worked on in prior efforts (which I don't see have ever come before the entire community). Thanks for being bold, LHVU. -->David Shankbone 13:24, 8 October 2009 (UTC)

I agree. The toughest thing for me to decide upon is "where". We have several good "de-adminy" things floating about out there. Tony has done some great work in that area, as have all the other folks who contributed to the evolution of his page. Same thing with Roux. EVula's page I thought was quite good as well - although not as publicized, or trafficed - trafficatedd (hell, I'm having trouble pronouncing the damn word - forget spelling it) ... didn't see as many folks posting to it. Good idea to post to Jimbo's page ... there's a tool somewhere that counts the number of folks that watch it - and it's a pretty big number - so it's going to get a view there. Actually it was the original post there that led me back to Mark's page and some chit-chat and thought about this whole thing. — Ched :  ?  19:38, 8 October 2009 (UTC)

Of course I knew it to be ironic. :P --Izno (talk) 19:40, 8 October 2009 (UTC)

Another one

I just noticed Wikipedia:Community de-adminship, and asked over there if they want to join in on this project. We really need to consolidate our efforts folks. We should mark a half-dozen of these things as historic, post a link to the one we're working on, and present a united effort. Just IMHO. — Ched :  ?  05:49, 9 October 2009 (UTC)

Not yet. I think we're still early enough where lots of proposals in laboratory is a good thing. My opinion anyway.--Tznkai (talk) 06:00, 9 October 2009 (UTC)

I'm fine with that. I'm one of the least experienced people here, so I'm quite content to follow along. ;) — Ched :  ?  06:11, 9 October 2009 (UTC)

I disagree with Tznkai that "were still early enough" as this idea has been around a looong time, but has never been presented. I want to again caution duplicating work already undertaken at prior stabs at a proposal, and needlessly reinventing the wheel. If anyone hasn't taken time to read those proposals, I suggest they do so instead of re-doing what has already been done (not directed at you Tznkai :-) - just general comment). -->David Shankbone 15:40, 9 October 2009 (UTC)

How many RFAs does it take to become an admin?

Does anyone know about how many people are promoted on their 2nd RFA? (How many on their 3rd? How many on succeed on their 4th?) --Alecmconroy (talk) 14:06, 9 October 2009 (UTC)

Also helpful would good data on average edit count of successful rfa from the past year or so, average length of time spent on wikipedia prior to promotion. (and of course the data mentioned above about "number of failed RFAs before promotion to admin".)

I don't know if it'd tell someone to bother to calculate it just for this, but if numbers (or good estimates) are easy to come by, they might help. --Alecmconroy (talk) 14:31, 9 October 2009 (UTC)

This is just blind guessing based on my own observations, but I bet you 2 is about average. Sadly, users who don't pass the third time tend to be perpetually opposed as "wanting it too bad," so not many have the stones to pursue it past the third try. I think the record is 7, anyone remember a nom with an 8 on it? Beeblebrox (talk) 21:04, 10 October 2009 (UTC)

7 seems to be the record, [2]. Cenarium (talk) 21:13, 11 October 2009 (UTC)

The answer is, of course, is that it only takes one. The unknown though is which one. --Malleus Fatuorum 21:22, 11 October 2009 (UTC)

The first one is sufficient, if the candidate has observed RFAs for a while, and "checked the appropriate boxes" before even applying. (That's the most cynical version--I prefer to think of it as "a candidate seeking out the requisite experience and seeing whether adminship is even anything he is interested in"). Jclemens (talk) 22:26, 11 October 2009 (UTC)

That's the case now. 'Twas not always thus, of course. Black Kite 22:38, 11 October 2009 (UTC)

So, using Cenarium's method, I see that we have admins who took four or even five RFAs before being promoted. ([3][4][5][6][7][8]) --Alecmconroy (talk) 15:30, 12 October 2009 (UTC)

If we could do it over again, what would we name the Admins?

At this point, the terms "admin" and "sysop" are so entrenched in MediaWiki culture that I don't think we could ever get them changed.

But suppose we could easily rename the group currently known as "Administrators". What should we have called them? I don't think it matters too much within the established user community who know Adminship is not a Trophy. But I remember when I was a newbie, and when I would meet someone who was one of the "Administrator of Wikipedia", the title carried a lot of weight and baggage. It certainly didn't sound like a big deal, it sounded like a major deal.

The "janitor" metaphor is one I think does do a great job of capturing the real nature of the post (particularly for we nerds who love the Weird Al film UHF). The janitor neatly evokes the image of someone who is both humble and heroic, neither combative nor passive. Perhaps we should have called them janitors from the start.

Or perhaps it would have been best to use a nonce word-- make up a brand new title to avoid ANY imported pre-expectations. "Wikiloompa" or "Kwyjibo" or something?

Think the horse is WAY out of the barn on this one, but it's worth point out past sub-optimums to learn from them. --Alecmconroy (talk) 16:10, 12 October 2009 (UTC)

I want to be a Wikiloompa!! :) Rd232 ^talk 10:16, 14 October 2009 (UTC)

Wikiloompa was my favorite too. :) It conjures of images of Willy Wonka's oompa-loompas, who go around doing good, keeping the factory running and solving the occasional industrial accident. Plus, I think there could be a lot of great songs that go with the job. --Alecmconroy (talk) 13:25, 14 October 2009 (UTC)

Lately I've been thinking WP:masochist, but I suppose a serious answer would be WP:Maintenance would be the closest thing I can think of, but as you said .. that ship has sailed. — Ched :  ?  18:10, 13 October 2009 (UTC) hmmm ... that's a blue link ... never saw that page before. — Ched :  ?  18:12, 13 October 2009 (UTC)

There would probably be a lot less drama if it were called "Requests for janitorial work." I don't mean to talk down at the non-admins here, but I have found that although I had heard the metaphor a thousand times, I didn't really understand how accurate it was until I had the mop myself. I wonder if we could discourage more "not now" nominations if we added something to RFA that explains that this is work that requires you to use your brain and that the slightest error (or perceived error) can set off a real shitstorm of drama. I think if we actually use those words "shitstorm of drama" we might be able to discourage the more obviously unqualified candidates. Beeblebrox (talk) 20:03, 13 October 2009 (UTC)

WP:Prefect? Like "perfect", but spelled wrong... LessHeard vanU (talk) 20:49, 13 October 2009 (UTC)

But not to be confused with prefect :) --Tryptofish (talk) 20:54, 13 October 2009 (UTC)

Actually, prefect or proctor are similar to the admins's functions and both are known for being officious, an accusation some make of the admin corps. BUt I don't suggest we use those names, it would only compound the problem and make this place seem like a school.   Will Beback  talk  10:09, 15 October 2009 (UTC)

Proctor isn't that familiar, it gets used a bit in Scotland for judicial roles, which is itself a disadvantage since not all will understand what it means... unless someone makes the connection with proctology. LessHeard vanU (talk) 12:36, 15 October 2009 (UTC)

Curator, guardian, caretaker, safeguard are all words I like... I have a penchant for "warden" as well, but that doesn't fit at all. --Izno (talk) 05:55, 14 October 2009 (UTC)

Custodian, major domo, steward (already taken), editor (already taken), herder, enforcer, tool-user, bit-holder, blocker, protector, mover, level 2 user, User Group Beta, trusted user, verbocrat (?). In maintenance, "custodian" is like a janitor, but with more responsibility.   Will Beback  talk  10:09, 15 October 2009 (UTC)

Nice work with tool-user, that's almost a Kenning. :) —Finn Casey * * * 22:21, 15 October 2009 (UTC)

• How about "Volunteer Jerk"? :) That's what most editors think of admins anyways. --Hammersoft (talk) 14:01, 15 October 2009 (UTC)
• I agree that this is a difficult question. The challenge would have been deciding on a name that communicated the responsibility inherant in the tools, coupled with the requisite lack of egotism and authoritarianism. I am persuaded that an original term may have been most effective, as suggested above (e.g. wikiloompa). —Finn Casey * * * 22:21, 15 October 2009 (UTC)
• I view all of this kind of work as "maintenance"—be it 'crat-work, CU, OS, AfD, merging, most any such activity thing except blocking is "cleanup on aisle 3" work, and should be done quietly and quickly—and would be happy with something like "janitor" or "maintenance personnel". -- Avi (talk) 16:30, 18 October 2009 (UTC)
• Gardener. I like the term so much I've adopted it on my own. In contrast to a 'janitor' who's job is to clean up messes, a gardener helps things grow and develop. Pruning and cleaning are parts of the job, but the overall goal is to nourish growth. henrik•talk 14:20, 24 October 2009 (UTC)
• Administrator It is basically what we end up doing here. It only means "someone who administers", it is simply the act of getting people together to accomplish desired goals and objectives. Chillum 14:22, 24 October 2009 (UTC)
• I think we'd name it "Moderator" or something like that, which is usually what users performing these functions are called elsewhere on the internet. Deacon of Pndapetzim (Talk) 09:18, 26 October 2009 (UTC)