Wikipedia talk:Requests for adminship/Archive 57

Apprentice and journeyman admins
An idea came to me the other day for an alternative to the present adminship process. While many people are happy with the way things are, the criticisms of the present process are:
 * It is a source of ill will and is generally unpleasant for many candidates
 * Some candidates are ultimately promoted even though we know little about them, because they have edited extensively and maintained a low profile
 * The process is increasingly politicized, with the result that admins are people who are good at behaving politically rather than being good at building consensus
 * Good candidates are being turned away if they have been involved in controversey, in some cases even though they were widely believed to have been doing the right thing
 * Many RFA voters find that they are voting on candidates they do not know
 * There is an overreliance on edit counts due to a lack of other metrics
 * Vandalfighters frequently become admins even though vandal fighting has little to do with most admin-specific activities

I remind those who were not here for the genesis of RFA that many of what are now hard-and-fast rules of RFA (that RFAs close in one week, that 75-80% is the promotion threshold) were arrived at originally with relatively little thought.

I propose an alternative process:


 * 1) People interested in adminship must find a sponsor who is already an admin.  Each admin may sponsor up to four candidates at a time, including journeyman candidates (see below).  It would be expected that an editor would be involved in Wikipedia for at least 2-3 months before seeking out a sponsor.
 * 2) The sponsor and the candidate ("apprentice") work together for a period of at least 1 month during which the sponsor helps the candidate learn the "way we do things."  Sponsors may retract their sponsorship at any time if they feel the relationship is not working and this may well become common.  In like fashion, candidates may retract their apprenticeship if they believe things are moving too slowly.
 * 3) When the candidate and the sponsor both believe that the candidate is ready for adminship, the sponsor proposes adminship for the candidate.  A vote much like the current RFA process ensues, with due weight being given to the judgement of the sponsor and his/her past history of successful or failed sponsorships as well as the merits of the individual candidate.  (It is expected that a "reject" outcome would be rare)
 * 4) The sponsor and candidate continue to work together for a "journeyman" period of 60 days, during which the sponsor can retract adminship if he/she believes that the candidate is misusing it (expected to be very rare and could be handled on Requests for permissions).
 * 5) At the end of the "journeyman" period the new admin may continue without formal supervision and may sponsor other candidates if they choose.

This serves several goals. First of all, it makes the process less divisive since the candidates are mainly working with one admin who is expected to know them well and work with them to be sure they have the necessary judgement, and impart the necessary knowledge. Second of all, since sponsors are accountable to the community for candidates they present for adminship, there is a focal point - someone trustworthy who has been given the task of understanding the candidate, their abilities, and their motives. Third, it provides a more well-defined process for candidates. Candidates know what to expect, and can choose sponsors who are reasonable and will work with them to get through everything. Finally, it provides a means for newly minted admins to have a reasonable level of supervision until they've been using the tools for a while.

If this is something we want to do, I would suggest a transitional period of two months during which candidates may either follow the traditional RFA process or seek out sponsorship.

The Uninvited Co., Inc. 21:41, 12 May 2006 (UTC)


 * Hmm...I'd be open to a process like this one and I'd be willing to see what a trial run of such a process looked like . &mdash; Ilyan  e  p  (Talk)  21:53, 12 May 2006 (UTC)
 * As a non-admin who might at some point become an administrator, I was hoping that there might be some sort of mentoring system in place. This sounds like a great idea to help new administrators become proficient at mopping. Kuki ni [[Image:Flag of Hawaii.svg|30px]] 22:12, 12 May 2006 (UTC)
 * Sounds interesting. Reminds me of the Old Jedi Order. This seems a bit like the "Admin Coaching" I see WP:ESP is already doing. I'd be interested in mentoring a few admin candidates if this process were to see a trial period. --Lord Deskana Dark Lord of the Sith 22:17, 12 May 2006 (UTC)
 * It will be interesting to see what y'all decide. Kuki ni 23:29, 12 May 2006 (UTC)
 * Jedi Order perhaps you compare with, but speak like yoda you must avoid. Kim Bruning 21:59, 13 May 2006 (UTC)
 * Interesting idea. Prodego  talk  00:06, 13 May 2006 (UTC)

I don't have a major problem with trying something new, but that amounts to a ton more work, especially in the candidate stage watching over the user and teaching them the ropes. I know for one I would certainly not have the time that it would take to do that. But you may be on to something with the journeyman(woman) stage where it is considered easier and less a big deal to revoke the adminship if it's not going well. Though the problems you note are there to some degree, I don't believe RfA suffers from a very great false positive problem, and what false negative problem there is could be fixed by less editcountitis and less reticence to de-admin when needed. A change such as your proposal would amount to voluminous amounts of time to change a process that's basically working fairly well. I wouldn't oppose it as an additional option for those willing and or desiring to do it instead though. Promoting admins requires reading community consensus for the candidate, it doesn't require doing it exactly through the current method. - Taxman Talk 03:02, 13 May 2006 (UTC)

The idea that rejection would be rare under the proposed process is not valid (other than the fact that the clueless newbie candidates will be gone). Because a candidate who is a bad choice for adminship will still have some supporters and friends, one of those supporters will do the process with the candidate, and when the time comes, the candidate will fail, and the coach will be upset that the community did not appreciate his coaching efforts and abilities. Ultimately, it still comes down to whether we trust the candidate, no matter how much coaching, training, or probation they go through. Every single one of the recent controversial candidacies (that were rejected for anything other than not enough experience) would have had just as much resistance under an apprenticeship type process as they did under the current process. Training is a good thing, to be encouraged, but we won't be able to use the fact that one has been trained to get around the fundamental issues that make candidates fail. NoSeptember  talk  06:14, 13 May 2006 (UTC)
 * Hmm, those fundamental issues still haven't been explicitly elucidated, (or perhaps I may have missed them). Would you care to give it a shot, maybe on a new page, and link here? It might be insightful. Kim Bruning 11:26, 13 May 2006 (UTC)

I'd love to help out with this process. To an extent it already exists, (the infamous "irc cabal" sometimes sort-of mentors people already. :-) )

Maybe we'd like to have only specific people be mentors to start with, to catch some of the problems seen by NoSeptember.

I'm thinking this might help a LOT with the actual Big Problem behind the scenes, namely: lack of acculturation. Kim Bruning 11:26, 13 May 2006 (UTC)


 * Kim's point is a valid one. The lack of acculturation, as Kim puts it, is a major factor in several particularly vexing distractions we now face that are unrelated to adminship.


 * To Taxman's point, I agree that we would have to make very sure that there are enough people willing to act as sponsors. There are some people who very much enjoy that sort of thing, and like many Wiki-related tasks, people will choose to participate in areas they find compelling to them.  Given the growth rate of the admin group I don't think the process would be unworkable, but we would certainly want to be sure there are enough sponsors willing to participate before making such a change.   I would think that we would want to compile a list beforehand, in fact.


 * To NoSeptember's point, I think that the reasons most people are turned down fit into one of four categories: (1) inexperience, (2) misunderstanding or confusion on the part of the community, (3) candidates who don't respect the way we do things, or (4) candidates with irredeemably poor judgement. I think that my proposal addresses 1, 2, and 3; particularly, it helps acculturate users so that they don't make the stupid mistakes that can result in adminship being denied, and it provides a sponsor to be sure that experience requirements are met and that the community understands the candidate.  While there are those candidates who simply have irredeemably poor judgement who we would never want as admins, I anticipate that they would have trouble finding sponsors; any who made it all the way to the community approval stage would then be an embarassment to the sponsor making the process self-correcting.


 * The Uninvited Co., Inc. 15:34, 13 May 2006 (UTC)


 * I've already found some people willing to be sponsors. We can start any time :-) Kim Bruning 15:41, 13 May 2006 (UTC)
 * I like the idea of sponsors not being friends of the candidates (i.e. we should assign coaches randomly). If you can convince a thrid party that you are ready for adminship, that would be much more credible than a known supporter saying you are ready. NoSeptember   talk  16:06, 13 May 2006 (UTC)
 * The couple of experiences I've had with random assignment weren't too encouraging. It's better to have folks decide amongst themselves. This is not a big problem though, we can sort that out. :-) Kim Bruning 21:58, 13 May 2006 (UTC)
 * I think the four reasons people are turned down mentioned above are basically correct. We would reduce the number of rejections if we prevented people making stupid mistakes, but is our goal to reduce the number of rejections? Being rejected for stupid mistakes, learning, and then passing on the next attempt, is also an ok path to adminship. Is the fact that the candidate has to fail once first - all that bad for us as a community? Apprenticeship is fine, but it should not be mandatory. There is an Admin Coaching program here, which is really just a program to bring coaches and candidates together without a bunch of rules.  NoSeptember   talk  16:06, 13 May 2006 (UTC)


 * My reply is below. The Uninvited Co., Inc. 02:35, 14 May 2006 (UTC)

What would the candidate and sponsor actually do during that month? How do you prepare someone for adminship when they can't actually do the admin tasks because they aren't an admin? What kind of mentoring would candidates get? How is a sponsor meant to spot a bad admin any easier than just going through the last months contribs before voting on an RfA? --Tango 16:40, 13 May 2006 (UTC)


 * Funny how different people have similar thoughts at about the same time. My thought was that an informal approach to mentoring might already be occurring either via "IRC Cabal" or otherwise. It might be that experienced admins might be willing to offer and that a particularly bold pre-candidate might ask a respected admin for advice or at least an honest appraisal.
 * A sponosor good help with the critical thinking part of the job and helping the candidate understand policy. They could serve as a resource for questions about, "what would I do about this? They could also ask the sponsor to explain the reasoning behind decisions that require elucidation. :) Dlohcierekim 18:56, 13 May 2006 (UTC)
 * The biggest job a sponsor would have would be to get the candidate to learn our culture, and grasp the unwritten rules underlying the wiki. It's quite surprising to see the difference in efficacy between a normal but acculturated user, and a non-acculturated admin. Note that most tasks currently done by admins do NOT necessarily require an admin bit. Kim Bruning 22:06, 13 May 2006 (UTC)
 * How do you teach someone culture? I would think the only way to learn that is to get involved. A particular person to go to to ask specific questions might be good, but I'm not sure I can see the point of having that for a forced month. If people don't have any questions, the sponsor is useless. --Tango 22:36, 13 May 2006 (UTC)
 * You teach people a particular culture by working together with them and showing them how you do things and/or explaining how things can be done better/more efficiently. Hmm, is the Apprenticeship article of any use? Kim Bruning 12:29, 14 May 2006 (UTC)
 * Could you be more specific? How do you "work together" with someone on wikipedia?  What do you actually intend these mentors to do?  Not what you hope to achieve, what they'll actually do.  What kind of things would they be explaining that can't be learned just by watching what other people do? --Tango 13:36, 14 May 2006 (UTC)
 * We'd like to formalise the "watching what other people do" part a little, to be sure the candidate has actually seen everything seeable, and hasn't missed anything. We're also introducing a whole "watch and then try for yourself" part, with an expert standing by to fix any problems if you happen to make mistakes. Basically, um, in short, well, apprenticeship. It's a kind of learning system known since the middle ages. There's probably centuries (in the literal sense) worth of articles and studies on apprenticeship explaining *how* and *why* it works. We'll just be stealing the concept and using it in a simplified form. :-) Kim Bruning 14:56, 14 May 2006 (UTC)
 * Apprenticeships are useful because the things being learnt aren't done in the open - you need a special arrangement in order to watch someone making a table. You don't need any such arrangement to watch an admin closing an AfD (or whatever) - there are even archives so it isn't time dependant either.  As for "having a go" that's what the concept of WP:BOLD is all about - we should (and usually do) encorage everyone to just have a go and someone else will come along and fix it if you go wrong.  That happens already - we don't need a formal system for it. --Tango 16:17, 14 May 2006 (UTC)
 * Indeed, in an ideal world, everything admins do would be out in the open.
 * To my immense chagrin, there are actually people out there who are constantly pushing to close, obfuscate, and hide large parts of the admin processes. This includes at least the Counter Vandalism Unit, and certain foundation employees.
 * There's also some unavoidable ad-hoc interaction, which you won't be able to track on wiki.
 * But even on-wiki, it is often difficult to reconstruct a particular incident. See how long it takes the arbitration committee to come to a finding of fact. If you only have a partial view, you can quickly draw incorrect conclusions, and get a very biased picture.
 * For all these reasons, it's actually quite handy to have an expert along, and to watch what's happening.
 * One of the key skills to learn is in fact where to look in the first place, to be able to map out the entire set of relevant interactions for a given situation. If nothing else, that's a useful skill to learn, which several of our current admins might in fact not posses :-) Kim Bruning 10:08, 16 May 2006 (UTC)

Any attempt to get rid of self noms must be opposed. It would be one of the first steps to setting up a real cabal.Geni 19:09, 13 May 2006 (UTC)
 * I recently asked some consultants to look in on wikipedia culture. They basically concluded that it was mostly clans with a little hierarchy thrown in for good measure. So self noms are basically not doing what you'd hope they do, all you're getting is multiple fractured cabals fighting each other, which isn't really useful either. I'd like to get some more ad-hocracy into the mixture, which at least *is* possible to introduce with a sponsorship system, I think. Kim Bruning 22:06, 13 May 2006 (UTC)
 * I'm guessing that their comments are (unfortunately) to remain private? If, by some chance, you could make them public, I think they'd probably make for fascinating reading. ;-) Kirill Lok s h in 22:31, 13 May 2006 (UTC)
 * Thanks for reminding me, I was meaning to ask 'em. Though hmmm, I need to find a good moment... Kim Bruning 22:43, 13 May 2006 (UTC)
 * At least self-noms allow new clans to form - without them there is a risk that you'd be stuck with just the clans that currently exist, they'd fight it out and eventually one would reign supreme and become a cabal. A rather melodramatic prediction, certainly - call it playing devil's advocate. --Tango 22:36, 13 May 2006 (UTC)
 * Oh geeze, TINC already! :-P What "cabals" there are are pretty informal. There's no secret handshakes, or initiation rites, or anything like that. I don't like the concept of clans and cabals anyway. I'd prefer if everyone to work together with everyone. I'd also like for people to be allowed to be more creative. I think we can achieve that more easily with a sponsorship system. A self nom could just as easily be a person approaching a sponsor-type-person themselves. Kim Bruning 22:43, 13 May 2006 (UTC)
 * Self noms along with outher community driven processes tend to be a an imporant element in makeing sure TINC. You can't form one because you can't be certain who is going to be promoted next week and begin to make your life difficult.Geni 00:34, 14 May 2006 (UTC)
 * Fine. So keep self noms. Kim Bruning 10:22, 16 May 2006 (UTC)


 * I think the four reasons people are turned down mentioned above are basically correct. We would reduce the number of rejections if we prevented people making stupid mistakes, but is our goal to reduce the number of rejections? Being rejected for stupid mistakes, learning, and then passing on the next attempt, is also an ok path to adminship. Is the fact that the candidate has to fail once first - all that bad for us as a community? Apprenticeship is fine, but it should not be mandatory. There is an Admin Coaching program here, which is really just a program to bring coaches and candidates together without a bunch of rules.  NoSeptember   talk  16:06, 13 May 2006 (UTC)

NoSeptember raised several interesting points above, which I'd like to discuss. First of all, NoSeptember asks whether it is a goal to reduce the number of rejections, citing what I would call a try it--fix it--try it sort of approach to adminship. The point that I would make is that while such a path is indeed valid, it is tough on candidates, because a failed RFA is frustrating and leads to wikistress for the candidate, the candidate's supporters, and all too often bystanders caught in the line of fire. In my view, a failing RFA is a rather blunt tool for helping candidates to improve.

Noseptember also points out the Esparanza admin coaching program and suggests it as a voluntary alternative to the proposal I make. While it is true that Esparanza's program offers mentoring in a less formal atmosphere, I believe that there are three key limitaitons to the Esparanza program. First of all, it is under the auspices of Esparanza rather than being a truly project-wide initiative. While this is not the place to discuss Esparanza's strengths and shortcomings, suffice it to say that Esparanza is not universally revered. Secondly, I believe that the informal nature of that program is a weakness. A key element of my proposal is a sort of sense of accountability of sponsors for candidates that they present. An informal process doesn't do that. Finally, as things stand, candidates are free to ignore the admin coaching program, and many if not most do exactly that. We don't even ask about mentorship or coaching in the standard RFA questions.

Others have raised concerns about the possible rise of clans or cabal(s) as a result of such a structure, particularly if self-noms are discouraged. I'm not concerned about this but there is the possibility that an overly insular culture could develop over the course of years and the project could get stuck in some sort of rut. I think that's a real risk albeit one that will take five to ten years to become a factor. As Wikipedia policy becomes less fluid, that sort of timeframe is more important than it was. But as things stand today, I believe that losing the valuable institutional memory and wiki culture is a far greater and more present risk. The Uninvited Co., Inc. 02:35, 14 May 2006 (UTC)
 * Wiki culture has and will evolve over time. Its evolution should not be considered a risk. Rather, the people who fear that evolution need to adapt to the reality; this project has and will evolve. For a number of reasons, I will oppose any RfA reform effort that makes self nomination impossible. Over the last 11 months, there have been 95 successful self-nominations. That's 1 of every 5 successul nominations. By forcing such people to be non-self nominations, you are further increasing the politicization of the process as candidates will now have to in essence campaign for someone to nominate them, rather than simply nominate themselves. --Durin 17:59, 15 May 2006 (UTC)
 * Meh, you can keep your selfnoms. We can run both systems in parallel.
 * Would you care for a small wager on which system will be the most successful? :-)
 * Kim Bruning 18:20, 15 May 2006 (UTC)


 * I reluctantly agree with Durin. I was a self-nom, and I clearly remember that I wasn't chummy with any admins in particular at the time of my nomination. That isn't to say the apprenticeship idea is a bad one - it sounds like it's worth a try - but we have to account for the fact that some people just edit in areas of the encyclopaedia where you don't run into admins very often. Perhaps they might observe your work, but you wouldn't be aware of it unless they talked to you. Even today, if not for IRC and the mailing list (which I now avoid as much as possible due to a lack of time), I would probably know very few admins personally. The apprenticeship system being put forward relies a bit too much on personal relationships which, while not a bad thing, should not be a minimum bar for RfA. For instance, we have an active Malaysian/Singaporean subcommunity on Wikipedia, but only about three of us are admins, and two of these are from Singapore. Any Malaysian editor wanting to apply for adminship would probably know very few admins, if - as most Malaysian editors do - he keeps to Malaysia-related topics. However, because I don't have much time to devote to editing/mentoring and I don't devote all my energies to Malaysian articles, I would - at best - know these people only superficially, so I would have difficulty sponsoring any particular candidates. (And in case anyone is wondering if we need admins who spend most of their time on one topic, rollback is a pretty handy tool - and in really heated arguments, page protection as well.) I'd like to see this apprenticeship system work, but as usual, the main barrier it must surmount is getting people involved. Johnleemk | Talk 18:25, 15 May 2006 (UTC)
 * Wouldn't an apprenticeship system give you the opportinity to quickly get together some more .my admins? You could seek out people who you think would be able to learn. :-) And I know folks who can help so you wouldn't have to do it on your own. Kim Bruning 18:29, 15 May 2006 (UTC)
 * I'd be very much opposed to this as a requirement and I wonder if it weren't a requirement whether it would really fly. To begin with, I'm emphatically opposed to the idea that there's anything wrong with self-noms. If you're confident you're ready, good for you—let your record and behaviour speak for itself. Secondly, I'm not an admin after 15 months and lots of edits and I think I'm acculturated rather well. If I ever nom myself, I'll go over the reading list, re-check the policies top-to-bottom, and then just do it. There are probably more non-admins in the 5000 to 10000 range than people think; the sponsor would be an unneeded formality in many such cases. Marskell 18:43, 15 May 2006 (UTC)
 * It's not a requirement. Kim Bruning 10:22, 16 May 2006 (UTC)


 * In reply to Uninvited's comment: as I run the Admin Coaching program at Esperanza, I've ran into several hurdles that can make a Wikipedia-wide project collapse. One of the problems I've seen is that a few coachees are "overenthusiastic" and actually irritate their coaches because they want to have a perfect RfA, to the point where several of them complained to me as the coordinator. While coaches should be accountable for their coachees, the tendency brought up above effectively causes coach burnout. However, the most pressing issue I've found is that admins are not always willing to help other users. Even in a smaller project like Esperanza, we are running a significant backlog, as I don't want to overload coaches, nor to pile on users to the few active coaches, thereby forcing David Gerard's Law to be true. If a project-wide process for this is done, I would really like to know where all the mentors would come from, or otherwise force applicants to provide their own. Tito xd (?!? - help us) 18:51, 15 May 2006 (UTC)
 * Apparently I'm still on the coaching program (I did join), but I only ever got assigned one person, who consequently never did anything and for all I know has left. What's with that? Kim Bruning 10:22, 16 May 2006 (UTC)
 * The pool of sponsors is any current admin. Of course, this has two immediately apparent problems; one, an additional expectation for admins that they be able to handle sponsoring people effectively. Think; "Oppose - This editor would make a very poor sponsor of potential admins" could be a reality. Second, and considerably more significant, we create a group (admins) where the only people that can get in are people approved of by one or more of those that are already in. If you're going to do that, you might as well just restrict RfA to admins only. --Durin 19:08, 15 May 2006 (UTC)
 * Meh, no, I'd like the sponsor pool to be a bit smaller than that. Some non-admins can be sponsors too.
 * I think adminship via sponsorship might actually work faster than via self-nom. But whatever, if you think self nom works better, it's not going away anytime soon. :-) Kim Bruning 10:22, 16 May 2006 (UTC)

Provisional admins?
(new subsection as it's related, but different)

I think having sponsers/apprentices is a little on the complex side, due to the many reasons outlined above. However, I also think the current Rfa system could do with improvement as at the moment, it is a leap of faith. All we have to go on are the persons history as an editor, but the skills required to be a good editor and good admin are quite different, in my view. Also, a person could be seen as very good at compromising with other editors, but we have no idea if this will carry over to admin space. Will they become a Roosevelt admin, "Speak softly and carry a big mop", or a rogue admin "I don't have to speak softly anymore because I have a big mop"? We just don't know. Maybe we should have Rfa's as now, and once successful, they become an admin, but have "provisional" status for the first X months, at which time there are reconfirmed (simplier process, with a lower threashold, so it's not a second Rfa, as even good admins tend to receive flak.) Any comments? MartinRe 14:33, 14 May 2006 (UTC)
 * You put your finger on the core question here--will an editor-come-admin change once they have the adminship? Or, to put it another way, how can we tell whether someone's hiding their "true" character in order to gain promotion?  The problem is that instituting more stages, whether through apprenticeship or a probationary period, does not really solve that problem.  An apprentice can keep on his best behaviour until the apprenticeship is over; so too can an admin-on-probation.  One way to keep people on their best behaviours is to have a sunset clause  for all admins: their admin status lapses after a year and must be renewed.  But this raises another question.  There are sometimes when, frankly, it's better that an admin is willing to step on peoples toes: to go into an edit war and be quite firm with one or both sides--to be willing to lose a bit of popularity.  The current process has its flaws, but I'm not sure that any other will be able keep bad people from being, well, bad.  Bucketsofg✐ 15:41, 14 May 2006 (UTC)


 * Good idea, although I'd still have the threshold at a decent level for the 'confirmation' - say 65% or so instead of the 80% for an RFA - since any admin who annoys over one third of the people he comes into contact with is doing something wrong. Cynical 16:24, 14 May 2006 (UTC)

Another way to take pressure off of RFA?
I think that as long as becoming an admin is a priority for a lot of editors (which it is...a big deal) any probationary adminship or mentoring system will just cause editors to smile, nod and supply answers they know are acceptable/expected. Moreover, groups like CJ and Esperanza will advocate for their members without being as critical as an outsider might be. I think the only real solution is to raise standards (which won't happen) or put into place some easier way of policing admins. We need to have a way to temp de-admin people that have transgressed on their admin tools. The prospect of a day/week/month as a non-admin will keep people closer to the pack I think. That takes the pressure off of RFA and makes the process not quite so final. Rx StrangeLove 16:52, 14 May 2006 (UTC)
 * What kinds of transgressions would warrant a short de-admining? If the mistake is in good faith, a warning should suffice, if it's not, they need to be de-admined until there is a reason to trust them again.  I don't see the point of short breaks. --Tango 17:19, 14 May 2006 (UTC)
 * Wheel warring for example. As it stands, once you're an admin it's pretty hard (and time consuming) to reverse. So unless you go way overboard or get noticed by the wrong person you're an admin for life. We need real consequences for admins that are acting out. I'm not talking about good faith mistakes, it's the bad faith ones that need attention. What normally happens when an admin gets blocked is that half the time someone else unblocks right away, or the block is a short one that doesn't have any impact. I think it's be much more effective to force the admin to participate in Wikipedia as a normal user for a while to bring home the consequence of his or her actions. Now, if permanent de-admiing becomes easier then this wouldn't be necessary. The point I'm trying to make is that there are nothing really stopping an admin from acting out right now...and as I was saying above probationary adminship or mentoring won't be effective. Rx StrangeLove 15:34, 15 May 2006 (UTC)
 * Surely wheel warring requires permanent de-adminship. I thought this could be done through ArbCom.  I've never really been involved in ArbCom and have only read a few archived discussions - is it too hard to get admins demopped that way? --Tango 16:01, 15 May 2006 (UTC)
 * Well, it depends on the issue I think. A couple months ago we had several de-sysoppings happen nearly instantaneously over the pedo-userbox thing. But generally it takes a full ArbCom action which I think many people agree is unwieldy and too time consuming. It doesn't stop at wheel warring though, there's things like ongoing and inappropriate blocking, ongoing disruption and/or deletion abuse that would warrant a block of this kind. I think we need to be able to warn admins in some manner that falls sort of a long drawn out ArbCom action. The problem with using normal blocks to so this is that even if they do stick, they cut off all editing permissions. It might be more useful to show the target of the admin block what life is like without the admin bit for awhile. Just to repeat myself a little, we should have some way to give admins a timeout short of a full blown ArbCom finding. Rx StrangeLove 17:58, 15 May 2006 (UTC) Rx StrangeLove 17:58, 15 May 2006 (UTC)

Adminship renewal
I've written a proposal, Adminship renewal. Please feel free to read it and comment/make amendments as you deem fit. - Cheers, Mailer Diablo 03:52, 15 May 2006 (UTC)
 * For the one millionth time, no. Putting aside the obvious (and very often stated) problem that the admins who do the most beneficial work are often the ones who aren't terrible popular, it is simply logistically impossible to renew 1000 (or more) admins every 365 days (or about 21 per week in addition to the 20-or-so per week that already go through RFA). Raul654 03:57, 15 May 2006 (UTC)
 * Being done at some other wikis. Not very encouraging results. Wasn't there a perpetual proposals page somewhere? Kim Bruning 06:34, 15 May 2006 (UTC)
 * "Wikimedia projects, including two of the better-known ones (Meta and Commons), have provisions that are already in place for inactive sysops to be removed/sysophood to be subject to confirmation." &mdash; I think this would be a better system than that proposed herein. Only inactive sysops would need to renew their sysop bit, all others exempt unless Arbcom says otherwise. I'd suggest sysops inactive for more than two years be de-adminned automatically, and when they return, they would need to have a "renewal" RfA (with a different success threshold?), in order to prevent their powers from being misused if their accounts fall into vandal control. Kimchi.sg 08:29, 15 May 2006 (UTC)


 * (Sorry if this should have been proposed already, but...) How about a request to remove adminship? With a treshold of let's say ...insert consensus here... for "successful" removal (85%?). And probably with a maximum repeat rate of ...insert consensus here... per year (once?). The community should be able to remove the sysop bit, after all it was the community who granted it. --Ligulem 09:03, 15 May 2006 (UTC)
 * Or we could think positive and do a "Request to confirm adminship" (logically the same, just with inverted support/oppose). --Ligulem 09:12, 15 May 2006 (UTC)

I agree that we should put something in place to deal with inactive admin accounts. We have over a hundred admin accounts that haven't edited in the year 2006. We're leaving ourselves wide open if any of those accounts end up in the hands of someone with malicious intentions. Also, some admin is leaking to a certain site that shall remain unnamed ... this is likely an inactive sleeper admin account just being used for this purpose. -- Cyde Weys 09:15, 15 May 2006 (UTC)


 * An inactive account is no more dangerous than your account or mine. If we have insecure passwords, we can be hacked.  Ral315 (talk) 09:23, 15 May 2006 (UTC)


 * I doubt it is a sleeper. Apart from anything else there are certain gaps in their knowlage that suggest they are not long term.Geni 12:08, 15 May 2006 (UTC)

This is why we don't do adminship renewal:


 * 1) To renew 800+ admins yearly wouldn't work logistically.
 * 2) Loss of adminship based on inactivity doesn't really make sense; we have lots of good admins (Zoe, for one), who take extended breaks.
 * 3) Most importantly, it causes admins to worry for their job. This is the same reason that many judge positions in different countries are life terms- otherwise, politics plays big into decisions. "If you block me, I'll oppose your renewal, and so will my friends."

Ral315 (talk) 09:23, 15 May 2006 (UTC)

Just renew the ones that don't respect policy. Get a petition with X number of signatures that say a certain admin doesn't respect policy, and then a renewal vote would be held. No need to have routine revotes on the vast majority of them who cause no trouble. Everyking 09:43, 15 May 2006 (UTC)
 * Yes, I support that idea. A request to renew with a minimum of X petitioners. --Ligulem 09:51, 15 May 2006 (UTC)


 * At the risk of stating the obvious, this would allow people like the GNAA to remove admins that were causing them problems.Kim Bruning 10:02, 15 May 2006 (UTC)
 * Of course, we'd have to expect people to have some record as constructive editors to be able to sign one of these petitions (or vote). Everyking 10:53, 15 May 2006 (UTC)


 * In reply to Kim: No. A normal RfA voting would be held after such a successful petition and the GNAA is not the community. The rate for such renewals should be limited, tough (minimum 3..6 months after the last confirmation). We also have TfD, Afd, whatever. This would help to keep admins in touch with the goals of wikipedia (and by also respecting the community needs for respectful communication). If the community is intelligent enough to elect admins, then it can also confirm them when needed. --Ligulem 10:59, 15 May 2006 (UTC)
 * Nope, this was tested, well totally poisoned by a misbehaving user who tried something quite like that ;-) Kim Bruning 12:22, 15 May 2006 (UTC)

I agree completely with Ral315 and Kim Bruning, and would not like a system like this to be implimented. --Lord Deskana Dark Lord of the Sith 11:06, 15 May 2006 (UTC)

The only reason I can see for desysopping someone is abuse of the tools. If there is extensive abuse, the offending admin will sooner or later be taken to arbitration, and can be desysopped by the ArbCom. A system that would make admins afraid to block disruptive users who might oppose their renewal RfA (and get their friends to do the same) would be a bad idea in my opinion. AnnH ♫ 11:57, 15 May 2006 (UTC)
 * I agree with your first sentence, but after that we differ. Realistically, the only abuse of the tools that's going to get you to arbitration is either of the rampage variety or the ticking-off-the-wrong-people variety. But there's admin abuse beyond this: blocking people for poor (but not entirely insane&mdash;that would be "rampage") reasons, using the rollback button against edits aside from vandalism, and things such as closing a vote improperly (by which I mean in a way that is unpopular and generally contrary to standard policy interpretation, but is not so outrageous that it is impossible to defend). This is the sort of thing we need a way to deal with: the admin who uses the admin tools improperly, in a way that makes him/her unpopular and undermines policy, but isn't extreme enough to get him/her to arbitration under current practice. Also, I think that getting admins to be more cautious&mdash;to worry about a revote&mdash;would be a great thing. Problem #1 with regard to admin issues is the way some admins will act individually and aggressively, without community deliberation or agreement, essentially with impunity. Anything that would get them to worry about their popularity a little more, and have some feeling of being responsible to the community, would be a tremendous improvement. Everyking 12:41, 15 May 2006 (UTC)
 * Anything that would induce an administrator to make the 'popular' decision is not a good thing. This would increase the likelihood of admins pandering to a) trolls, b) POV warriors or c) asshats, as they know that all of those three groups remember every admin who ever blocked or ticked them off, and all of whom would waste no time in turning any kind of revote into a fiesta of bitching and ill-will.  Administrators don't have to make the popular decision, they have to make the correct decision.  Wikipedia is not a democracy, it's an encyclopaedia, and the correct decision does not have to necessarily tie in with the most popular one.  There are many examples of this.   Proto  ||   type   13:04, 15 May 2006 (UTC)
 * I'd like it if they pandered to the community a little more, really. I guess we have a philosophical disagreement. You want every admin to be a law unto himself, making "correct" decisions on anything according to his own say-so? Is this really a proper way to determine what's "correct"? Somebody has to decide the difficult issues, but it should be the community, not the individual admin. Everyking 13:25, 15 May 2006 (UTC)
 * So I guess we can toss out WP:NPOV, WP:V, etc. in favour of consensus, then? Consensus makes the right call 99% of the time, but the 1% of the time is what we tend to focus on, so we tend to allow wedge cases - the most unlikely eventualities - to define how we ought to behave for the other 99%. Admins are not a law unto themselves - rather, they are given the mop and bucket to utilise in however they best feel necessary within reason. If they make editors slip on wet slicks, or flood the hallways, they ought to have the mop and bucket taken away. This dangerous either/or philosophy - that admins should either be totally unaccountable to the community and accountable only to the arbcom, or totally accountable to the community alone - is not one that should ever be allowed to take root. We have mechanisms for ensuring that admins utilise their mop and bucket responsibly. Anyone can start a discussion at AN or ANI about a particular admin, and through this, the community is able to communicate its (dis)approval of an admin. Likewise, the arbcom is the only way to desysop admins because people have confidence in the arbcom to do the right thing - both sides will typically respect a decision (unless one of them is a troll or just plain disruptive), whatever it may be.
 * I'm sure the one admin always hanging on everyone's mind where this is concerned is . I don't mince my words, and I have no compunction about naming his name. I'm sure everyone is silently thinking, "Yeah, the arbcom didn't desysop Tony despite his history of wheel warring especially over userboxes". The problem again is that we have allowed Tony to be defined by the wedge cases - the exceptions rather than the rule. Tony does a lot of work with his mop and bucket outside the realm of userboxes, and remarkably wheel wars a lot less than he used to. Likewise, is famed for simply deleting any userbox he felt like deleting, but this obscures a lot of work he does maintaining TfD and performing other menial tasks. Since being banned from involvement in userboxes, he has continued to do good work.
 * The lesson is, if we the community had our way every time, we would act like the primal beings that we are in a large number of cases, and act punitively rather than preventively. Many rogue admins can contribute to Wikipedia without being desysoped, but a riled up lynch mob is the last creature on the face of the earth to look for compromise. This is why Jimbo delegates the trust we, the community, place in him to arbitrate disputes to the arbitration committee, and not back to the community ourselves. It hurts to be considered incapable of even-handedly resolving a polarising dispute, but this is exactly the last thing an angry - or even worse, angry and divided - community can ever be capable of doing. Individually, we may act rationally, but on a group scale, we act irrationally. As such, it is irresponsible and simply crazy to give the community full discretion in banning admins - and likewise, it would be irresponsible to only allow the arbcom input on whether admins are handling the mop and bucket well. We must have compromise, and the only way to do this is to elect impartial and capable arbitrators - not invent new forms of instruction creep. Johnleemk | Talk 14:20, 15 May 2006 (UTC)
 * Wow. Very well said. You should save that diff for when you run for ArbComm. Thanks for saying this, it needed saying.  + + Lar: t/c 16:32, 15 May 2006 (UTC)


 * With regards to Everyking's comments, I must politely disagree with him. I have yet to see any evidence that this "problem" with rogue admins really exists. Yes admins handle their affairs with some degree of individuality--within the larger guidelines and rules of Wikipedia--but that is a good thing and something that makes Wikipedia strong. What would scare me is if there were an "admin police" out there ready to smack down any admin who screwed up at some point (and we all do) or dared to use individual judgement when an issue arises which is in one of Wikipedia's grey areas (and it sometimes seems as if a lot of Wikipedia is grey). To quote Everyking, "I think that getting admins to be more cautious&mdash;to worry about a revote&mdash;would be a great thing." No, that would be a horrible thing. I am absolutely opposed to this proposal.--Alabamaboy 13:20, 15 May 2006 (UTC)

Renewing admins is a solution looking for a problem. Oleg Alexandrov (talk) 14:11, 15 May 2006 (UTC)


 * In case someone wants to take a look at the dustbin for past failed proposals of this type (the "perpetual proposals page" Kim mentioned), it's at WP:RFDA. For me, the above arguments quite convince me that to date there isn't any rogue admin problem the RfC+RfAr system cannot handle. (edit conflicted) Kimchi.sg 14:20, 15 May 2006 (UTC)


 * Oooh... I agree with Oleg. --Lord Deskana Dark Lord of the Sith 15:07, 15 May 2006 (UTC)

Any process that arbitrarily enforces admins to go through a stressful cross-examination by the community would be detrimental to the community. In particular it would be very harsh on our best vandal fighters, RickK would have had far from a smooth ride if he had to go through such processes during his heyday. Secondly, Admins like myself, and there are quite a few of us, would also probably be stripped of the buttons because we no longer have the free time we once had. No one has ever accused me of misusing the sysop tools, but I know many would oppose me because I do not have the time to RC patrol or close *fDs anymore (I am a final year undergraduate). It would be of no obvious benefit to put admins through a second request simply because we can, it makes no sense; if there is a problem ArbCom can, and will, sort it out. Rje 15:24, 15 May 2006 (UTC)
 * Additionally, any admin who has been around for a sufficient period of time and made any kind of decision will usually have rubbed at least a few users the wrong way. I know that I have had 194 vandals attack my user page just from doing something as ordinary as reverting vandalism, and any sort of tough administrative decision would add a few users to a potential future lynch mob. I don't see a need to confirm adminships, except for cases of extreme inactivity (someone who hasn't made a single edit for more than two years or a similar benchmark). Tito xd (?!? - help us) 19:04, 15 May 2006 (UTC)
 * As I made clear before, this proposal would not accept the votes of new users, and certainly not vandals. If you have rubbed constructive editors the wrong way through the use of admin tools, then those editors ought to have a way to oppose your continued adminship. Whether or not that opposition was successful would depend, perhaps, on the number of constructive editors you'd rubbed the wrong way.


 * Wikipedia is either going to be run by its community, or it's going to be run by a few admins who are particularly assertive and individualist in their decision-making. Admin confirmation after petition would swing a degree of power back to the community. Every admin needs to understand very clearly that they are responsible to the community and they must obey policy. Failure in those respects absolutely should carry with it some risk to one's adminship. Everyking 04:42, 16 May 2006 (UTC)
 * "If you have rubbed constructive editors the wrong way through the use of admin tools, then those editors ought to have a way to oppose your continued adminship." I'll give you credit - that was a very slick attempt to dictate the terms of the discussion. It's not hard to see why you put that particular qualification in there. This above all, to thine own self be true, James. Raul654 04:52, 16 May 2006 (UTC)
 * I eagerly await Everyking's reply, as to why he felt it necessary to include the proviso that only admin actions may be considered. Raul654 22:43, 17 May 2006 (UTC)
 * What are you trying to imply, Mark? When considering whether someone should remain an admin we should look at her admin actions. This is axiomatic and it is the principle which the ArbCom and you yourself have applied in the past. There's no reason to paint this as some sort of self-serving ploy on James' part. Haukur 14:05, 19 May 2006 (UTC)
 * I am implying that Everyking realizes he has plenty of baggage with regard to his non-admin actions and that he has devised this requirement as an attempt to head-off the inevitable opposition his own adminship renewal would generate as a result. Thus, for him to claim we need to crack down on admins who "rub others the wrong way" - but only for their admin actions - is hypocritical in the extreme.
 * As far as my own opinion on the subject, I oppose this whole proposal because it's an extremely bad idea for a the same reasons I have said the last 100 times it was proposed. But that's no reason I cannot point out the motives of people arguing in favor of it. Raul654 05:15, 20 May 2006 (UTC)
 * James has never abused his access to the administrator tools. You know this and it has been reflected in past ArbCom rulings which you have been a part of. When James calls for desysopping of people who have abused their privileges there is no hypocrisy on his part.
 * You were attacking his integrity&mdash;something likely to provoke him into a conflict with you&mdash;at a time when he is under draconian restrictions not to criticize administrators. And when he wisely declines the bait you try to taunt him into replying to you.
 * There is normally no reason to "point out the motives" of fellow editors. It is better to discuss the proposal on its merits, even if it is covering old ground. Haukur 16:28, 20 May 2006 (UTC)


 * "When James calls for desysopping of people who have abused their privileges there is no hypocrisy on his part." - bzzzt, wrong, but a nice attempt to suger-coat what he actually said. EK's actual words (as I quoted verbatim above - a quote your reply apparently flatly ignored) is that we should desysop admins who "rub others the wrong way" - but only based on their admin actions. There is nothing in any arbcom decision pertaining to adminship renewal, despite your attempts to bootstrap our comments vis-a-vi desysopping into the discussion. So it's clear this caveat was cooked up by Everyking. Why would he do such a thing? Could his actions possibly be designed to prevent him from having to respond to all those irritating questions about his numerous past non-administratorial misdeeds. At which point, it beggars the imagination to hear him complaining about admins who "rub people the wrong way" when he is no doubt all too aware that he has rubbed an enormous number of others the wrong way. Those "draconian restrictions" (which, ironically enough, don't actually restrict any of his actions besides the unacceptable ones) did not materialize out of thin air for no good reason.
 * "You were attacking his integrity" - again, wrong (nice attempt to poison the well though). I'd be "attacking his integrity" if I called him a liar, which anyone who actually read my post can see I did not. I am, however, pointing out his transparent attempt to insert a caveat designed to benefit himself. And lastly, if EK doesn't want to be "baited" by people pointing out the transparent patent hypocrisy of his comments, he should not make hippocratic comments!! Raul654 01:20, 21 May 2006 (UTC)
 * I was wondering if we could just drop this discussion? Adminship renewal ain't happening, and this particular debate on the topic just causes more ill-feelings and upset editors. Let's just drop this. -- Mark Neelstin (Dark Mark)  01:43, 21 May 2006 (UTC)


 * My objections to the proposal have nothing to do with new editors or vandals messing up the vote. They are based on the fact that rational people acting as a group often end up unintentionally morphing into an irrational lynch mob. Many established editors called for (and occasionally, some still call for) Kelly Martin, Tony Sidaway and MarkSweep to be desysoped for their actions on userboxes, but this is a peripheral issue which has nothing to do with the encyclopaedia. People often allow wedge cases to define their thinking - these admins have contributed and continue to contribute positively to the encyclopaedia without being stripped of the mop and bucket. Community input is valuable, but it should not become the be-all and end-all of adminship. The mop and bucket should not become political tools. (And yes, contrary to popular belief, rational people do become political given the right circumstances - and one such situation would be the formation of a lynch mob. Group mentality based on the input of rational people often outputs irrational decisions.) Johnleemk | Talk 06:07, 16 May 2006 (UTC)

A minor anomaly?
I guess that no bureaucrat had noticed this but the recent RfA for Whouk has a slight mistake. Bhadani voted TWICE for this RfA (See vote number 36 and 62)! Anyway it did not make much of a difference. However, this further clarifies the point in which we need more bureaucrats in Wikipedia. -- S iva1979 Talk to me  19:06, 15 May 2006 (UTC)
 * Well, that's a good catch, but I don't think more b'crats is the only way to deal with problems like that. --W.marsh 19:10, 15 May 2006 (UTC)
 * Siva1979, I don't see how having more bureaucrats would have changed anything about the RfA you note. Could you clarify what you mean please? --Durin 19:13, 15 May 2006 (UTC)
 * What we need is a robot bureaucrat. It parses the votes, eliminate duplicate votes, which a person won't see. Afterwards,

identify (sockpuppets); if (percentage > 75) promote; else not promote; end
 * Oleg Alexandrov (talk) 20:00, 15 May 2006 (UTC)
 * Robocrat... lol. Kimchi.sg 10:17, 16 May 2006 (UTC)
 * That was my mistake in missing it. I didn't check every vote because it wasn't remotely close. More bcrats seems a little besides the point though because we aren't necessarily going to catch something like that even if we cross checked each other's closings. For sock concerns and double votes we have to lean on the community to at least bring concerns to our attention in a civil way. Politely comment below a vote what the concern is and then we can deal with it. - Taxman Talk 20:12, 15 May 2006 (UTC)


 * Well, Oleg Alexandrov stated it quite clearly for me. What we need is a robot bureaucrat. However, we must not neglect the human involvement in RfAs as well. I feel that the responsibilities of a bureaucrat as grown given the number of users in Wikipedia. As these issues require attention, more bureaucrats would be able to deal with bureaucratic issues such as RfAs and changing of user names more efficiently. -- S iva1979 Talk to me  02:56, 16 May 2006 (UTC)


 * I saw this post & started to go through my recent RFA & found that User:Anonymous_anonymous has supported me twice (no. 49 & 55). Just thought I'd bring it to your notice. Thanks. Srik e it ( talk ¦  ✉  )  03:39, 16 May 2006 (UTC)
 * This seems to be symptomatic of a more general problem- people don't pay enough attention/ think enough about the RfAs they vote in. JoshuaZ 03:55, 16 May 2006 (UTC)
 * I don't think more bureaucrats are required for cross checking & other such things. I suggest that while nearing the end of an RFA, an admin, an experienced user or even a bot should check the RFA for duplicate votes, suspected sockpuppets, tallying problems etc & leave the closing 'crat a note in the comments section. Although I don't think giving a bot bureaucrat powers is the way to go. Bots, if created, should assist bureaucrats not do their job for them. --Srik e it ( talk ¦  ✉  )  05:33, 16 May 2006 (UTC)


 * The best solution to these kinds of things is for users who are interested in the particular RfA (regardless of which "side" they are on) to keep the RfA on thier watchlist. Bureaucrats have a dozen RfAs to watch; most users have, at most, 1 or 2 that they really care about at a given time. If users watch for things like double votes, vote tampering, and sockpuppets, and bring these things to the attention of the bureaucrats, it greatly decreases the chance that a mistake will be made. Given that many if not most RfAs are having more than 100 voters, and some reaching up into the 200 edit range by the time commenting, question answering, and the like is finished, asking bureaucrats to review each and every edit with a fine toothed comb is expecting just a bit too much. We're a community here; just because you don't wear the bureaucrat title doesn't mean you can't and shouldn't help where you can. Arbitrators have clerks; bureaucrats have the voting base. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 07:16, 16 May 2006 (UTC)
 * Actually Essjay's suggestion has given me an idea. Why not start an RFA watchlist commitee? The commitee can assign members to watchlist each RFA. The 'watchlisters' (for lack of a better word) can update tallies, check for duplicate votes, sockpuppet votes & generally keep an eye on the proceedings. In case of delayed RFA closings they can strike out late votes & after an RFA period is over leave a general summary of the voting proceedings for the closing 'crat in the comments column. Any user having a good knowledge about RFA's & voting procedures will be accepted as member. What do you think? Srik e it ( talk ¦  ✉  )  08:41, 16 May 2006 (UTC)
 * But won't this be extra bureaucracy for a trivial task? The voters who watchlist each RfA are quite good at pointing out discrepancies or fraud votes already. Kimchi.sg 10:17, 16 May 2006 (UTC)
 * Yet we have cases like those mentioned above where duplicate votes slip through. In my RFA as mentioned above the duplicate vote was tallied & missed by the bureaucrat even though it was quite close to the orignal vote (no. 49 & 55). If we check all archived RFA's we might probably find many such votes. I doubt that such a commitee will increase bureaucracy. All the committee will do is organized RFA watchlisting. And all the 'watchlisters' (still searching for a better word) will do is keep an eye on their assigned RFA's. The only proper task is to add a voting summary to the comments section like "No duplicate votes found. No sockpuppet votes apparent." at the end of the RFA. The committee will just make life slightly easier for the 'crats. Srik e it ( talk ¦  ✉  )  10:49, 16 May 2006 (UTC)
 * No instruction creep, please. There's no evidence that an official committee will substantially improve the situation, which - at any rate - does not appear critical. If the double votes are enough to tip the balance, people would be combing the RfA (and its subsequent closure) for any discrepancies whatsoever. On the other hand, if it's clear that the candidate will pass with or without the double vote, whether there was a double vote or not is immaterial. Adding an extra layer of bureaucracy is patently unnecessary, especially since it assumes there will be people eagerly volunteering to help out. On Wikipedia, most things get done ad hoc due to its decentralised nature, which is largely a good thing. The people most interested in an RfA will already be watching it, with or without the committee, and there's no evidence that forming a committee will get more than these interested people to participate. Johnleemk | Talk 11:41, 16 May 2006 (UTC)
 * There's no need for an official committee if it just becomes part of the RfA culture to check for and point out double votes, sockpuppet concerns etc. People of course need to have solid backing before pointing out sockpuppet issues because if there's no basis for the concern it's quite rude to claim sockpuppets. But if it becomes part of the culture, newer RfA participants will see the polite way to point out issues in an RfA and we then have a much better checking procedure than all the bcrats put together could do. - Taxman Talk 11:19, 16 May 2006 (UTC)
 * There's no need for an official commitee, but an unofficial one might be good. Similar to how RC Patrol works - there's no burocracy around it, but there are particular people that regularly do it. --Tango 12:34, 16 May 2006 (UTC)


 * Then have at it! :) I take it upon myself to help ensure RfAs are stamped with the proper end time, are closed properly, and other sundry tasks around RfA related pages. Didn't take a committee :) So, go for it. You don't need approval :) --Durin 14:20, 16 May 2006 (UTC)
 * I think we can handle the closing unless you're talking about failed ones removed under WP:SNOW. :) But the rest would be great and would make closings nearly flawless. Lets see, the tasks would be sock checking, double vote checking, proper end times, making sure the numbering isn't messed up due to comments. Anything else? Maybe we could put that up at WP:RfA/RfA Patrol because this talk page discussion will be lost in the archive eventually. - Taxman Talk 16:43, 16 May 2006 (UTC)
 * I don't think we need anything formal. --Durin 16:47, 16 May 2006 (UTC)
 * Now that I think about a bit, I feel the guys above are right. A committee is unnecessary & will be instruction creep. Like Durin says above, anyone interested can take up the job themselves & they don't need any approval. Well so much for my brilliant brain fart, guess I'll go back to patrolling RFA's myself. Srik e it (  talk ¦  ✉  )  16:28, 16 May 2006 (UTC)
 * There is no reason to expect a bureaucrat to spot minor errors in an RfA that is not close. Double votes happen more often than you think. Mine had two cases of double voting. Anyone who spots it should fix it. the more 100+ vote RfAs we have, the more common double voting that goes unnoticed will become. NoSeptember   talk  17:49, 16 May 2006 (UTC)
 * Indeed. Even though such mishaps are more common than one might think, the overwhelming majority of them gets spotted and pointed out before the RfA is closed, so it is rather rare that one of those will get by &mdash; and if they do, it's usually because the RfA went very smoothly, so the chance that any real harm might be done is quite slim.  Redux 21:57, 17 May 2006 (UTC)
 * I think the candidate would be most likely to spot it, I spotted it when MONGO voted twice on my RfA. Prodego  <sup style="color:darkgreen;">talk  23:56, 18 May 2006 (UTC)

Making points on RfA pages
Well, its come up again, Ardenn's voting oppose everwhere because "nothing against the candidate" we just don't need more admins. Ardenn is clearly trying to make a point, and I find it a little disruptive. Do we want to adopt a policy for crats marking disruptive votes as such so hopefully the pratice can be discouraged -- Tawker 05:34, 17 May 2006 (UTC)


 * It's simple: Trust us. We're smart enough to know when votes are in good faith and when they're not. I can't think of a single person who was denied adminship because of a vote by Boothy443, and I highly doubt this instance will be any more sucessful. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 05:51, 17 May 2006 (UTC)


 * I agree with Essjay :). It really fills the page with needless rambling, AT BEST. IMHO people should stick to disputing reasons instead of the intent of the user, because they only direction that goes is down :\. It is as it always was T 05:54, 17 May 2006 (UTC)


 * I think its not about final outcome and is more about a user making a point by adding his views (which no-one agrees to) everywhere. I strongly feel that its disruptive. -Ambuj Saxena (talk) 05:57, 17 May 2006 (UTC)


 * Then just ignore it. As long as it doesn't have any tangible effect, there's no point paying attention to it. Johnleemk | Talk 07:13, 17 May 2006 (UTC)


 * He should be able to vote for any reason he wants, provided he has a sufficient history as a constructive contributor. It's worrying when people try to dictate what reasons are valid for opposing. A person's vote is an expression of their opinion and should not be disregarded. I find it rather irritating that Essjay would say that holding such an opinion marks him as voting in "bad faith". Everyking 09:09, 17 May 2006 (UTC)


 * If an opinion runs counter to policy it ought to be disregarded. Not ignored, but nevertheless, considered invalid. Otherwise we might as well discard our policies and just let mob rule take over. (Note: There's a difference between IAR and running counter to policy. Many actions taken under IAR/WP:SNOW in the correct manner as outlined on their respective pages do not run counter to policy, but fill the gap of wedge cases as I mentioned above in the discussion of renewing adminship/community deadminning.) Just because you can express your opinion doesn't mean it should be heard. It is policy currently to continue adminning people, or else we would shut down RfA. If Ardenn wants to make a real change, he should propose to change this policy, not try to make a point by abusing the existing policy/process. Johnleemk | Talk 09:19, 17 May 2006 (UTC)
 * I can't think of what policy could prohibit his votes, unless it's the argument that the mere existence of RfA prohibits his votes, in which case I think that's preposterous. And quit talking about "mob rule" (you were using that same line in the previous argument)&mdash;I happen to be a big believer in "mob rule", although I don't call it that or perceive it as being at all contrary to policy. Everyking 09:25, 17 May 2006 (UTC)

RFA IS NOT A VOTE The whole point is reaching consensus, which is not the same as "80% support", that's just a convinient rule of thumb - that means that all votes are not equally valid. The weight of the vote depends on the validity of the reason. Everyone is entitled to their opinion, certainly, but that's irrelevant - it's the opinion of Wikipedia as a whole that matters. Prehaps we should have a single debate and reach concensus on what kind of votes are valid in one go, rather than repeating it every RfA. "Candidates need more than 2000 edits" is no more valid or invalid on one RfA than another, so why don't we determine it for all RfAs at once? --Tango 12:29, 17 May 2006 (UTC)
 * Seems to me you're using the idea of consensus to dismiss a person's opinion. The fact is, RfA is a vote&mdash;it's unfair to treat it as a vote, to have it function as a vote, and then try to deprive people of their votes when they use them to express something unpopular, on the grounds that it's supposed to be about "consensus". Everyking 13:56, 17 May 2006 (UTC)
 * Surely the closing 'Crat would discount Ardenn's vote. How useful in seeking consensus is his vote if he always votes no? Is he violating WP:POINT? Is there not a better way for him to address the issue? :) Dlohcierekim 14:26, 17 May 2006 (UTC)
 * Were I the 'crat (I'm not), User:Ardenn and User:A ding ding ding ding ding ding ding would cancel each other out... ;) Radio  Kirk   talk to me  15:29, 17 May 2006 (UTC)
 * If RfA isn't a vote, why does the template say "vote here"? Hehe. It's not a pure vote, because sometimes people with 75-80% can pass, but 80.0% and up now essentially always succeeds, and b'crats have said this, so it is basically a vote. --W.marsh 15:41, 17 May 2006 (UTC)
 * why does the template say "vote here"? That's a weak point, because some nominators do change the "vote here" to "discuss here". Kimchi.sg 15:47, 17 May 2006 (UTC)
 * My point is that we can call it consensus and discussion, but ultimately a 82% RfA will practically never fail. If you succeed by getting enough votes, then it's a vote... even if we say it's determing consensus, which it is too. --W.marsh 15:55, 17 May 2006 (UTC)
 * 82% will never fail because b'crats have to essentially promise to always abide by the 80% rule and never think for themselves if they want to get through RfB. We should give the b'crats more room to use their descretion - if we didn't trust them to do that, they wouldn't have the job.  How about expanding the buro descretion range from 75-80% to 65-90%?  Promotions under 75% and failures over 80% would still be very rare, and would probably require some explanation from the buro, but they would have room to do it if it's needed. --Tango 16:19, 17 May 2006 (UTC)

Frankly, this seems like a reasonable reason to oppose, although one I would disagree with. Presumably if the position has only limited support then it will have little effect on RfA outcomes. Ardenn's votes may very well be in bad faith (I know nothing about the user) but on the surface I don't see the problem, particularly. Christopher Parham (talk) 21:24, 17 May 2006 (UTC)
 * They're not really in bad faith (I think--see his talk page) but his summaries do leave something to be desired. I don't think this is really a WP:POINT to be made. --M1ss1ontomars2k4 | T | C | @ 04:11, 21 May 2006 (UTC)

RfA Analysis
I picked up a few hints in the "A minor anomaly?" section above, so I decided to create a tool to deal with the duplicate voter problem. Although not a "Robocrat" by any means, it can find duplicate voters in any RfA you specify. It should support most users' signatures, but if you find any user listed as [Signature not found], please drop me a line. The tool's available at http://tools.wikimedia.de/~tangotango/rfa.php. (Code will be available shortly, when it's been cleaned up). Cheers, Tangot a ngo 14:43, 17 May 2006 (UTC)
 * At a glance, that looks awesome. Detects the duplicate votes in the RfAs we know about, at least. I think from the gist of the previous "minor anomaly" thread, it would be best if a member of the community tried to run this towards the close of every rfa, and mention the results in the comments for the b'crat to read (or maybe something mathbot-like could do it automatically, just don't call it Robocrat). Of course b'crats might want to think about using it themselves. --W.marsh 15:47, 17 May 2006 (UTC)
 * Hmm... going over NoSeptember's RfA with it doesn't find any double votes. Kimchi.sg 15:55, 17 May 2006 (UTC)
 * I think that's because in that RfA, the duplicate votes were basically removed and no longer show up as #'d items in the list? So the script sees them as comments, not votes. --W.marsh 16:01, 17 May 2006 (UTC)
 * Yes, that's right. The two duplicate votes don't register as votes under the tool. (It only looks for valid votes that count towards the tally). — Tangot a ngo 16:05, 17 May 2006 (UTC)
 * W.marsh's idea about using a bot is a good one. Maybe the robot which updates RFA summary and which parses the votes anyway could also flag the duplicate votes. Oleg Alexandrov (talk) 17:20, 17 May 2006 (UTC)
 * I caught both of the duplicate votes in my RfA, and either told the voter or struck the vote myself. I note that Merovingian's vote (#14) did not get picked up by name in my RfA though. NoSeptember   talk  17:56, 17 May 2006 (UTC)
 * Btw, this is a very nice output, useful even without further improvement. :-) NoSeptember   talk  18:04, 17 May 2006 (UTC)


 * Seems good, but misses struckout votes. This happens frequently as people change their votes. --Durin 17:24, 17 May 2006 (UTC)

Can the vote tally be alphabetized? I was thinking it might be neat if we could summarize each regular voter's voting habits. (that is determining that User X voted support 19 times, oppose 4 times, and neutral 2 times in the month of April) I can see it now: voteritis, a focus on the past habits of voters :-P. NoSeptember  talk  17:56, 17 May 2006 (UTC)
 * That'd make it easy to pick out the Boothys, Ardenns and Massiveegos - or the other way round. And make it easier for insomniacs like me to sleep. :-P Kimchi.sg 18:32, 17 May 2006 (UTC)

Another error: the tool chokes on CSCWEM 3, saying it has "4 sections". Kimchi.sg 19:03, 17 May 2006 (UTC)

Also in JoshuaZ's RFA it shows that PseudoSudo has voted both Support & Neutral but what happened is when Connel Mackenzie changed his vote from Oppose to Neutral, PseudoSudo copied the vote to the Neutral section & added his sig to the end saying "copied from above". The tool does not register Connel Mackenzie's vote. Srik e it ( talk ¦  ✉  )  00:31, 18 May 2006 (UTC)

Thanks for all the feedback, everyone! I've fixed the errors that have been reported so far. There are some other known issues.
 * 1) The error Kimchi.sg reported with CSCWEM's 3rd RfA was due to one of the sections being bolded by a semi-colon instead of the usual three apostrophes. The regex has been updated to handle this.
 * 2) The anomaly Srikeit reported with JoshuaZ's RfA has been fixed with various changes to the signature-finding code. It will now detect multiple timestamps and look for the one that was most likely added by the first user. (It will also ignore timestamps used inline for non-signature purposes)
 * 3) Another problem Srikeit reported (on my talk page) with Xoloz' RfA was due to extra spacing after one of the headers. The code has been updated to handle this.

Durin, I'm interested to find out which RfAs suffer the problem you report. If the new version of the tool still suffers from the same problem, please tell me the details. Also, if anyone finds any other problems or oddities, please feel free to contact me or write here (with a link to the RfA where the error occurs). Thanks. Tangot a ngo 08:23, 18 May 2006 (UTC)


 * Thanks for rectifying the above mentioned errors. I was just checking Requests for adminship/Master of Puppets with the tool where quite coincidentally it says that I cast a duplicate vote. It was really Mysekurity's vote who gave the following comment;


 * Metallica SupporT :). " Oh, and in response to Srikeit, I believe it might have something to do with this. - Mysekurity <font style="background:black" color="white"> [m!] 21:30, 26 April 2006 (UTC)".


 * Here I think my user page link has triggered the tool to recognize it as a duplicate vote. Srik e it ( talk ¦  ✉  )  09:05, 18 May 2006 (UTC)


 * Thanks for your report, Srikeit. I'm still thinking of a way to fix that particular problem - it happens due to you being mentioned in a normal <tt> Srikeit </tt> link, and the voter using a <tt> User:My/sig </tt> style signature, which has less precedence than the User: links. (This is because someone could easily link to a subpage of their (or someone else's) userpage, and sign their name later). I'll have to change the regex/way signatures are found in some way. I'll try and deal with it when I have more time. Meanwhile, I've published the code, which is available at, so if anyone can give me a hint as to how to fix it, I'd be very grateful. Cheers, Tangot a ngo 16:51, 18 May 2006 (UTC)


 * Okay, I've fixed the issue by treating <tt> User:XX/sig </tt> links separately, by prioritizing signature-like links that come at the end, and by lowering the priority of <tt> User:XX/pagename </tt>-style links. I'm a bit concerned that some users' signatures might not be caught by this, but I think many of the false detections have been weeded out. If anyone finds any errors, please report them to me. Thanks, Tangot a ngo 17:50, 19 May 2006 (UTC)

I am worried by vote counting. Kim Bruning 10:26, 18 May 2006 (UTC)
 * I am not so eager to make "voteritis" (to use NoSeptember's words) easy, either. My RfA analysis tool is intended to help Bureaucrats with finding duplicate votes, and while nothing stops someone from quite easily modifying the code to make a tool that could cause another Wikipedia-related disease, I don't think I will implement this myself. -- Tangot a ngo 16:51, 18 May 2006 (UTC)

Removing RfA comments
I've noticed a general trend of simply reverting RfA comments that users feel the burecrats will simply disregard anyway. (Note: I'm not pointing this out to attack any editor or group of editors, just pointing out a general phenomenon.) I think we should have some kind of discussion about what should be done about such votes. While I don't mind removing votes that are obvious vandalism or trolling, my opinion is that less clear-cut cases should either be struck out, a note attached, or simply having faith in the closing burecrat. (With more emphasis on the latter two, as having your vote struck out can be demoralizing.) There are a few points that I feel would be relevant to this argument: Any thoughts? E WS23 | (Leave me a message!) 17:22, 17 May 2006 (UTC)
 * Our user account benefits page lists "the right to be heard in votes and elections" as a benefit to having a user account. While some people may abuse this system, simply reverting the edits prevents the editor from being heard at all.
 * On a similar note, common practice of AfD's that are riddled with sockpuppets has been to let the comments stand as-is, rather than reverting them or striking them out.
 * Starting the practice of reverting rather than commenting on potentially bad-faith votes might be a slippery slope. For example, one might feel justified in reverting the votes of users who feel the need to oppose nearly every single RfA candidate.
 * I tend to agree. Quite a number of people have struck votes, removed votes, commented on them being fractious, etc. I trust the bureaucrats to sift through the RfAs they are closing and weed out what is unhelpful. --Durin 17:28, 17 May 2006 (UTC)
 * Agreed, but as discussed above, it is helpful for users to comment on situations, as we can't see and know everything. Reverting votes or striking them out is not good though. Duplicate votes and sockpuppets of banned users can have an extra # added so they don't contribute to the count. - Taxman Talk 18:00, 17 May 2006 (UTC)

Frankly, comments or votes should never be struck out, even by bureaucrats. If a comment is in such bad faith that it should not be read (the implication of striking out) then it should just be removed entirely. Comments that make a good faith contribution to the process (e.g. votes from anon users) should be commented as such, or moved to a more appropriate section of the page, but should not be struck out. Christopher Parham (talk) 21:31, 17 May 2006 (UTC)
 * I agree with Christopher Parham's first sentence but not with what follows. I think the only comments that should be removed are those that are personal attacks.  Judging good faith or bad faith is a slippery slope that we should not venture out on. -- DS1953 <sup style="color:green;">talk  21:57, 17 May 2006 (UTC)
 * There are cases where bad faith is dead obvious, e.g. simple vandalism. Christopher Parham (talk) 22:17, 17 May 2006 (UTC)
 * Agree with Christopher Parham. Comments that are clearly bad-faith (i.e. "candidate so-and-so is an !!!", etc.) or are from banned users can be reverted. Otherwise, good faith comments (or comments that could be interpreted as good faith) should never be reverted or deleted. Thanks! Flcelloguy (A note? ) 22:20, 17 May 2006 (UTC)

Don't alter other peoples' statements. If you catch someone, I think a block would be appropriate. Kim Bruning 22:21, 17 May 2006 (UTC)
 * I see no point in removing a comment from an RfA. Such will either stand or fall on their own merits. :) Dlohcierekim 14:16, 18 May 2006 (UTC)
 * So are we to remove all the votes from User:A ding ding ding ding ding ding ding or keep them? joturn e r 03:26, 19 May 2006 (UTC)
 * This is exactly the sort of thing that should be removed from the tally, but which doesn't need to be struck out or removed. Although if such absurdity grows to an epic scale we may need to take stronger measures against disruption. Christopher Parham (talk) 11:17, 19 May 2006 (UTC)
 * Once again, just make a comment underneath them about why you think it should be disregarded. The closing bureaucrat will make a decision on what to do with it. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 11:19, 19 May 2006 (UTC)

Consensus
Is there a definition of 'reaching consensus' on an RfA? Does that require a minimum percentage of support votes, a supermajority?  SCH ZMO  ✍ 21:00, 17 May 2006 (UTC)


 * >80% = pass
 * <75% = fail
 * >75% AND < 80% = bcrat's decision, based on strength of votes. If everyone voted weak oppose/strong support, the clear consensus is support. --Lord Deskana Dark Lord of the Sith 21:05, 17 May 2006 (UTC)


 * Actually, it's always left to the common sense and good taste of the closing bureaucrat. Typically less then 70% is a failure, more than 80% is a sure win, in between it varies. But this is not a hard and fast rule. --Stephan Schulz 21:08, 17 May 2006 (UTC)


 * Perhaps I'm wrong, but I'm sure I read what I said somewhere. --Lord Deskana Dark Lord of the Sith 21:11, 17 May 2006 (UTC)


 * Deskana's is closer to what seems to be taken for granted. There was a recent 'crat candidate who said 70% to 80% as discretionary and he got hammered for it (does anyone recall who it was?). JoshuaZ 21:15, 17 May 2006 (UTC)


 * Surely that was Essjay. -lethe talk [ +] 21:17, 17 May 2006 (UTC)


 * Yes, it was. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 23:34, 17 May 2006 (UTC)


 * Where do you find this information, and why isn't it in the article? I had thought all you needed was a majority.  SCH ZMO  ✍ 21:25, 17 May 2006 (UTC)


 * Unfortunately I can't remember where I read this, nor can I seem to find it anywhere. Shame. I'll keep looking. --Lord Deskana Dark Lord of the Sith 22:22, 17 May 2006 (UTC)
 * People make it up. It's actually supposed to be a consensus process, but people have been cowing the bureaucrats to turn it into a vote. Grr! Kim Bruning 22:23, 17 May 2006 (UTC)
 * No, people don't make it up. People don't write it down anywhere (and if they do it doesn't stick) because some would like to uphold the impression that RfA is not a vote, even though it is (or is as close to it to make little difference in practice). And it's easier to be delusional about a policy page than about people's minds. -- grm_wnr Esc  09:59, 19 May 2006 (UTC)
 * In what article? If you look on the RfA page it mentions that range explicitly. A little wider range would make the process less of a vote, but 75-80% seems to be what's been decided. If I recall, Cecropia was the first to state the number, though they would go below 75% for some promotions. It didn't happen often though. Solidifying at 75% seems to have happened later. - Taxman Talk 22:29, 17 May 2006 (UTC)
 * It was previously stated in the "Instructions for sysoping someone" section of Bureaucrats but was removed by Uninvited Company as instruction creep. It remains the standard used in practice. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 23:47, 17 May 2006 (UTC)

For those interested: This is not where I first read it, but it is on Supermajority. --Lord Deskana Dark Lord of the Sith 22:26, 17 May 2006 (UTC)


 * which, as you may note, is NOT a consensus :-P Kim Bruning 22:30, 17 May 2006 (UTC)
 * What is the significance of neutral votes? Are they ever reckoned? Anwar 22:34, 17 May 2006 (UTC)


 * In the "vote tally", they are considered by closing bureaucrats in close cases; if there are a lot of oppose-ish neutral votes and the support percentage falls between 75-80%, then they can tip the balance to "no consensus." In most cases, however, they're used by individuals who want to raise awareness of issues, but express no particular opinion about promotion, such as "Neutral - Doesn't seem very active in...". I would venture to label them as most useful for helping other voters to decide how to vote. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 23:34, 17 May 2006 (UTC)

Actually, are bureaucrats still using consensus as their criterium at all? It would be nice if someone came out and actually stated that supermajority was the official criterium for Requests for Adminship, because at least then we could kill it cleanly. Kim Bruning 00:33, 18 May 2006 (UTC)


 * The last time I ran statistics, I was responsible for 50% of the promotions since 1 April (I don't have figures on percentage of total RfAs, only total promotions). So, I can say definately that in all those I've closed, this has been the standard: less than 75% fails, 75-80% is the bureaucrat's descretion, 80% is promote. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 01:15, 18 May 2006 (UTC)
 * Did someone mention bcrat stats? Go here NoSeptember   talk  13:35, 18 May 2006 (UTC)
 * Ok, that's terrible. I'll chalk that up as Requests for admniship being totally broken. Kim Bruning 10:29, 18 May 2006 (UTC)
 * You can say that, and I agree it should be less strictly numerically based, but the fact is most people agree RfA is working pretty well. There aren't many false negatives or positives, and certainly not enough that we should be wringing our hands about it. - Taxman Talk 11:08, 18 May 2006 (UTC)


 * Users may be interested in the numbers at Guide to requests for adminship. Tito xd (?!? - help us)

I don't see why the fact that it's a vote bothers people. Of course, I respect that you feel that way, but I don't understand it. --Lord Deskana Dark Lord of the Sith 06:53, 20 May 2006 (UTC)

Question, please, on how to correctly add comments to an RfA ?
Sorry to interject this dumb question here, but I am still fairly new, and don't know where else to ask it. I noticed that Requests for adminship/Zpb52 contains numerous comments in the voting sections. I commented a strong oppose in a previous RfA, which another editor moved (away from my vote, and to the bottom of the page, under comments). The follow-up I received on my concerns was after the RfA closed (from an editor who took difference with me on my talk page). For the future, I would like to understand the correct usage of comments supporting one's vote: should my comments have been moved to the bottom by another editor, or do comments belong in the voting section, as in the RfA referenced above? TIA. Sandy 23:58, 17 May 2006 (UTC)
 * Personally, I wouldn't move a comment without asking the commentor first, but I understand why someone moved your comment - it was LONG! Generally, you should use no more than a couple of lines explaining your support or oppose vote and elaborate further, if necessary, in the comments section. (You could also say "See Comment below".) Otherwise, long comments would clutter up the Support and Oppose sections. — Cuivi é  nen  T, Thursday, 18 May 2006 @ 02:08 UTC
 * LOL - Ok, thanks for the info. If there's a next time, I'll aim for brevity, and squeal if someone moves my comments without asking. Thanks again.  Sandy 02:30, 18 May 2006 (UTC)

Refactoring RfA totals to remove sockpuppets after they have been closed?
User:Dragon's Blood was found to be a Zephram Stark sockpuppet. This user also voted on a large number of RfAs as part of their long term encroachment to make the sock plausible. As a sock of a banned user, I would presume that the votes it made are not valid. Therefore, should the vote totals be adjusted on the admin stats pages? JoshuaZ 00:16, 19 May 2006 (UTC)


 * Not really. The totals *do* accurately reflect the totals used for promotion, and that is what matters. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 01:38, 19 May 2006 (UTC)


 * Well the totals in all technicality don't mean anything since it isn't a true vote (it isn't truly consensus based either) since the number of support or oppose in actuality seems to act more to a guide for the bureaucrats to look at when seeing if there's a consensus or not to promote. Pegasus1138 Talk 03:27, 19 May 2006 (UTC)


 * This may be true, but the totals in any case do represent the state the RfA was in when it was closed. The tags say "This is preserved as an archive" for a reason: Specifically, so people know what the state of things was at the time the decision was made. If you rewrite the history, then it will only be natural for people to come along and say "Why did this happen?" The bottom line is this: Regardless of whether we find out someone was a sockpuppet two months later, at the time the bureaucrat decided what to do, the vote was considered legitimate. That is what should be recorded. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 03:46, 19 May 2006 (UTC)


 * Is there anyone that was so close that one opinion mattered one way or the other? Those would be the only ones I'd even consider supporting doing anything about... and although I haven't checked I am thinking there haven't been any such in a while.  + + Lar: t/c 03:57, 19 May 2006 (UTC)


 * Of course, there could be. However, I was recalling an RfA I know of recently where there were three sockpuppets. Three oppose votes stricken could affect a tally significantly. Taking two fairly reasonable numbers, 50/15, which is 76.92% support, could quite easily be a no consensus; it would be 50/12 with those three removed, which would be 80.64%, a clear promote. So yes, it could be that close. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 04:36, 19 May 2006 (UTC)


 * Thanks for the analysis. So what do you think? I'm still thinking that even if it "could be that close" mathematically that the changes wouldn't make a difference (and maybe closer to the original question) and shouldn't make a difference. What's done is done, leave sleeping dogs lie, etc. Do others agree?  + + Lar: t/c 15:53, 19 May 2006 (UTC)

If the RfA failed when it should have succeeded, a 2nd nom is all that's needed to fix the problem - they should succeed this time. If it succeeded when it should have failed, then it's a little cruel to take back someone's mop. If they do something wrong, they can go through ArbCom like anyone else, otherwise let's just consider them as lucky. --Tango 16:06, 19 May 2006 (UTC)

IP Votes
On my RfA an IP user mysteriously came along and voted. What is the procedure for handling this - should I just remove the vote and inform the user that IP votes are discounted? Or should I leave it as it is with the minor comment stating so? Cowman109<sup style="color:darkgreen;">Talk 01:31, 19 May 2006 (UTC)


 * Comment it out, that is, replace just # with #: so that it indents it but doesn't interrupt the numbering of votes below it. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 01:37, 19 May 2006 (UTC)


 * Done, thanks! Cowman109<sup style="color:darkgreen;">Talk 01:40, 19 May 2006 (UTC)


 * A caveat: Replacing # with #: doesn't work if the vote is the first one. It still gets numbered so you'll have to remove the # altogether. Kimchi.sg 02:28, 19 May 2006 (UTC)


 * Yes, exactly. However, if the first support vote is an IP vote, it may well be the nominator logged out, as the nominator traditionally casts the first support vote. Might be a good idea to notify the nominator, unless they are signed below. <font color="#7b68ee">Essjay <font color="#7b68ee">( <font color="#7b68ee">Talk  • <font color="#7b68ee">Connect  ) 03:23, 19 May 2006 (UTC)

Surprise RfA nom's
Yesterday there were a couple of unsigned RfA's that were removed quickly. User:Funnybunny had a previous unsigned RfA. Is it possible for someone, for whatever reason, to nominate a user and the user not know about it until everyone had opposed per not answering the questions? Or worse, to not know about it till someone brought it up on a subsequent RfA? Thanks :) Dlohcierekim 12:40, 19 May 2006 (UTC)
 * If a candidate has not yet accepted a nomination, it is inappropriate to put it up for voting, or to vote on it. So, hopefully. Generally, I don't think anyone would hold an unaccepted nomination against a candidate under any circumstances. Christopher Parham (talk) 12:53, 19 May 2006 (UTC)