Archive 215

←

Archive 217

Archive 218

Archive 219

Archive 220

Archive 221

→

Archive 225

"The additional features granted to administrators are considered to require a high level of trust from the community" So it says here - WP:RFA. Many RFA fail not due to competence issues, but because a sufficient number of editors have concerns about the trustworthiness of a candidate.

Editors have different notions on the subject of trust, it is a personal matter. If we are being asked to trust a candidate for a limitless period, without effective recall and for every tool in the box (see above) it is highly likely that for some !voters the threshold of trust will be higher. In comparison, a time-limited or probationary period with a simple get out after a few months, on either side, might encourage a lower level of trust based purely on the lower inherent risk.

Since much tension at RFA revolves around the issue of trust (manifested as maturity, witnessed behaviours, past interactions etc.) it seems an obvious thing to consider that by replacing the instant job for life with an initial agreed probationary period, the demand for absolute trust would be reduced in some cases. It doesn't absolve editors from all responsibilities in assessing a candidate. It simply reduces the risk of perceived long term harm and therefore may enable some editors to be less tenacious in their candidate assessment.

So what is there to prevent the framing of a suitable experiment / trial period? Is it the community at large or long-standing Admins concerned that they will eventually be subject to reconfirmation? Is the founder or WMF preventing a trial of a fixed length (4-6 month) probationary period? Maybe you will say that it would not reduce the problem. If not, why? Leaky Caldron 21:36, 14 October 2012 (UTC)

I personally think a limited duration adminship is a great idea except that the RFA process is such a nightmare few will want to do it if its a limited duration. So if we do this the RFA process needs to be much more painless. Kumioko (talk) 21:50, 14 October 2012 (UTC)

Trial adminship is a great idea in concept form. But it won't happen. The simple fact is that no one will want to follow after the probationary admin and watch over their edits to make sure they are appropriate. And further, it relies on the community actually trusting those who would be doing such checking. As Kumioko notes, RfA is already a process of subjective analysis. This would create a process of even greater scrutiny, which would have post-discussion factors, and require a follow-up process. For these reasons, and others, I personally think that "trial adminship" has no chance of achieving consensus. However, if there is some way to deal with these issues, who knows? - jc37 22:07, 14 October 2012 (UTC)

Upon initial consideration, my main concern is similar to that standing in the way of establishing a formal recall process: administrators would be encouraged to fly under the radar (and refrain from attempting to help solve difficult problems) until the trial period's conclusion, lest they anger editors who retaliate by opposing their adminship's permanence. As a result, their behavior during this time might not serve as accurate reflections of what we can expect from them.

So for this to work, we'd have to nail down some very specific evaluation criteria, thereby instilling confidence that mere dislike could't derail a confirmation. —David Levy 22:16, 14 October 2012 (UTC)

But David that is already a problem for editors. If editors want to be admins then they must walk a careful path and not do anything contentious or else they won't get enough votes to pass RFA. Its happened many many times and IMO a lot of good editors don't have the tools because they choose to participate in areas like Copyright infringement, WikiProjects or other areas where people get very sensitive. Its really not about trust, its about the perception of trust. Many have gotten the tools and had them revoked, others have taken 3, 5 and even 6 times to get the tools and they do a great job proving that the RFA process of picking the favorite editor for admin doesn't work. Using myself again as an example. I have never vandalized anything and have always been a net benefit but I have pissed a lot of people off because I strongly believe in the absence of article ownership. I have argued fervently with several editors who feel they own certain articles or projects that they do not and for this I have been marked by many as a menace. Which means I will never be able to get enough votes to be able to edit protected pages, pull more than 25000 articles into AWB at a time or to see certain Database reports that are protected like Unwatched articles. I don't you would find many that think I would abuse any of the tools but some feel my temperment is not appropriate which may be true because I strongly believe in the 'project not in all the editors. Kumioko (talk) 22:34, 14 October 2012 (UTC)

I agree that this is a problem (and hope that we can devise a solution). —David Levy 23:25, 14 October 2012 (UTC)

I think we all need to focus on finding candidates before overhauling the process entirely. I know some editors are unwilling, but there has to be some who are willing to run the gauntlet, so to speak. I don't think we can have another month without a promotion (a Wikipedia first?). We have about ten days to start a request for adminship on someone so it passes before November, if we don't want another Christmas without Santa. Regards, — Moe Epsilon 01:46, 15 October 2012 (UTC)

You should run Moe!. Kumioko (talk) 01:57, 15 October 2012 (UTC)

Oh, god... Regards, — Moe Epsilon 01:57, 15 October 2012 (UTC)

I was about to suggest the same thing. I've considered running myself, but I think that given my lack of experience, it would do more harm than good. I am willing to run if convinced otherwise, but I don't expect that to happen. AutomaticStrikeout 02:01, 15 October 2012 (UTC)

I would say you should run! However don't do it unless 1) You feel you are ready. "Being the one is like being in love, you just know it..." and 2) think you would continue to edit if you don't get it. A lot of good editors have left after a failed RFA and it would be a shame to have it happen to you (Even though you did vote Oppose on my RFA! :-) Kumioko (talk) 02:06, 15 October 2012 (UTC)

There have been many attempts to find suitable candidates over the last two years. The responses were overwhelimingly that the editors are not prepared to submit them selves to the humiliation of th of the current environment at RfA. Even those who would probably have a walk in the park. So change needs to come first. Kudpung กุดผึ้ง (talk) 02:04, 15 October 2012 (UTC)

Fair enough, I last ran in 2006, and I have no doubts in my mind about the amount of detractors I have gathered along the way since then anyways (considering I had users and an administrator socking on mine). It would all time out pretty nicely since the 24th is my birthday, but I think I have a root canal that looks awfully tempting as well.. Regards, — Moe Epsilon 02:06, 15 October 2012 (UTC)

I think I may very well give it a try. If it goes downhill quickly, I'll just withdraw the request, and wait again until next April, as I initially planned. For the record, I think I would support you, Moe. AutomaticStrikeout 02:09, 15 October 2012 (UTC)

Thank you, I appreciate that. I've seen you around as well and I think I'd swing a support your way as well if you decided to. :) Regards, — Moe Epsilon 02:11, 15 October 2012 (UTC)

(edit conflict)A solution is needed, as all here agree, but I see very roughly three factions:

1. Those who are admins who are not in favour of unbundling the tools - possibly because they have experience of what it is like to be in the firing line once they get the bit and are prepared to work in contentious areas.
2. Among those those who are in favour appear to be some who are not admins, among whom are some who have made an RfA attempt and failed, and others who have not been bold enough to throw their hat into the ring.
3. Those who suggest radical alternatives and that in the case of impasse the WMF should design and impose a new system or set of regulations by fiat; they may have missed the statement by the Foundation only 3 months ago (with their bold text) in response to what was possible one of the best, most coherent, and well presented proposals.

With the level of trust being a criterion for according the bit, those who oppose unbundling say that if a user is to be trusted with the tool he should be trusted with all of them. In a way, this makes sense, because trust is trust, and I can't think why lower levels of trust should be accorded to those who want what would be unbundled, and hence 'minor' rights à la WP:PERM. Thus applications for those unbundled admin rights would indeed require as much scrutiny as RfA and cannot be accorded on the decision of one admin alone.

I have followed many (but not always participated) discussions in which Kumioko was involved. I am 100% certain that he would not abuse the admin tools, but he is telling us himself why he possibly hasn't been accorded them. Trust is not only a requirement for admin tools, but also for admin behaviour.

Trust should also be required from the voters to comment objectively and in a polite, honest, and humane manner; some of them don't. As I see it, the solution is threefold: better publication (without canvassing of course) of current RfAs, clerking, and minimum qualifications for voters - and contrary to the statements of conjecture by some, I'm certainly neither advocating a yet higher bar, nor an abolition of the 'oppose' votes. The actual system not broken and does not need to be changed or supplanted by devolution. Kudpung กุดผึ้ง (talk) 01:53, 15 October 2012 (UTC)

I certainly agree with a need for better publication. I don't check all that often and most RfAs just fly by without me noticing. Eeekster (talk) 02:01, 15 October 2012 (UTC)

You could watchlist the main RfA page. AutomaticStrikeout 02:03, 15 October 2012 (UTC)

That won't get word out to and draw in a wider community, however. -— Isarra ༆ 03:11, 15 October 2012 (UTC)

Well no, it wouldn't. But it would help Eeekster to cut down on missed RfAs. AutomaticStrikeout 03:19, 15 October 2012 (UTC)

No, I've long had a on my watch list. But my list is huge and I don't often see RfA edits. Eeekster (talk) 03:38, 15 October 2012 (UTC)

Oh, I see. Well, I think some editors have transcluded this page onto their user page or user talk page, maybe for that very reason. AutomaticStrikeout 03:44, 15 October 2012 (UTC)

Alternatively, you could always transclude the RfX report box which doesn't take too much room and gives you all the current nominations. Regards, — Moe Epsilon 03:54, 15 October 2012 (UTC) I should click links before I post :P same box. Regards, — Moe Epsilon 03:58, 15 October 2012 (UTC)

I have advocated clerking and limiting threaded discussion to !vote, challenge, response with everything else transferred to the talk page - all of which would reduce the battle mentality of supporters and opposers. So how caan it be made to happen? Not sure what Kudpung means by minimum qualification for !voters but if it manages down some of the "fuck yea!" or "no, not ever" comments added without any obvious thought, it might be useful. Not so sure about advertising but media-wiki is the obvious place. Leaky Caldron 09:44, 15 October 2012 (UTC)

I've mentioned this several times before. If you've missed it, it's here. Kudpung กุดผึ้ง (talk) 11:36, 15 October 2012 (UTC)

It is reasonable to expect a minimum level of experience from a voter. The German & Italian thresholds seem fine to me but I'm not convinced about the net benefit - which would require a lot more analysis to prove. On the subject of trust, bearing in mind this section was intended to discuss probationary periods not unbundling, in an ideal world trust levels would be the same and to some people they might be. But circumstances can determine an individual !voters trust, it is not the same for everyone. For example, not all editors can be trusted with Rollback, not all editors can be trusted to be Admins and not all Admins can be trusted for higher office, which is why there are approval, selection & election processes for everything. It mirrors real life. My threshold for Admin would be lower if it was a responsibility given for a fixed term and I know I am not alone in that position. Leaky Caldron 12:01, 15 October 2012 (UTC)

I think its rather odd that so many in the current RFA are asking and requiring the user to submit more AIV's and spend more time in the admin related areas when at least 4 out of the 5 that looked at from July and august didn't have any or very few edits in these areas and 3 out of the 5 I looked at (a different 3) most of them were unanimous approvals. It also seems like there is a time factor that people want candidates to be editing for a year. So if people think this is a criteria that must be followed here is what I suggest to cut down on the fuss and to save editors from wasting their time in submitting based on the reasoning of the majority of the votes in this RFA. Add these 2 criteria to the existing criteria for adminship.

1. Editors must have at least 1 year before submitting for adminship
2. Editors must have a documented history at AIV, AFD or other places where adminship is needed. Kumioko (talk) 10:43, 16 October 2012 (UTC)

Suggestions for minimum requirements for running for admin always get shot down here. -- Boing! said Zebedee (talk) 11:08, 16 October 2012 (UTC)

And so they should. Why 1 year? What is someone is perfect after 10 months - how does the community benefit by making them wait 2 months? GiantSnowman 11:33, 16 October 2012 (UTC)

Yes, at the moment, I generally tend to agree that we don't need minimum criteria. If we had a large number of clearly unready candidates running, then something might be justified - but since warnings about not running before you're really ready went up, it hasn't seemed to be a significant problem. -- Boing! said Zebedee (talk) 14:47, 16 October 2012 (UTC)

These criteria come and go, and there are different criteria in fashion at different times (people used to voted against people who had no FAs or who had no experience with images). I don't really see that making any of them permanent would improve anything. I don't think there is a hard requirement other than "not currently blocked or banned from the English Wikipedia". —Kusma (t·c) 11:40, 16 October 2012 (UTC)

I think a part of it has to do with the answer the candidate gives for question #1. If they say they want to work in deletions, people will look for XFD !votes and CSD tagging; if they want to work with vandalism, people want to see AIV reports; and so on. When the candidate is not intending to work in an area, people don't care so much about lack of experience in that area, particularly if the candidate has shown general experience elsewhere and it seems we can trust them to use appropriate caution if they do jump in to one of these areas. Anomie⚔ 14:36, 16 October 2012 (UTC)

I think the easiest way to explain it is to start with the statement that the process of determining suitability is somewhat similar to the scientific method. A theory that a candidate will make a good administrator is proposed, leading to other editors searching for evidence in an attempt to disprove the theory. If no evidence is found within a certain period to disprove the theory that the candidate will make a good administrator then the theory is considered proven. So the process is "negative" not "positive" driven. Given this process, it is inevitable that "undocumented requirements" will come up - and they should. If they didn't, then adminship would be a checklist driven exercise that anyone could game. QU ^TalkQu 15:10, 16 October 2012 (UTC)

Perhaps we should do away with question 1 entirely. It makes no sense to confine our pool of admins purely to people who have specific admin tasks in mind. If there are well qualified editors out there who have no objection to being admins but no definite plans on what to do as an admin, we shouldn't be actively excluding them. --regentspark (comment) 17:53, 16 October 2012 (UTC)

Quite. That was the basis I ran upon in 2006,^[1] and from the reaction it got you'd have thought that I had either just arrived from the Moon, or was swearing in church, or both. It took two more attempts^[2][3] (making the same points each time) before I got through in 2007, at "peak RfA" - it seems I was lucky. It's a principle I believe in passionately, but one that's lost on almost all the RfA commenters. We would do better if it was widely understood. — Hex (❝?!❞) 18:26, 16 October 2012 (UTC)

I think a lot of folks who try look at is as I do unfortunately. I tried and the answer was no. I tried again about 4 years later and the answer was still no. So now if I run again at all, which I likely won't, it will be a good long time again. Although, since more and more content is getting protected and wait times for things like CSD and AIV get longer, there may be no point in running at all because the system will be so overwhelmed I won't want to do it at all. Plus it becomes an effort of futility. If the editor is told no enough times in a harsh enough manner, which has happened repeatedly in the past, they just quite editing and we lose them completely which is also not a good thing. I would also not Hex that using you as an example of how badly broken the RFA process is you had a fair number of opposes all three times. Since as far as I can tell you have never abused your tools, deleted the main page or caused anything remotely resembling Wikiarmageddon, that our process for choosing Admins isn't particularly effective and we are dismissing a lot of good candidates as unsuccessful because of too many opposes. This might be a time to go with a simple majority promotion rather than the percentage basis we have been using.Kumioko (talk) 18:38, 16 October 2012 (UTC)

That's essentially the opposite of what I believe should occur. As I commented at the village pump, opposition based on ridiculous rationales (e.g. "has never written a featured article") should be disregarded. —David Levy 20:40, 16 October 2012 (UTC)

I agree; clerking for RfA should be extremely strict. — Hex (❝?!❞) 08:14, 17 October 2012 (UTC)

Today's Signpost has a piece proposing an "Administration Committee" to be in charge of admin selection. Quote:

What we need is a system that accomplishes three goals: (1) the community is in charge; (2) adminship is easy to give and easy to take away; (3) admins do not need to be politicians.

Having an AdminCom would accomplish all of these goals. The community would ultimately be in charge, because the community would elect the members of AdminCom. We would simply replace direct democracy with a representative democracy. Handing out the buttons and taking them away would be as easy as AdminCom decides to make it. Admins would be directly answerable to AdminCom rather than to the editing public at large, and so would get clearer guidance than they currently do.

The primary power of AdminCom would be to give and remove adminship. Other powers, such as overseeing and advising admins, would naturally follow from this. I do not believe it would be a good idea to formalize internal procedures for AdminCom: the members should be able to decide for themselves how to make things work. Naturally they would express their beliefs about proper procedure when they run for office, and the Wikipedia community could make its choices accordingly.

— Looie496, AdminCom: A proposal for changing the way we select admins

It's already generated a fair amount of discussion on the page. Personally, I think it would be a bad idea for giving adminship but a good one for removing it in cases of abuse. — Hex (❝?!❞) 08:11, 17 October 2012 (UTC)

Since bureaucrats (along with stewards, etc.) are the only users with the permissions able to give and take adminship, it would essentially be the same group of about ten to fifteen (active) users deciding who gets it and who doesn't. However, if it isn't designed to be bureaucrat-only, and the community gets to participate in that process, then how is it inherently different than RFA? Regards, — Moe Epsilon 08:54, 17 October 2012 (UTC)

In the case of de-sysop-ing (I can't find a way to type that word that looks right to my eyes), could it not be the case that bureaucrats would be bound to follow the decisions of such an elected group? In other words, the task of making the decision to remove the sysop bit would be devolved to the new committee, but not the physical power to do so; that way elections to the AdminCom could be held separately. Bureaucrats would retain the discretion to close RfAs and create new administrators. — Hex (❝?!❞) 09:06, 17 October 2012 (UTC)

I suppose that would work in theory, but I think the community is already in disarray with the few committees it does have, and it's unlikely we (the community as a whole) would support a small group that has that kind of power. Now when I say small, I mean about the size of ArbCom. If this was a very large committee that was elected in, say about the size of 100 users of mixed roles (non-admins, admins, bureaucrats, etc.) to represent the community as a whole, where abuse of the committee could be kept in check, then it might have a chance at working. Regards, — Moe Epsilon 09:26, 17 October 2012 (UTC)

We seem to be asymptotically approaching ancient Greece. :) — Hex (❝?!❞) 14:07, 17 October 2012 (UTC)

Any system that is implemented to make it easier to desysop must also come with a system to make it easier to promote admins. If we elect a board to make quick desysops, but leave the current RfA system as the only way to promote admins, that would be a huge mistake. -Scottywong| speak _ 15:24, 17 October 2012 (UTC)

I suggested that if such a committee had a majority of non-admins it could be a useful gatekeeper for RFA/de-RFA community votes. Making evaluation more consistent and deliberative would be a good thing. And this approach is far more plausible than handing over all power over adminship. Rd232 ^talk 11:04, 17 October 2012 (UTC)

In many cases folks aren't admins because their judgement isn't trusted (this especially applies to those who take an active interest in the process) - would this not affect the effectiveness of such a committee? -— Isarra ༆ 14:29, 17 October 2012 (UTC)

Presumably those folks wouldn't get elected to the committee. And even if one or two oddballs did, they shouldn't be able to prevent the rest of the committee reaching sensible conclusions with majority approval, whilst maybe providing a different perspective that could actually be valuable. Rd232 ^talk 15:40, 17 October 2012 (UTC)

Something is needed that abolishes the incivility, flippant and disingenuous voting. This isn't it. Kudpung กุดผึ้ง (talk) 14:49, 17 October 2012 (UTC)

With apologies if this has somehow been discussed to death previously, but has there been any thought given to administrators actively going out and mentoring and recruiting potential other administrators? Maybe instead of tinkering with the process we can, I dunno, apply some elbow grease to the problem?--Tznkai (talk) 22:01, 17 October 2012 (UTC)

Yes, SNAFU, welcome to Wikipedia. — Preceding unsigned comment added by 88.104.5.244 (talk) 23:10, 17 October 2012 (UTC)

Admins and other editors are constantly on the lookout for potential candidates. As discussions on this page and elsewhere will reveal, the majority of those canvassed declined because they do not wish to run for office as long as RfA is allowed to be a venue for unpleasant behaviour. This was mentorship programme at Wikipedia:Admin coaching which is no longer active, but Wikipedia:Editor review can often provide useful feedback and advice. Kudpung กุดผึ้ง (talk) 23:21, 17 October 2012 (UTC)

That is not a new idea, Wizardman, myself and others have done so in the past. My two cents on the matter is that I always looked for people not interested or considering adminship, and I don't see any reason to honestly recommend to somebody like that to go thru a RfA in the environment we've created there for the past couple of years. Snowolf ^{How can I help?} 11:37, 18 October 2012 (UTC)

Out of all the renewed discussion that's taken place here in the wake of the month without any promotions, there have predictably been a range of suggestions, both old and new, for reforming RfA. Equally predictably, each of them has received a thorough measure of support and criticism. However, the one thing that I've been able to detect so far that we generally agree upon is that bad/incorrect commenting in RfAs needs to be stopped.

It's worth reading the latter half of this comment by MBisanz where he suggests that [crats] should take a more active role to nip/hat discussions earlier, block more readily, and strike faulty comments/!votes quicker, but that we may be [l]acking the crat staff/will to do this. For that reason, I think it's time to give another round of attention to the proposal of RfA clerking.

My initial thought would be that, should we go ahead with it after hashing out an official specification (Kudpung's selected list of tasks seems a good place to start), clerks could be selected by a similar "RF..." ("RFRC"?) process to what we have already. Being a less contentious role - no superpowers or special bits/flags, - it shouldn't be too hard to have people succeed. There would be no "no need for the tools" opposes, and it could be explicitly set that candidates didn't need to be evaluated on the basis of their article writing, which would eliminate "1 GA" opposes and their ilk. — Hex (❝?!❞) 09:14, 18 October 2012 (UTC)

I'm not keen on restricting how editors can comment on RfA. Even issues I see like extraordinary long votes will be seen by many as acceptable. I'd be quite interested to see what happens when Mbisanz raised changes at BN, and depending on the outcome there, might even reconsider running for 'crat. Worm^TT(talk) 09:20, 18 October 2012 (UTC)

Thought I think a key problem with any RFA reform suggestion is the lack of a trial period. I think an RFA with clerking on the comments would be a great idea, personally, but without the community seeing a non-hypothetical, full-fledged test RFA with a suggested reform, I don't think we'll ever get any. With new tools like pending changes, it's been pushed through and is going to happen simply because we got a trial period where we got to test it. The broader community would never support a trial period of any RFA reform where an editor may be promoted through it, or for that matter, equally as discouraged coming out at the end. Maybe that's just my opinion, though.. Regards, — Moe Epsilon 10:59, 18 October 2012 (UTC)

Well of course there should be a trial period. I don't see why you think the community wouldn't support an RfA that happened during it, this is hardly the most contentious of proposals. — Hex (❝?!❞) 11:35, 18 October 2012 (UTC)

Personally, I don't think it's contentious at all. :) There is, however, a small fraction of the community (apparently not small enough to make any reform impossible, though) that would see a promotion outside of the RFA process other editors or themselves had to go through, and would immediately think of it as contentious. To the best of my knowledge, this is why I think no trial periods have ever occurred. Regards, — Moe Epsilon 11:44, 18 October 2012 (UTC)

Ah, I see. And yes, that fraction. Well, y'know, the good thing about introducing clerking would be that it wouldn't fundamentally change the process of being selected, so hopefully (hopefully!) that would be sufficient to assuage the more change-averse elements that a trial would be worthwhile. — Hex (❝?!❞) 12:38, 18 October 2012 (UTC)

I don't think moderation of one's opinion at RfA is a good solution. It's a very thorny issue for me personally: I think that anybody should be allowed to say anything that's not a personal attack at RfA, and we shouldn't use in effect a majority rule to shut their opinion, however unpopular or misguided it would be. It instead should be up to the community to judge with their support or opposes, with their comments and discussion if the candidate is okey or not, if the points one raises are correct or not, if they're oppose-worthy or just minor concerns we don't have to be bothered by. Frankly, as long as one is not trolling nor making personal attacks, we should welcome all feedback, however silly or misguided it sounds to each of us. I'd like to point out that personally, I still fail to see the problem even with Kurt Weber's infamous opposes. I don't share his ideas, quite the contrary, but I think he was fully convinced of them, and had a right to voice them. And yes, we can disagree with him, or people that say "Need 1GA" but it's their own opinion, and we accomplish nothing by shutting them out of the RfA process. It is up to the community to simply say that it's silly when people raise such objections, if the community feels that way (and I'm not so sure about it these days). Snowolf ^{How can I help?} 11:45, 18 October 2012 (UTC)

• An oppose is a personal attack and opposers are expected to go into some detail. The point of WP:NPA is that we should "Comment on content, not on the contributor." This constraint does not apply at RfA because we're not dealing with content there and the point of the process is to assess the personal qualities of the candidate. Warden (talk) 11:59, 18 October 2012 (UTC)

• An oppose is most certainly not a personal attack. GiantSnowman 12:06, 18 October 2012 (UTC)

• Oh yes it is. One can soften the blow by, for example, saying that the candidate would be better suited to some other role such as a content creator, say, but by opposing their desire to become an administrator, one is necessarily attacking their competence or temperament. This is a systemic problem because most users don't want to be nasty and so they will only oppose when it seems safe to do so - after there is enough weight on the oppose side for there to be safety in numbers. See groupthink for more details. Warden (talk) 12:23, 18 October 2012 (UTC)
• Not all opposes are, but it is certainly true in many cases. — Hex (❝?!❞) 12:40, 18 October 2012 (UTC)

• I think that interpretation of WP:NPA is far too strict. Just because it says, "Comment on content, not on the contributor" doesn't necessarily mean that any comment about a contributor is a personal attack (because then support votes would be personal attacks as well), nor does it mean that any negative comment about a contributor is a personal attack (because then constructive criticism would be a personal attack). There are ways to express opposition without personally attacking the candidate, and many people find this challenging. This is where RfA clerks could help immensely. -Scottywong| soliloquize _ 13:58, 18 October 2012 (UTC)

• The last few opposes at my RfA have been like a punch in the gut. I can understand when someone opposes because of some matter of not meeting their criteria, however arbitrary, but when it's borderline slander that a candidate has to sit there and take, I think there could be a line where it becomes a personal attack. Gigs (talk) 20:48, 19 October 2012 (UTC)

• I am sorry to read of Gigs' dismay regarding the comments at the current Rfa. However, count me among those expressing considerable concerns about the moderation of comments at Rfa. It is a slippery slope, once it begins, and inasmuch as the issue of admins having lifetime tenure and the power to block editors resonates strongly for many rank-and-file Wikipedians, I think the clerking proposal could generate a lot of heat and just wind up on the cutting room floor. I would support active admins who are uninvolved with the candidate in any way whatsoever doing civility work at Rfa, like warning obviously uncivil commenters, and in extreme cases refactoring their clearly offensive commentary such as cursing, but otherwise I believe we have to grin and bear those with strongly expressed and even very lengthy comments in their !votes. Jusdafax 21:05, 19 October 2012 (UTC)

  As a general remark to anyone reading this, I think the best way to be able to understand when a candidate complains about the comments in oppose votes is to have gone through an RfA yourself. I think when someone has experienced what Gigs has for themself, they should be more inclined to be polite and helpful when casting their !vote. Perhaps the best strategy for !vote rationales is to ask yourself if you would find your comments offensive if they were made about you. AutomaticStrikeout 21:21, 19 October 2012 (UTC)

  I have no problem with strongly expressed comments, and I definitely wouldn't want to see "civility policing" either. The comments that bothered me are the ones that are patently false. I'm not sure there is a solution within our current RfA framework. Gigs (talk) 21:35, 19 October 2012 (UTC)

  Yeah, having just gone through an RfA, it's easier to understand when another user does not want to do so. AutomaticStrikeout 21:40, 19 October 2012 (UTC)

  (ec) I've been through 2 RfAs and an RfB, and I'll agree with Gigs that it's the misrepresentations and the flat falsehoods which are typically the most "annoying". That said, as someone else said on this page somewhere, admins (unfortunately) have to deal with that more often than they should. So it's at least good that we get to see how a candidate handles that. Besides, what clerk would be able to "fact-check"? Should we expect clerks to check contribution histories for the other commenters who may choose to not bother, and just drive-by "vote"? Of course not. The "broken" part of RfA is the people. What their expectations and motivations are. We can spin our wheels on WT:RFA all we want, but RfA won't change unless we find a way to change that. - jc37 21:45, 19 October 2012 (UTC)

Pardon me if this question is redundant; I've tried a quick archive search and nothing quite touches on what I'm thinking of. I'm wondering if it's possible to confirm an editor for adminship with restrictions on how or where he or she could operate, something like a topic ban. It may seem unusual that you'd want to give such an editor the mop at all, but let me present a scenario. I promise this is hypothetical; I've never interacted with the two current RfA candidates, so this has nothing to do with them. For ease of writing, we'll presume a female user:

Suppose there were an editor who you generally respected. She exercises good judgment in venues such as AfD, she's a respected contributor of mainspace content, and she demonstrates good demeanor and temperament when it comes to interactions with other users. She also has an impressive edit count and account history. You think she'd be a great administrator. But suppose she were fanatical about the Macedonia naming dispute and couldn't keep cool or be neutral on such topics, so you wouldn't want her exercising administrative powers there.

First of all, how would you vote if she came up for RfA? Specifically, suppose an editor proposed she be made an administrator but prohibited from acting as one on Macedonia-related articles. Should an isolated tendency keep her from adminship altogether, or should her talents be exercised where appropriate? If you feel strongly about Macedonia, substitute another issue. The specifics aren't important. --BDD (talk) 01:39, 19 October 2012 (UTC)

If she can't keep cool, she probably shouldn't be an admin. AutomaticStrikeout 01:45, 19 October 2012 (UTC)

I don't see why an admin can't be topic-banned. Or even being an admin have the self-control to abstain from approaching that topic in the first place. -- œ^™ 02:51, 19 October 2012 (UTC)

I think our expectation of admins is that they should have self-control, no? If they demonstrably don't in an area, then I doubt they'd pass an RfA. — Hex (❝?!❞) 14:22, 19 October 2012 (UTC)

To sum up my feelings on this: Why would we agree to give someone a mop while at the same time saying that they can't enter certain areas of the house as they are a threat to spill crumbs there? And Hex is right, I doubt someone like that would pass an RfA. AutomaticStrikeout 20:59, 19 October 2012 (UTC)

• Wouldn't the normal rule apply, to never use the tools to gain an advantage in a dispute in which the admin is involved? --213.196.214.215 (talk) 02:55, 19 October 2012 (UTC)

I suppose that depends on how broadly you interpret that phrase. Our hypothetical user could be considered to be "involved" in any dispute over Macedonia, whether she's personally involved or not, but it's not hard to think of gray areas. --BDD (talk) 15:17, 19 October 2012 (UTC)

• Should it, assuming they stick to a topic ban on the named article(s)? No. Will it prohibit them from passing? Absolutely. Regards, — Moe Epsilon 06:08, 19 October 2012 (UTC)
• Admins shouldn't have to be topic banned - if a candidate was topic banned, or the topic ban was a condition of their candidacy, I would definitely oppose. It shows poor judgement and there would be too many concerns about abusing tools. GiantSnowman 15:33, 19 October 2012 (UTC)
• I'm an admin and have some limited areas of WP where I have done quite a bit of editorial work - Wicca is an obvious example. I would not exercise my admin powers there to block anyone because by definition I'm heavily WP:INVOLVED. Had anyone asked me at my RfA whether I'd use my admin powers in that way, it would have been a reasonable question and I'd have happily said no. However I hope that even there I remain cool and neutral. An editor who couldn't do so, whether on an area of intense personal interest or elsewhere, is not one I'd want as an admin. So, should admins be careful about using the mop in areas of personal interest? Yes. Is it safest if they agree to a self-denying ordinance not to do so? Yes. Can I imagine a person being trusted enough for the mop, but so untrustworthy in one area that they need a topic ban? No. Kim Dent-Brown ^(Talk) 15:51, 19 October 2012 (UTC)

I've developed a tool that attempts to grab a bunch of information about a user, and spit out a score value that indicates their readiness for adminship. It looks at various stats that are popular among typical RfA criteria posted around the 'pedia. There are certain things that it obviously can't factor into the score, like signs of competence/maturity, history of vandalism or sockpuppetry (for which the user wasn't blocked), civility, understanding of policies, etc. However, it is good at scoring the raw experience level of a user, and I think it could be useful to eventually add to WP:GRFA and other such pages.

It is a new tool and it definitely needs some refinement to be useful (particularly in how many points are awarded for each criteria). If you have a chance, try looking up some users, see what you think about the score values assigned to them, and post some comments here. I can also post the specific scoring rules at some point in the future, so that we can all comment on them and tweak them. Also, if there are any other criteria that you think the tool should look at and factor into its score, please suggest it here (but anything that would substantially increase the time it takes to generate the score will probably be denied).

It currently doesn't care if the user you're searching is already an admin, so you can use it to look up admins as well (and hopefully their scores should be high!!!). The tool can be found at http://toolserver.org/~snottywong/adminscore.html You'll see the breakdown of all criteria measured and the scores for each criteria on the results page. Thanks! -Scottywong| spill the beans _ 00:49, 4 October 2012 (UTC)

It also doesn't take into account if the blocks were good or bad - it picked up where I blocked myself by accident less than two years ago. --Rschen7754 01:27, 4 October 2012 (UTC)

Interestingly, if Scottywong had the same user rights as when he passed his RFA, he would fall below the 1000 threshold. Also an interesting benchmark, Jimbo is less then halfway to being an admin. Monty₈₄₅ 01:36, 4 October 2012 (UTC)

You would think being the "founder" would give you +9001 points. LegoK^ontribsT^alkM 01:44, 4 October 2012 (UTC)

(edit conflict) Yeah, I've blocked myself a bunch (easier than blocking a test account, at least for tech-clueles me) and the tool does not like that much. Is there a way you can filter self-blocks or blocks undone in a certain time period (under 5 minutes, say?) Keilana|^{Parlez ici} 01:51, 4 October 2012 (UTC)

Ha, I score just 1217, so I guess I'm only marginally qualified. Then again, Scotty did oppose me, maybe he was on to something ;) Fun little toy, but I prefer my judgement over an algorithm's when judging suitability. Dennis Brown - 2¢ © Join WER 01:56, 4 October 2012 (UTC) added - After looking at a couple more admin, I don't feel so bad.

The real problem with this is that it turns adminship which is very non-statistical at its core into mere numbers. As for the blocks, it counts a block that's a ban the same as 3-hour block--not to mention whether the block was valid in the first place. This tool is interesting, but honestly, I see great potential for misuse by partcipants and abuse of candidates by its use. I do not plan to use it.PumpkinSky talk 01:57, 4 October 2012 (UTC)

I have several comments after playing with the tool for a few minutes.

• The tool is vulnerable to XSS, here for example (NoScript users can use [1] for the one without images). Users should be wary of clicking links that lead there until Scottywong runs his output through cgi.escape() or a package such as MarkupSafe.
• Checking a user who has had no edits to article space will result in a ZeroDivisionError, here for instance.
• Checking an account that has made no edits and has made no log actions results fails due to ts (maybe it stands for 'timestamp'?) being a NoneType instance, which does not support __getitem__.

Please fix these issues. →Σσς. 02:07, 4 October 2012 (UTC)

• As many video games have shown (in my opinion), you simply can't quantify experience. The more this tool takes into account (like number of automated edits, AIV edits, RPP edits, etc.), the closer it will be. However, it should never be used as an official metric for adminship, because it will only be a rather rough estimate.--Jasper Deng (talk) 02:10, 4 October 2012 (UTC)

(edit conflict) Scotty, you might also want to take into account whether the account is currently blocked - it just so happens that I scored a single point lower than JarlaxleArtemis! --Dylan620 ^{(I'm all ears)} 02:13, 4 October 2012 (UTC)

Does Scotty really need to make fixes for trivial cases in which the person being scored would never be promoted anyway? Although, who knows, maybe a day will come when we see: "Support. He scored over 1000! Who cares if he's banned?" Someguy1221 (talk) 03:27, 4 October 2012 (UTC)

Lol, I just thought it unfortunate that I scored lower than a LTA. I had figured that having the program recognize current blocks would come in handy, but then a currently blocked user wouldn't be running for adminship, now would they? :/ Gah, I've been away too long. --Dylan620 ^{(I'm all ears)} 04:14, 4 October 2012 (UTC)

I think I should get way more than 1 point for having a userpage that's exactly 28 bytes. 28bytes (talk) 03:35, 4 October 2012 (UTC)

• You're cheating, with your 13,175 bytes worth of template. Your desysop is in the mail, and it's not much longer than 28 bytes, you frothy rump-fed vassal. Drmies (talk) 14:30, 4 October 2012 (UTC)

Same with my userpage, which is also a transclusion.--Jasper Deng (talk) 03:37, 4 October 2012 (UTC)

Ouch. I wasn't planning on running anytime soon, and by that I mean never, but figured I'd score higher than just past half an admin... --Onorem♠Dil 03:43, 4 October 2012 (UTC)

I've tried to make it as clear as possible that this tool is not intended to be an "official" metric for adminship, but only a rough estimate, particularly targeted towards hopeful candidates who are not quite ready to run, in the hopes that it can give them some guidance on what they might need to work on. While the tool allows you to look up editors who are already admins, I don't think it would be worth the effort required to make the tool detect (for example) instances of an admin blocking themselves accidentally, because the tool is not for admins. I appreciate the bug reports and I will fix those tomorrow. Finally, the 1000 point "threshold" is a very early estimate, based on very few searches. I expect that threshold will change, once we have more chances to search a wider range of users, tweak the point system, and add/remove criteria to make the tool more accurate. At this point, I believe that someone who scores over 1000 is not necessarily someone who would be a good admin, but rather they are probably someone who would realistically be able to start an RfA without having it closed early, and perhaps have a better than average chance than succeeding.

Have you found users who scored less than 1000 who would otherwise be realistic candidates? Let me know. Do you have personal RfA criteria by which you measure candidates that this tool doesn't use? If they are practically measurable by an automated tool, let me know. Thanks again. -Scottywong| speak _ 03:59, 4 October 2012 (UTC)

Yes you did post that caveat, but the problem is there are way too many people who will ignore it. PumpkinSky talk 09:40, 4 October 2012 (UTC)

Can you have it check the number of edits to particular pages such as WP:AIV and WP:RFPP? Those seem to be commonly used metrics, in that they indicate activity in areas that require the admin tools. —Torchiest ^talk_edits 04:16, 4 October 2012 (UTC)

Not knowing how things are scored, it's tough for me to give feedback. My account age is apparently the most important factor in my score. My edit count was second. My user page length matters? Guess I'll put a bunch of worthless info there just to make my 'credit' score higher. That I use edit summaries 99.9% of the time is worth the same amount as having one 6 year old block on my account. I'm hoping where I got killed was on the whole last 12 month thing. I'm not suggesting I'm a realistic candidate, but scoring is the 600s was not what I was expecting. --Onorem♠Dil 04:20, 4 October 2012 (UTC)

I'm an admin having got the bit earlier this year and I get a score of 705. I'm fairly sure that 50 of that is because I've since given myself the abusefilter right so presumably I was no higher than 655 when I passed. I will admit that my success at RfA had a lot to do with working in an esoteric area (copyrights) but it still goes to show how coarse a tool like this will be. Dpmuk (talk) 06:25, 4 October 2012 (UTC)

Dpmuk, I found yours and Onorem's scores to be interesting, and I think they highlight that the scores for activity in the last 12 months and edit counts to various namespaces are overvalued in comparison to other criteria. I arbitrarily decided to limit the range of scores for each criteria between -200 and +200, but since the activity in the last 12 months is actually 3 criteria, that counts for a maximum of +600. Same for namespace edit counts. So, if you are a bit lower than normal in both of those criteria, your score will be drastically lower than it should be. I'd like to post the scoring algorithms somewhere for comment, but I'm trying to think of a non-overwhelming way of doing that. Hopefully will get that done in the next 8-10 hours. Thanks for the feedback. -Scottywong| verbalize _ 13:34, 4 October 2012 (UTC)

I also wouldn't recommend even trying the tool passed a certain edit count. I have around 90,000 and I tried to use the tool. I fell asleep and I woke up to my internet timing out before the tool finishing. Regards, — Moe Epsilon 08:32, 4 October 2012 (UTC)

Yeah, it might be advisable to reject any queries on users past a certain edit count, or limit how far back it goes. I bring up the point about the blocks as something to potentially add to the disclaimer, since there's no way for a bot to know if a block is good or bad. --Rschen7754 08:38, 4 October 2012 (UTC)

Probably not a bad idea, although limiting how far back it goes will produce somewhat inaccurate results (i.e. if you created a lot of articles early on, but none lately, it might list your created article count as zero and deduct a lot of points). Most of the time it takes to pull the information is based on edit count, but it can be adversely affected by toolserver's load at the time. If someone was running an expensive SQL query at the same time you looked yourself up, that might have been the reason for it timing out. I just looked up Moe Epsilon and it took over 10 minutes to complete. Your score was 1019. -Scottywong| gab _ 13:52, 4 October 2012 (UTC)

How strange, after reloading the page about three times and manually typing my name into the URL, the tool finally accepted the query. With a load time of thirty-seven seconds, I got a score of 1018. I'm not particularly pleased with: "% of edits in Article namespace: 67.8% +29" and "% of edits in Wikipedia namespace: 5.1% -49". Not sure if we should send the message of article writing is less important than editing the Wikipedia namespace. Regards, — Moe Epsilon 08:43, 4 October 2012 (UTC)

And not being blocked gives 200 points alone; I am quite sure almost every admin tried blocking himself at least once...Lectonar (talk) 08:59, 4 October 2012 (UTC)

Some useful aspects, but you're right that it does not make for a 100% replacement. My main account scored just under the 1000 threshold because the I seem to have a couple of 0 edit months over the last 12 ... well, um, yes that's true :-) dangerouspanda 08:47, 4 October 2012 (UTC)

Damn, I only got a score of 585! Pretty funny when you consider I got +200 points for having other permissions. I should resign my administrator rights, and probably should resign my bureaucrat rights too. --(ʞɿɐʇ) ɐuɐʞsǝp 08:54, 4 October 2012 (UTC)

I am an admin, crat, oversighter and a member of the Ombudsmen. I've been around for seven years. I have 40,000+ edits, mostly manual, and a stack of featured articles to my credit.

I scored 1306.

Based on my own score, and those reported above, I think if you're looking for new admin candidates to score 75% of what I do then either the tool is currently worthless or the bar you're suggesting is atrociously high.

On the detail:

• the tool could check for Featured material. WP:WBFAN is a good source for this, although it would still ignore other featured material.
• the %s for different namespace edits seem wonky. I'm a collaborative editor, who liaises with others to find consensus or to work on articles. I therefore have a large % of edits to talkspace. This seems to be penalised, when I think it should be encouraged. I have somewhere more than 10,000 mainspace edits, so it's not like I'm just here to chat.
• You also seem to undervalue length of time here.
• An ability to differentiate between manual and automated edits might be a useful refinement.

Hope that helps --Dweller (talk) 10:30, 4 October 2012 (UTC)

I would argue that the qualities which make a good admin are not necessarily well correlated with doing lots of work on FAs. bobrayner (talk) 10:54, 4 October 2012 (UTC)

They're not "necessarily" correlated. I've supported a fair number of gnome admin candidates who couldn't develop an FA if their life depended upon it. But it's impossible to develop an FA without a proper grasp of all sorts of useful adminny things, including understanding a slew of policies, getting to grips with consensus building and demonstrating an ability to stay cool. -Dweller (talk) 11:45, 4 October 2012 (UTC)

The tool should have considered user talk page length (including archives) rather than user page length. --Anbu121 (talk me) 11:01, 4 October 2012 (UTC)

Getting this sort of model right is always tricky - I was wondering if it would be possible to have a variation that tests someone's admin score at a given point in time? If that was possible, you could run it against successful and unsuccessful RfAs to try and refine the model to increase its value as a possible indicator of success. - Bilby (talk) 11:03, 4 October 2012 (UTC)

• To answer scottywong way above, yes, I specifically found an excellent candidate that scored well under 1000. Actually, as someone else pointed out, your score isn't so high. It is an interesting toy, granted, but there is no way to quantify the most important aspects of being an admin, regardless of the code: Clue, understanding of policy, patience, willingness to help others, fairness. I would rather have those traits and someone with a crappy user page and who had to go TDY in the military for a month once a year (both point losers), than someone who edits a lot and snaps at new users, which would look better on "paper". Dennis Brown - 2¢ © Join WER 13:28, 4 October 2012 (UTC)

I thought there was supposed to be an Admin shortage and backlogs to clear? Evidently not, if time and effort can be expended on developing and discussing something no more likely to be adopted into general use than I am of becoming an Admin. Leaky Caldron 13:39, 4 October 2012 (UTC)

Well, I don't know about anyone else, but I'm currently considering (more than considering, really) a run at RfA. Got a couple solid nominators, imo, so I can't be totally crazy or off-base. Scored in the 400s. So, there's that. I'd add that, if the edit breakdown's based on the same thing that TParis's is, then it doesn't count deleted contribs, which murders (so to speak) the ratios of anyone with a large CSD log. The placing of the SD tag in mainspace doesn't count towards the ratio (since it gets deleted), but the log entry and notification to the user does. Writ Keeper ⚇♔ 13:50, 4 October 2012 (UTC)

Writ Keeper - run for RFA is you feel ready, don't let this tool dissuade you. As someone said above, even Jimbo wouldn't "pass" based on this. It's nothing more than a bit of fun. GiantSnowman 13:55, 4 October 2012 (UTC)

(edit conflict) Monthly edit counts and namespace edit counts are currently way overvalued. I'll be overhauling the scoring system soon, so please don't let the score dissuade you from running. -Scottywong| talk _ 13:56, 4 October 2012 (UTC)

Oh, no, I wasn't going to let it dissuade me from running or anything. Just another datapoint in the plot. Writ Keeper ⚇♔ 13:58, 4 October 2012 (UTC)

In fact, if things appear like they're starting to get out of hand, I will specifically disallow looking up anyone who is currently at RfA. Feel free to request that in advance if you like. -Scottywong| chat _ 13:59, 4 October 2012 (UTC)

Nah, to anyone who would actually base their !vote entirely off of this number (few and far between, one would hope), it would probably look like I have something to hide or something. The stats are what they are; it's not a big deal, really. It's not like the actual statistics are inaccurate (other than the namespace breakdown, but that's an issue with all of these kinds of tools, not this one in particular). I don't really understand the valuation of each statistic, but that's whatever; it's a prototype, and not intended to be the be-all-end-all anyway. I'd take people's advice if they think blacklisting is a good idea (not having had the experience of RfA myself to judge from), but I don't think it's a problem. Thanks for the offer, though! Writ Keeper ⚇♔ 14:07, 4 October 2012 (UTC)

It's an interesting toy, and I commend ScottyWong for the initiative in developing it. It has a lot of weirdness in it, though. (Since when has anyone ever evaluated an admin candidate based on the size of their userpage? What's that doing there?) As far as its value in evaluating candidates, I think it may over-emphasize content creation, which is important but does not necessarily fit a person for adminship. Personally I scored 1226 - almost as high as the admin/bureaucrat/oversighter/ombudsman Dweller. That would seem to make me a prime candidate. And yet I am deficient on a lot of the criteria I see people apply here at RfA; there are vast areas of Wikipedia where I have no experience or knowledge at all. IMO this might have value as a pre-screening tool, after tweaking, but it's not going to help much in the actual decision to grant tools or not. --MelanieN (talk) 14:45, 4 October 2012 (UTC)

BTW the fact that many current administrators are scoring poorly is not necessarily a defect in the program. Once a person becomes an admin, they are likely to spend a lot of time doing things that aren't counted by this program - which is designed to evaluate people who CAN'T do housekeeping chores, and who certainly can't block themselves. --MelanieN (talk) 14:50, 4 October 2012 (UTC)

It is not so simple about the content creation: I have over 90% of my edits in the article space, and I scored negative on all items related to the percentage (article, WP, and talk). Apparently, the tool has is some ideal distribution of edits, punishing deviations from it.--Ymblanter (talk) 15:03, 4 October 2012 (UTC)

Yes, this is probably the weakest aspect of the scoring algorithm at the moment. The ideal percentages I chose are somewhat arbitrary and need to be improved. I'm working out how to best put the scoring algorithm to a vote, and I will post that here later today. -Scottywong| chat _ 16:00, 4 October 2012 (UTC)

@Scottywong: Have you considered writing an AI-based tool that doesn't start with preset criteria, but instead compares stats it finds for current admins, failed admins, and normal users, and uses this to determine what criteria are relevant, and in what proportions? Even better, it could compare stats of current admins as of the time they made their successful run for adminship. (I've heard that being an active admin tends to change one's editting pattern, leading to an increase in non-mainspace edits.) The most interesting thing about such a tool, to me, would be getting insight into what matters in practice, as compared to in theory. Kobnach (talk) 17:33, 4 October 2012 (UTC)

That would be great, but also far beyond my capabilities. -Scottywong| talk _ 18:06, 4 October 2012 (UTC)

Well, I know nothing about wikipedia bots, but have just started an AI class. Perhaps I'll learn. Kobnach (talk) 15:46, 5 October 2012 (UTC)

Admin scoring workshop

It has become clear that my preliminary scoring algorithm doesn't do a great job with representing an actual candidate's readiness for RfA, for a variety of reasons. For that reason, I would like to get additional input from anyone who is interested, for the purpose of refining the algorithm. So, I have started a subpage at User talk:Scottywong/Admin scoring workshop. In it, there are a bunch of tables to which you can add rows indicating your own preferences for how important each criteria is in determining whether a candidate is ready for RfA. Please follow the instructions on the subpage for voting, and please only vote once. Thanks a lot for taking the time to improve the tool; hopefully something useful will come of it. -Scottywong| confer _ 20:47, 4 October 2012 (UTC)

Scotty, can you make a table of all the gathered data for every RfA, or the largest random selection you can make? As it would have looked at the time of the RfA? there's a friend of mine I'm sure would enjoy making a machine learning algorithm that would give you the ideal weights, insofar as predicting the outcomes of previous RfAs. Someguy1221 (talk) 00:10, 5 October 2012 (UTC)

Dear Scotty, I think this could be a very useful tool for RfA but I think trying to yield a score is pointless. Presenting relevant metrics is one thing but trying to rate them with points is doomed to failure. The userpage issue is classic. Userpage length is a poor indicator of admin qualities. Some editors believe a short and unostentatious one is better than a big one, you rate it the other way! Who is right? Is anyone right? If this tool could just present various statistics in a neutral way without making necessarily subjective value judgements it would be much better IMO. Setting up a scoring system (even if decided by a committee just like a camel vs race-horse) will court continuous controversy and derision. So...just give us the numbers and easy ways to investigate them (links etc) rather than a score. Jschnur (talk) 00:26, 5 October 2012 (UTC)

I agree about the raw numbers over analysis. If it will include analyses, the weights should be decided by the community, not one person, and that would be a never ending discussion. Eau (talk) 00:56, 5 October 2012 (UTC)

Aye, what Jschnur said. Rating is subjective and often meaningless, but the stats themselves could certainly be of use especially if they serve to encourage folks to look more closely at anything that seems... odd. Looking into such things sooner rather than later can often avert considerable potential drama down the road. -— Isarra ༆ 03:03, 5 October 2012 (UTC)

I think it's a fundamentally a bad idea to try and make assessing candidates easier via tools like this. Adminship is ultimately a question of trust and competence, which are basically impossible to algorithmically quantify. These stats mostly measure activity, and if people adopt this tool as part of their analyses of candidates, the stats calculated by the tool would be given more weight, by voters and hopefuls alike, than they should be (in my opinion). Other editor analysis tools have already seen such problems, but this one makes it so quick and easy and has such an appearance of comprehensiveness that it may have a greater negative impact than any other. wctaiwan (talk) 05:16, 5 October 2012 (UTC)

Used properly, I think the tool has some merit. Obviously it's not going to work as a selection tool but it would make a nice screening tool. There are plenty of editors out there who would make good admins but just don't have the social network aspect in place to get nominated. Or, when you see someone you think would make a good admin, you could run the tool and use the results to see if it is worth nominating the editor. --regentspark (comment) 15:53, 5 October 2012 (UTC)

There is no used properly - the numbers are essentially meaningless by themselves, and adding them up is no more meaningful. Using such numbers for screening would only serve to rule out those who would be excellent and give hope to those who haven't the clue, the same clue that is necessary to determine if someone is worth nominating and the same clue that has little to nothing to do with numbers. -— Isarra ༆ 23:04, 5 October 2012 (UTC)

Anyone expressing the view that this is a great, usable idea and who goes for RFA in the next 6 months will get a straight oppose from me on the grounds of having a complete lack of judgement & no common sense. It is a potentially confusing, divisive and almost worthless utility. I suggest leaning back in your chair, feet on the desk, crumple it up and try to spin it into the waste basket. Don't worry if you miss, the cleaners will sweep it away. No offence intended. 914 Leaky Caldron 16:05, 5 October 2012 (UTC)

Well, I tried the relatively face saving way, and that was rejected, so here goes. I'm watchlisting Wikipedia:Requests_for_adminship/Leaky_caldron just so I remember to oppose that nomination if it ever comes to pass, barring a personality transplant. You made an egregiously dickish comment towards a fellow contributor who has done nothing to you, but try to improve in his or her own way, the encyclopedia. You don't like it? Fine, neither do I, I always look at behavior and attitude, and find the tool (any tool, really) cannot measure those things. See how easy it was to neither condescend nor wave my metaphorical dick around? I'm all for the traditional polite note on a talk page normally, but there was no excuse for your comment. You should apologize, but I'm almost certain you will not. So I'll have to settle for publicly calling you out on your utter failure to live to Wikipedia standards (check out WP:NPA, WP:CIV, WP:BATTLE for starters) and basic human decency, in hopes that others may learn.--Tznkai (talk) 16:20, 5 October 2012 (UTC)

I can guarantee that the redlink will remain for ever red. Given that personal attacks and gross incivility are now effectively condoned by many Admins and remain un-actioned by our toothless Arbcom I think (with regret) that a re-write of those policies and guidelines will be required. The only thing that concerns me is that we have Admins (existing and new) that know WTF they are doing. If Scotty tells me on my TP that he prefers my opinion to be expressed differently I will do so. It's certainly not for you to remove any of it without first asking me per WP:TPO. You're an Admin and should know our guidelines better. Moving forward, anyone has my consent to hat or remove entirely this paragraph, preferably along with your polemic which is equally unrelated to the main subject of this section. Leaky Caldron 17:18, 5 October 2012 (UTC)

Incivility doesn't require rude language, just disrespect. (As a rule of thumb, if you say "no offense intended," you're probably being disrespectful but trying to avoid accountability for it) You showed another contributor great disrespect with no reason. Our policies, including the ones on personal attacks and incivility that you complain are underenforced are firmly against your behavior. One of the classic tools administrators used was to remove them, which of course is obliquely mentioned in the very guideline you cite. We could get into a dueling policies, but that is only slightly more productive than silly buggers. You disrespected another contributor, and there was no excuse for it.--Tznkai (talk) 18:41, 5 October 2012 (UTC)

Please tone it down. Such responses won't help anything, and just because one editor was less than civil is no reason to do the same yourself, especially since that just tends to aggravate things. -— Isarra ༆ 23:04, 5 October 2012 (UTC)

Wondering

Ok, I got a result of 999. (After over 6 years of having an account, I'm only 1 point away : )

I'm wondering about the various values. Some of the numbers feel a bit arbitrary. (But as I don't know how they are computed, I can't tell : ) - jc37 23:16, 5 October 2012 (UTC)

I got 1300-something yesterday, if I recall - which seems too high, being honest. GiantSnowman 23:38, 5 October 2012 (UTC)

I got 1364 but considering I'm by far and away one of the worst candidates for an RFA these days, I'm pretty pleased.... The Rambling Man (talk) 19:28, 7 October 2012 (UTC)

I scored 844 -- Boing! said Zebedee (talk) 15:14, 8 October 2012 (UTC)

1033. I think it likes my article creation :-) But someone I have my eye on as a potential candidate scores just under 1000. Now why does it rate you so low?? Yngvadottir (talk) 18:12, 8 October 2012 (UTC)

Dunno - I didn't keep the details, and I can't really be bothered running it again. -- Boing! said Zebedee (talk) 08:03, 9 October 2012 (UTC)

529? That's unexpected. I think it's mostly because of my article work. :-) Double sharp (talk) 06:09, 10 October 2012 (UTC)

For clarification, I'm not an admin, and I don't intend to request to become one for some time. Double sharp (talk) 07:49, 10 October 2012 (UTC)

Arbitrary Break

I also have an issue with deciding that RFPP, AFD, and AIV are the only "Admin areas" to use. Especially given that the other XfD and other AI noticeboards are just absent, (not to mention PERM). There's also an issue that the only "Front End" content areas that the metrics use are Articles. Portals, Templates, and Categories (which are all seen by readers, just like Articles) are so terribly underrepresented. It's an issue that comes up all over the project (that Articles are deemed "more important" than other Front End contributions), and one it would be wise to no longer propogate. Achowat (talk) 14:35, 16 October 2012 (UTC)

The tool is not intended to drive users to participate in under-staffed areas, it is simply meant to determine their suitability for adminship. It is not perfect by a long shot, and you might be right that AfD, AIV, and RFPP are not the only "admin areas" that RfA voters care about. However, the tool already takes a long time to complete its search, and adding searches for other pages will only make it longer. -Scottywong| communicate _ 14:48, 16 October 2012 (UTC)

No tool can determine anyone's suitability for Admin. Leaky Caldron 14:56, 16 October 2012 (UTC)

Your point has been made ad nauseum. I understand and recognize your opinion of the tool, and frankly, I don't care about it. I'm not abandoning the tool because one very vocal, and somewhat obnoxious user doesn't like it. If you have anything constructive to add to the conversation, please feel free. Otherwise, please find better uses for your time, and please don't waste mine. -Scottywong| converse _ 16:04, 16 October 2012 (UTC)

Attitudes like yours are exactly why RfA has the reputation it does. Instead of giving constructive criticism about something you don't like (or just not commenting on it at all), you feel compelled to post snarky comments like the one above in as many places as possible, presumably to draw attention to yourself. This is analogous to RfA voters who lack the ability to give a candidate constructive criticism when they oppose them, instead opting to publicly defame the candidate and point out exactly how unqualified/stupid the candidate is, presumably to draw attention to themselves. Moral of the story: just because you oppose someone or something doesn't mean you have to be an asshole when you express that opposition. -Scottywong| spout _ 16:11, 16 October 2012 (UTC)

(edit conflict)Do not claim that your tool has any sort of quasi-official role in determining Admin. suitability. It is at best a toy and will deter many suitable candidates as well as identifying wholly inadequate ones. It is inaccurate and gameable and will set false expectations - none of which concerns you seem willing to address which is why I will continue to criticise it until I receive an acceptable response. I would also appreciate removal of your very direct personal attack. Leaky Caldron 16:22, 16 October 2012 (UTC)

(I hope you're not misreading and thinking that he call you an "asshole" ... that would take some bizarre parsing of the sentence) dangerouspanda 16:50, 16 October 2012 (UTC)

I don't agree. The way I parse that sentence, he clearly implies that Leaky Cauldron is being an asshole whilst expressing his opinion. What's more, in the previous comment he's stated that "one vocal, somewhat obnoxious user" clearly referring to Leaky Cauldron. I'm awaiting Scotty's comment on the matter at his talk page. Worm^TT(talk) 16:55, 16 October 2012 (UTC)

Scotty's comments above are an example of why rank-and-file editors find lifetime admin appointments a high-risk proposition. Until editors see surer ways to redress abusive behaviour, RfAs will continue to be hard to pass. --A. B. ^{(talk • contribs • global count)} 17:01, 16 October 2012 (UTC)

FWIW I read the section the same way WTT did. Using "you" clearly personalizes it. Intothatdarkness 17:05, 16 October 2012 (UTC)

Maybe it's my linguistic background that read "you" meaning "one" ... such as in "if you can't do the time, don't do the crime" certainly is not personalizing it - it's a generic statement. I can now see how some could have read it otherwise, but it's still a bit of a stretch in my mind. Nevertheless, it's been struck and clarified as to the intent dangerouspanda 17:22, 16 October 2012 (UTC)

Maybe when you've just been described as obnoxious 7 minutes earlier it is entirely reasonable to assume that the subsequent arsehole reference is also aimed at the same target. Leaky Caldron 17:32, 16 October 2012 (UTC)

• To be clear, rather than rely on various sentence parsing strategies, I did not intend to call Leaky an asshole. In any case, I have struck the comment above to avoid any possible misinterpretations. -Scottywong| soliloquize _ 17:17, 16 October 2012 (UTC)

And where exactly did I claim that my tool has any quasi-official role in determining admin suitability? I have actually avoided that at all costs, and posted multiple warnings on the landing page for the tool to that effect. -Scottywong| talk _ 17:19, 16 October 2012 (UTC)

The very fact that you are referring to it at all in RfA comments gives your toy - Leaky was correct in that assessment - an air of credibility which it does not deserve. Please cease doing so forthwith. — Hex (❝?!❞) 18:05, 16 October 2012 (UTC)

No. -Scottywong| spill the beans _ 19:14, 16 October 2012 (UTC)

In this discussion you've clearly demonstrated the conversational skills that earned you a slew of oppose votes when you ran for administrator status. Well done. For your reference, I for one shall take pains to point out on any future RfAs where you refer to your creation that it is not reliable or useful, let alone official. I hope that others shall do the same. — Hex (❝?!❞) 19:20, 16 October 2012 (UTC)

That is your prerogative, however I would suggest that a more productive response would be to suggest improvements to the tool. -Scottywong| comment _ 19:38, 16 October 2012 (UTC)

I'm afraid that I feel the entire concept is a flawed one, sorry. — Hex (❝?!❞) 19:56, 16 October 2012 (UTC)

I have a "tool" that I use for World Cup Soccer pools. It's ugly, convoluted, and most people think it's stupid. It has, however, won me much money over the years. Perfect? Never. Useful in part? Yes. Just as nobody would build a house using only a hammer, one should never make your entire RFA decision on a single tool - and nobody is suggesting that they should. However, the more tools one has in their toolbox, we may get good choices in admins if used wisely and in conjunction with othersdangerouspanda 20:54, 16 October 2012 (UTC)

What might improve the tool and eliminate a lot of the acrimony here would be to simply remove the "score" part and leave it as a collator of information. People can then decide how much value to assign to all the numbers on their own. —Torchiest ^talk_edits 21:02, 16 October 2012 (UTC)

Good idea. If it doesn't set an artificial target/benchmark/comparator that can be misinterpreted by would be candidates or their nominators. Leaky Caldron 21:13, 16 October 2012 (UTC)

The tool does exactly that — for users that are currently at RfA. Otherwise, I disagree that the score is unhelpful. It seems that some of you approach this tool assuming that the average RfA voter is a complete dolt, and we need to build in babysitting measures to make sure these imbeciles don't use it incorrectly. I tend to have more faith in the community, and I think that the score is understood by most for what it is; simply a value judgment made by me (well, in reality, made by several users with whom I collaborated to refine the scoring). I have attempted to design the scoring algorithm such that the score values resemble the average judgments of RfA voters. If your values don't match the tools, then simply ignore the score and use the data. There has also been a constructive suggestion on my user talk to allow you to define your own scoring values. I'm interested in how that might be implemented, but not promising anything at this point. -Scottywong| confess _ 22:05, 16 October 2012 (UTC)

Someone already did a study of successful RfA's and compared them with various parameters and found that some characteristics had little impact, some had significant impact, and by their algorithm they could tell 80% of the time if someone was going to pass. Off the top of my head that is what I am remembering. So as an admin wannabe, I can participate only in that manner and tip the scales. Or I can check my number through the tool and find # of edits to Article namespace: 2211 +4 # of edits to Wikipedia namespace: 362 -28 - hmm how many thousands of edits does it take to get a positive score? But I would actually suggest shutting down the tool whenever there is an open RfA or RfB - it is pointless to have everyone just check the tool score and vote accordingly. Anyone can rack up whitespace edits just to get a better score. Apteva (talk) 04:36, 21 October 2012 (UTC)

Somewhat different proposal:

!Votes should not be accompanied by extensive commentary, but should have only a brief comment or two. Further discussion of any comments should not be made in a threaded discussion within the RfA, but on the talk page set aside for such an RfA. If any editors feel that extensive exposition on their !votes is needed, then links to pages in their userspace where their extended opinions are given is proper

Which I suggest answers the issue just at hand recently (the posting of excessive commentary). And also the unfortunate habit of some to engage in "extensive colloquy" on the RfA page itself. Collect (talk) 12:18, 9 October 2012 (UTC)

Now this is a proposal I absolutely can get behind. Well said. — Hex (❝?!❞) 14:09, 9 October 2012 (UTC)

Support Support of the broad concept, details might need fine-tuning)--SPhilbrick(Talk) 15:32, 9 October 2012 (UTC)

Oppose - Many editors cast their !vote in an Rfa in a drive-by manner, merely looking to see if someone they know is supporting and doing little independent investigation. An extensive commentary in the list of !votes is often the only place they will see serious objections. This proposal is, I assume, a reaction to the recent Rfa in which just such a comment, which effectively turned the tide of the !voting, was made. Bad idea to restrict such comments, and bad idea to have such a proposal when feathers are ruffled. This is a reactive appeal to disgruntled supporters of the rejected candidate, and should be removed as disruptive. Jusdafax 15:52, 9 October 2012 (UTC)

Huh? I !voted "oppose" on that RfA - this is not a "reaction" by a "disgruntled supporter" but a reasoned recognition of a real problem and an attempt to reach a rational solution to the problem. Collect (talk) 23:49, 9 October 2012 (UTC)

Support outlawing threaded discussion of votes on RfA and requiring them to take place on the talk page, but oppose length restrictions on anyone's vote. -Scottywong| squeal _ 16:04, 9 October 2012 (UTC)

• How brief is brief, though? Many comments can get somewhat long while still in the realm of what I'd consider reasonable, and plenty of on-page discussions also remain in the ream of relevant and productive. It's the ones that aren't that cause issue, and the only real way to pick those out is to use one's best judgement. Perhaps we should try kidnapping some of arbcom's clerks for the duration of these things; they're supposed to be good at that sort of thing. -— Isarra ༆ 16:23, 9 October 2012 (UTC)

  Three lines should be ebough to link to any serious cavils listed in userspace. Collect (talk) 23:47, 9 October 2012 (UTC)

  That would limit normal reasonable comment. Better people be required to use their brains - less definable and enforceable, but also less detrimental to the usual stuff. -— Isarra ༆ 17:35, 11 October 2012 (UTC)

• Part of what I am thinking about is to limit the comments on the actual nomination page that accompany the editors !vote to the minimum possible amount and have all lengthy discussion be taken to the talk page.--Amadscientist (talk) 19:04, 9 October 2012 (UTC)

Oppose as stated, but I'd Support a pure "no thread" proposal. RxS (talk) 03:24, 12 October 2012 (UTC)

• No threading in the ¡votes is something I've suggested. And ¡votes should be kept fairly short, although not severely. Say a hundred words and five diff/links. They want more? Link to all the noise they want on the talk. Links to userspace is a bad idea as there is not part of the RfA. All badgering to the talk page would go a long ways towards reining in the stupidity. Br'er Rabbit (talk) 03:45, 12 October 2012 (UTC)

Just pointing out that at the end of the day RFA is supposed to be a discussion on merits, and the whole support/oppose game we have now somehow came up as a side effect of how we discuss. A number of people including myself promised long ago that if the "!vote" ever became a "vote", we'd shut down this process. If we drop the last pretense, then I figure it's time to shut down for sure. --Kim Bruning (talk) 04:12, 21 October 2012 (UTC) 03:57, 21 October 2012 (UTC) I wonder if I could still find co-noms for the MFD though. I reckon everyone else who made the promise may have already left :-/

I am planning to launch an RfC for the future of this RfA. I am inviting all editors to create statements and proposals on what can be changed here at RfA. Please do not vote or endorse any statements as the RfC is not officially opened yet. The RfC can be found Wikipedia:Requests for adminship/RfC (new link: Wikipedia:Requests for comment/RfA, per comments in that section. mabdul 01:35, 21 October 2012 (UTC)).—cyberpower _Offline^{Trick or Treat} 18:48, 17 October 2012 (UTC)

The opening sentence is completely non-neutral and in article space would have ^{[citation needed]} and pov tags added. The result of the RFA was an unsuccessful candidate. Any other interpretation is your personal opinion and as such, I would contend, inappropriate for a RFC. Leaky Caldron 18:55, 17 October 2012 (UTC)

What we really don't need is for people to come in and say that this won't work and that RfA will never change. There is all this talk about how something needs to happen, an RfC is an opportunity to make it happen. But if we begin with an attitude that we are doomed to fail, we won't accomplish anything. AutomaticStrikeout 19:04, 17 October 2012 (UTC) To clarify, I was not to referring to any remarks that have been made yet in this section, I was remembering things I have read in the past that reflect a mentality that this is basically a lost cause. My apologies for any confusion. AutomaticStrikeout 19:13, 17 October 2012 (UTC)

Shouldn't it be at Wikipedia:Requests for comment/RfA? GiantSnowman 19:07, 17 October 2012 (UTC)

That would probably be a better place for it. Ryan Vesey 19:19, 17 October 2012 (UTC)

I think either place would be fine. Monty₈₄₅ 19:23, 17 October 2012 (UTC)

The current location indicates that a user called 'RfC' is standing for admin. GiantSnowman 19:27, 17 October 2012 (UTC)

I've moved it now, everything except the title is still exactly the same. Rcsprinter (speak) @ 21:44, 17 October 2012 (UTC)

The lack of neutrality Leaky caldron points out is concerning and should definitely be addressed before any of it becomes active. And while the RfA in question got rather ugly in some ways, the bulk of that was also largely not anything that could not have been resolved with some good clerking, something for which there is already provision in current policy and guideline. The more problematic issues with RfA, including the environment that led to that in the first place as well as the fact that next to nobody actually did anything to try to help the situation, are more general. This makes these issues both harder to pin down and also much more difficult to sensationalise, but they perhaps are what would merit a focus of any further extended discussion. -— Isarra ༆ 19:49, 17 October 2012 (UTC)

There is a reason why I didn't launch the RfA yet. This community input is great. The RfC is far from ready to go live but it's getting there. I just need community in order to fuel it and get it going. I will address the concerns stated before it goes live.—cyberpower _Online^{Trick or Treat} 01:07, 18 October 2012 (UTC)

Many RfC fail because the proposal statement is either too vague, or allows for too much side tracking and alternative proposals within the RfC. Any RfC pertaining to reforms of the RfA process or of the duties and responsibilities of admins themselves is best tailored to one single aspect. There have be many RfAs, both successful and unsuccessful, that have been a platform for incivility, personal attacks, and disingenuous voting, so it's unwise to base an RfC on one single RfA. There have been many RfC proposing change to adminship and its (s)election process, all have failed to reach a consensus. A thorough review of former proposals is preferable before making proposals for change, and probably one of the most comprehensive collections of research and discussion that can be drawn on is here. Kudpung กุดผึ้ง (talk) 02:25, 18 October 2012 (UTC)

I find using Sigma's RfA as a premise for the RfC very ill-advised. At best it would be a distraction, at worst it may change the focus of the RfC from fixing RfA to rehashing the arguments whether Sigma's RfA was unfairly derailed / uncivil / whatever. Really, I think the way it is done, starting an RfC at this point wouldn't help with anything. At the very least, it needs to be more focused and more editors representing each major view (RfAs are uncivil, opposers are unfairly badgered, RfA is working just fine, etc.) need to be involved in the structuring. Even then I'm sceptical it'll help. wctaiwan (talk) 02:36, 18 October 2012 (UTC)

I was basing this RfC off of Sigma. It was merely a motivational push. I had already announced that I plan to launch an RfC for this months prior to his RfA. As I said, I will be doing some more development towards this RfC before I let it go live. I will remove references to Sigma in that RfC. — Preceding unsigned comment added by cyberpower678 (talk • contribs) 02:56, 18 October 2012 (UTC)

I've got to say, this RfC is set up to fail, and is almost exactly how I would have set one up after reading about how RfCs work, but not actually looking at how RfCs work. There are too many options, which may work for soemthing like ArbCom, who can go away as a small group and decide which ones are the best to use, but not here. Allowing new proposals will mean that lots of little ideas which will be supported by 2 or 3 people will turn up. Having a section for a pure numerical vote will lead to "voting is evil" comments - and how many people are going to say "no, RfA is not broken at all". The question is "how is it broken?" not "is it broken?" Overall, I'm sorry to say that not only is this not going to work, it will perpetuate the issue. Worm^TT(talk) 05:32, 18 October 2012 (UTC)

Hmm, well my plan wasn't to run just one RfC to address and fix the issues here at RfA. Perhaps the proposals are too premature right now and should be placed in a different RfC. I am thinking this entire thing through. I am alright if it fails, but I'm going to devote my time, efforts, and community inputs from you guys to make this hopefully, the most effective RfC there is.—cyberpower _Offline^{Trick or Treat} 14:11, 18 October 2012 (UTC)

It seems that there are discussions about RFA everywhere; here, at Wikipedia:Village_pump_(policy), RFA reform, [Signpost] and various other locations. I cannot see a single RFA here capturing and maintaining enough of the community's attention to arrive at decisive outcomes. Leaky Caldron 14:22, 18 October 2012 (UTC)

RfA is broken has been a running theme whenever we get the sort of drama that happened with the recent one mentioned at the top of the RfC proposal. Anyone remember ecoleetage_3? Always worth keeping in mind that there are two processes driving an RfA. The first process is the social networking one. Well recognized editors with plenty of virtual friends start strong with lots of positives. The second process is the one in which the candidate is really examined. That's when you start seeing the negatives and, particularly when the social network aspect is strong as it was with Ecoleetage, the RfA tends to fall apart dramatically and it appears that the process is derailed. But, in reality, the process is doing exactly what it is supposed to do and doing it quite well. --regentspark (comment) 14:30, 18 October 2012 (UTC)

Ok, just taking part of this argument a few steps further. Just as WP:AFD is really supposed to be thought of as "articles for discussion", that might lead to deletion, here's a concept that combines a few things: RFA becomes two steps: "Proposed Administrator For Discussion", followed by (*gasp*) a vote after the discussion period closes.

Imagine the following format for a PAFD:

Nomination

This section would be pretty much as it stands today

Standard Questions for candidate

This section would consist of the standard questions only. Consensus could be reached to add/subtract/amend those standard questions prior to including on an PAFD in order to set the current default set of questions.

Community questions and concerns

This section would be numbered, just like out supplemental questions are now. Each question/concern signed by the raiser - just like we currently do. IMHO, each editor should be limited to one question or concern. The candidate being discussed would have the opportunity to respond/retort, but continuation of the thread would not be permitted. This permits the candidate to provide a diff or explanation.

After a certain number of days (add: probably 7), the discussion is closed, and the candidate can determine whether or not they wish to pursue the actual vote phase. If not, the PAFD is archived, and the candidate has a great list of things to work on for next time.

If yes, a voting page is setup that only says "I have read the linked PAFD discussion, and: __ Support __ Oppose __ Neutral", the editor signs their vote, and no further discussion or comment is permitted - add: I would assume the vote portion would also last for 7 days.

The RFA page would originally only transclude the discussion page until the candidate chooses to run the formal vote, at which point vote page is transcluded beneath it.

This therefore allows:

• pre-vetting of the candidate (almost like editor review)
• the candidate gets the chance to respond to negative/ancient concerns
• the community can discuss in a reasonably fair manner
• nobody is allowed to vote until discussion is complete.

Yes, it forces those who desire to vote to come back twice - but that's not a bad thing. (✉→BWilkins←✎) 15:02, 21 October 2012 (UTC)

Partly covers what I have expressed elsewhere, that voting should effectively be on hold until all standard and some additional Q's have been answered. Prevents early pile-on supports by friends. Also, limiting threading is good. The rest needs a bit more thinking about. Good ideas worth further discussion. Leaky Caldron 15:15, 21 October 2012 (UTC)

I support the concept, but more details need to worked out. 7 days would not be sufficient for this process. --Anbu121 (talk me) 15:45, 21 October 2012 (UTC)

• I would suggest adopting a policy of not opening the voting on RfA, RfB's for three days, and having any votes deleted that are cast before the voting is announced being open, and no other changes. No formal PRfA. PAFD sounds too much like preliminary article for deletion. Apteva (talk) 18:40, 21 October 2012 (UTC)
• Perhaps an editor review should happen first, with a transclusion at RFA. Graeme Bartlett (talk) 20:52, 21 October 2012 (UTC)

• So you're suggesting requiring an editor review? I think it sounds like a good idea in practice... but the truth of the matter is, so many people go for ERs that getting a review (for someone who isn't a "popular" Wikipedian) is nearly impossible... and even if you do, it's usually a) from your wiki-friends or b) a one line "Awesome job!/You suck!". If we were to work on increasing the actual usage of ER, though, then I think this has some potential... but. Well. I don't know. Theopolisme 01:01, 22 October 2012 (UTC)

• I'd like to see something like this. It reminds me of how we vote for arbitrators, a position of much higher authority, where we have separated discussion and voting (even when we did not have secret ballots). —Kusma (t·c) 07:38, 22 October 2012 (UTC)
• I've repeatedly suggested having a discussion phase before there is any voting. I'm not sure I like all the restrictions suggested here, but as long as there are no restrictions on discussion in the voting phase, it can't make things any worse. And it can always be tweaked once the principle is established. Rd232 ^talk 11:51, 22 October 2012 (UTC)
• Beware of how this backfired/was abused when it was tried once before: Wikipedia:Requests for adminship/Ironholds 2 Not to say it can't work, but in that RfA, the "ammo" that doomed the RfA was saved until the RfA went live.208.254.130.23 (talk) 16:22, 22 October 2012 (UTC)
  • That RFA failed because of concerns about incivility and a sort of hypocrisy about one action (doing an experimental RFA having criticised others for doing it). Those are basically issues that can't be greatly illuminated via Q&A with the candidate. If you force Step1 to be basically Q&A, then everything that doesn't fit into that will be left for later. There's a lesson there, but it's not "we'd better stick with the status quo, because we can't possibly do any better". Rd232 ^talk 11:04, 23 October 2012 (UTC)
• Well, I think it sounds more than reasonable except, do we really need the change in name or a two part section/review? Couldn't we use these ideas for the current RFA page without a complete change in formating and something that might prolong this more than is needed. How about just keep the same format for nomination and standard questions and we already allow questions by the community so we don't really need to make a seperation, but:
  • Amend the standard questions with consensus discussion.
  • Voting opens three days after nomination is opened.(This could be bot operated to show a change in status to "open" in voting)
  • One question limit with one answer no follow up.
  • No thread making, in the questions or the !voting section if another editor has a follow up question it should be made seperate.
  • Any and all off topic discussion should go on the talk page only if needed.

I could support something like this without having an actual review.--Amadscientist (talk) 09:13, 23 October 2012 (UTC)

I originally posted this at Wikipedia talk:Requests for adminship/Σ#Attempts of the community to fix the Requests for Adminship process but I don't think it got much attention. It's a slightly different take on our RfA drought:

I've been participating in RfAs for six years. I went through 2 RfAs myself - one here and one on Meta:

• Wikipedia:Requests for adminship/A. B.
• m:Meta:Requests for adminship/A. B. (note: I'm no longer an admin there due to inactivity on Meta)

I've seen a lot of RfAs. I've been an admin myself for almost 5 years.

Unfortunately, I've seen too many problematic administrators over the years. I'm sorry to hurt folks' feelings, but websites like wikipediarevview and wikipediocracy exist for a reason, whether or not you like the sites' behavioural norms or some of their participants.

My observations:

• Community standards for admins have increased in reaction to admin abuse over the years.
• Some of our most problematic admins have been approved by bureaucrats after RfAs with very weak support.
• Problematic admins are very disruptive and hard to remove. One troublesome admin can create enough drama to tie up other admins and editors for hours with each incident.
• The community has become increasingly skittish of approving administrators who might cause problems: "better safe than sorry".
• Administrator quality has risen since 2006. One informal measure is the reduced activity at wikipediocracy and wikipediareview.
• Admin selection by ArbCom or 10 other admins will just lead to charges of cronyism and erode editor confidence in their administrators.

If folks are really worried about declining admin numbers, make it easier for the community to remove problem admins; our editors, in turn, will be more willing to take a chance on RfA candidates. Better yet, let admins automatically stand for reapproval every 1-2 years.

We must never forget that our editors serve our readers and our administrators should serve our editors -- not the other way around.

(Note: I've cut some stuff from my original post that was germane only to that particular RfA).

Without getting into political debates over employment policies in different nations, I'll note that many economists have said that restrictions on employers' personnel actions lead to higher levels of unemployment; supposedly employers are more reluctant to hire employees if they're might be more expensive and difficult to fire. Whether or not that's the case in the labor market, I think there's a similar mechanism at play on en.wikipedia.

I'm no longer active on Meta, but Meta used to have a procedure where every admin was listed annually for reapproval. As long as some percentage (90%?) of commenters were positive about their continued service as admins, admins kept their adminship. That was usually the case. As a result, it was easier to become an admin on Meta. (They also had a policy, inapplicable here, that any Meta admin had to already be an admin on their home Wikimedia project). I don't know if Meta still has this procedure in place, but I always thought it was a great idea. It lowered the risk for the community associated with RfAs and it kept admins accountable for their actions once approved. Meta's RfAs were also more collegial in tone, in part because of the culture of that project but also because these RfAs were less frought with worry.

I strongly believe that the community must continue picking admins
--A. B. ^{(talk • contribs • global count)} 12:31, 11 October 2012 (UTC)

A problem with the "90% reapproval" thing is that some of our best admins are the ones who aren't scared to get stuck into difficult and contentious areas, and they rack up enemies who would jump at the chance to vote them down - I don't think Meta is remotely close to en.Wikipedia in that regard. As an example, a contentious area I work in as an admin is the Indian caste system. It is plagued by exceptionally nasty caste warriors and nationalist POV-pushers - the kind who actually track people down and issue death threats. (There are lots of good people too, I hasten to add, and some in between). A reapproval run for me would get a good dozen opposes just from editors I know, and there is organisation off-wiki that could easily gather 100 or more. Imagine the potential problems facing admins active on Israel/Palestine topics, or other problematic areas. We have to make sure we don't create a generation of admins who are too scared to do the hard stuff. (But then, this is just the same old same old being hashed out yet again) -- Boing! said Zebedee (talk) 12:48, 11 October 2012 (UTC)

Another problem with the "90% reapproval" thing is that it wouldn't necessarily deal with 'problem admins' - it could probably also see the removal of perfectly good admins who aren't seen as being absolutely perfect. GiantSnowman 13:06, 11 October 2012 (UTC)

Reconfirmation would be very labour-intensive and create more space for more bickering. We can do without this. I do agree about there needing to be more fluidity in adminning/de-adminning though. Casliber (talk · contribs) 13:16, 11 October 2012 (UTC)

I'm not necessarily arguing for a 90% threshold if we adopt a reconfirmation process. I also don't even think we need a reconfirmation process if we have some other easy mechanism for the broader community of rank-and-file editors to remove problematic admins. As for the Indian caste article example, approaches that could be adopted for any removal or reconfirmation process could include either or both of:

• A minimum number of commenters required such that a handful of nay-sayers couldn't overwhelm a discussion with few total commenters.
• Minimum edit counts and account longevities to filter out newly created sock or meat-puppets.

--A. B. ^{(talk • contribs • global count)} 13:28, 11 October 2012 (UTC)

We've uncovered dozens of sleepers already, and I don't doubt there are many more. Tweaking of the percentage pass rate and criteria for !voting just don't have the resolution needed to keep out all the baddies and allow in all the goodies. Whatever, say, the pass percentage, there will be good admins who fall below it and bad ones who fall above it. -- Boing! said Zebedee (talk) 13:36, 11 October 2012 (UTC)

• The quality of our administrators has certainly increased substantially, but we've also denied granting the bit to literally hundreds of editors who would have put it to very good use, oftentimes for the most trivial reasons. The results of that are a diminished efficiency in dealing with user conduct issues, unblock requests, and several other key areas. Kurtis (talk) 04:06, 13 October 2012 (UTC)

I'm not sure that it has denied access to the tools to 'literaly hundreds' of editors. Close analysis of 100s of failed RfAs doesn't seem to bear that out. In fact the data seems to suggest that RfA generally works quite well in that it passes those who should, and fails any for whom there are reasonable doubts - many of whom actually pass at a future attempt. However, except in the case of very close calls, at the end of the day, it's a community decision. There are concerns that the process is broken, but in spite of efforts to reform the system, which also include numerous proposals to unbundle the tools, no significant changes to RfA have ever been accomplished. Lowering the bar to adminship is probably not the answer as there are many who contend that many admins are incompetent already. The solution is to find a way to encourage more editors of the right calibre to run for office. Kudpung กุดผึ้ง (talk) 05:15, 13 October 2012 (UTC)

Most administrators (over 1,000 of them) were given administrative privileges for life before 2008. That was the period where children needed to do little more than ask to be promoted. The admin system can claim no credibility so long as admins have been appointed in this way for life. Just because the last handful of admins to be appointed are better qualified doesn't go anywhere near to addressing the issue. It is becoming ever more apparent that reform must happen. The fact that efforts to reform the system have been suppressed is no reason to think that reform is not going to happen. --Epipelagic (talk) 05:58, 13 October 2012 (UTC)

I agree that reform is needed and that it will happen one way or another (and one of my beefs is the attitude of some of those early admins who seem to think they have carte blanche to make up rules as they go along and that their decisions cannot be questioned), but I'm not really sure how such efforts have been "suppressed" - looking back over the past years of this talk page, it looks to me as if they've simply not come close to achieving consensus. (And it really doesn't look like it's been admins who have stymied them - the wider community does seem to have been very divided on every suggested reform that I've seen). -- Boing! said Zebedee (talk) 06:08, 13 October 2012 (UTC)

That's because you haven't factored in the admin wannabees, who are often the most vocal and persistent opponents of change. They often allow admins to stand aside while they do the work. --Epipelagic (talk) 06:14, 13 October 2012 (UTC)

Classing people who disagree with you as "admin wannabees" is not a good faith approach to discussion. But it does illustrate one of the ways that such discussions deteriorate and fail to get anywhere - the approach of turning it personal and insulting people. When you stick such labels on your opponents ("Because they disagree with me, they must be X"), you lose objectivity and damage your own credibility. When we work by community consensus, community consensus is the only thing we have, and when you abuse that by slagging off the opposition, you become part of the problem. I've always said that the Community itself is the problem, and you've just helped to reinforce my opinion. -- Boing! said Zebedee (talk) 06:25, 13 October 2012 (UTC)

I'm talking about editors who say they want to be administrators on their user pages. How is that "slagging off" anyone? That is a matter that can be tested objectively. Who is making the personal attacks here? --Epipelagic (talk) 06:33, 13 October 2012 (UTC)

OK, fair enough, you have my apologies - the approach I described is one I have seen a lot, and I'm sorry if I misunderstood you as being part of it. OK, you say it can be tested objectively, so have you done so? Have you counted the editors in favour of and against any specific proposal, and can you give us the actual numbers? I really would be interested if you can find a proposal where one side has been predominantly admins and "wannabes" while the other side has been predominantly not. If you can provide evidence for your assertion, I think that could be a very useful contribution. -- Boing! said Zebedee (talk) 06:43, 13 October 2012 (UTC)

Alright. That'll need a bit of time, and I can't do it tonight. But I'll get back to you on this. --Epipelagic (talk) 06:48, 13 October 2012 (UTC)

Sure, it's not a trivial task, but there's no urgency - look forward to the result -- Boing! said Zebedee (talk) 07:01, 13 October 2012 (UTC)

Here is a current proposal

	admin	non admin	total
support	0	2	2
oppose	4	1	5

--Epipelagic (talk) 20:50, 15 October 2012 (UTC)

And here is another current proposal

	admin/wannabee	non admin	total
support	14	38	52
oppose	11	10	21

--Epipelagic (talk) 08:56, 25 October 2012 (UTC)

The wannabes who say they want to be administrators on their user pages actually do themselves a disservice. Although there are rare exceptions, most of them won't be ready for adminship for years - especially the ones who add such userboxes to their userpages almost as soon as they register. Those who have been around a long time and who do great content work but who are not admins and exploit every opportunity to throw personal abuse at who have the bit, need to examine their social skills to understand why they are not admins. By disrupting genuine attempts to get a better class of admin on board*, both these kinds of users damage not only their own credibility in discussions, but the performance and credibility of Wikipedia as an encyclopedia. Most of the truly active admins are those who were given the bit since 2008 - since the invisible bar was raised. In my own very clear experience, most the few admins who misbehave are the pre 2008 ones. Kudpung กุดผึ้ง (talk) 08:13, 13 October 2012 (UTC)

Kudpung, specifically who are these people you refer to who "exploit every opportunity to throw personal abuse at who have the bit" and need to "understand why they are not admins"? --Epipelagic (talk) 08:51, 13 October 2012 (UTC)

I am probably at least one of the editors he's talking about but I am not "abusing" anyone. If I have been critical of some admins that's because they have done something to deserve it. Like make statements like the one above that make adminship seem like some lofty achievement that few deserve. Its a few extra tools, that's it and there is absolutely no reason other than vanity to paint it as anything else. We need to go back to the mentality that adminship is no big deal, make it easier to get, easier to take away and take the process back to the gentle one we had back in the 2007 days. I genuinely believe Kudpung means well and his heart is in the right place and I truly believe he means well for the pedia but the raised bar he is referring too is part of the reason RFA is now a battleground and near defunct. Kumioko (talk) 14:19, 13 October 2012 (UTC)

I suppose the high bar was introduced to keep the very admins out that Epipelagic is always complaining about. What would happen if the bar were significantly lowered and 100s more of them were let in. Kudpung กุดผึ้ง (talk) 15:18, 13 October 2012 (UTC)

• Fancy blaming me for the high bar you are promoting Kudpung! You also avoided answering my question. There is a clear implication in the context of your comment further above that I "exploit every opportunity to throw personal abuse" at admins, apparently because I don't "understand" why I am "not an admin". If that was the message you intended to give, then it is overdue for you to note that that I receive far more abuse than I ever dish out. Your persistent misrepresentations go too far. What possesses you to think I want to be an admin? I want to develop content. That's why I'm here. I've diverted to addressing admin issues because it has become too demeaning to contribute content on Wikipedia. I am totally committed to developing the best encyclopedia possible, and equally committed to enabling a workable admin system. I do have an agenda. It is to work towards an environment where content developers can work with dignity and some sense that natural justice prevails. Equally, people who work on the admin side should be able to work with dignity, and be better empowered to do what needs to be done. But people who work on the admin side should not be elevated to some special status. What is your agenda Kudpung? --Epipelagic (talk) 19:16, 13 October 2012 (UTC)

I don't think anyone is saying the bar should be removed completely, but it should at least be achievable without devoting "years" to the goal of becoming an admin. I also doubt that hundreds would come flocking but the travesty that occurred in the last couple weeks with Sigma should be avoided. Kumioko (talk) 15:22, 13 October 2012 (UTC)

Travesty? Leaky Caldron 15:35, 13 October 2012 (UTC)

Yes and frankly I consider your oppose for the reasons of his user name to be absurd. IF you oppose because you think he doesn't have the right temperment, maturity, skill, etc. then that's fine but to oppose based on a username that Wikipedia technically allows is absurd and reflects the abuses and poor civility of the RFA process itself. Its these sorts of silly and insignificant opposes that sabotages the entire process. Kumioko (talk) 15:43, 13 October 2012 (UTC)

Then I suggest you re-read my oppose which consisted of 4 valid reasons. I also suggest you examine the 15 or so support votes who also expressed concern about his username as well as the many opposition !votes that didn't care for it. Nice attempt at assigning blame but you're wrong - and off topic. Leaky Caldron 15:49, 13 October 2012 (UTC)

FWIW, I primarily opposed Sigma for other reasons, but since you bring it up, the user name, while a small matter, was emblematic of a larger concern I had. Administrators need to be very readily accessible to all editors, including those without the arcane knowledge of how to generate a "Σ". I know I had to spend a minute or two to figure it out. It didn't bother me that Σ didn't realize it was an obstacle for many others but I was surprised that Σ didn't move to change it the minute the issue was pointed out to him/her. A good admin needs to have a "service mentality". Editors serve our readers; admins serve our editors. There are might even be work-arounds Σ could have used: alternate accounts redirecting to his main account (I'd have to think about how you might get that to work).

A certain part of being an admin includes surrendering some of your own likes and dislikes when acting in an admin capacity. How willing and able is a potential admin to do that? When the person has such strong beliefs that they vandalise other websites on ideological grounds, you look even more closely for signs they'll bend as necessary here to be good admins. An admin is supposed to represent the community's needs and values before their own.

Content is king. We really need good content-builders more than we need admins. We can never do too much to recognize our good content-builders. Sometimes I think admins get too much attention.
--A. B. ^{(talk • contribs • global count)} 16:54, 13 October 2012 (UTC)

PS I became an admin before 2008 -- I hope I don't misbehave that much!

It's a shame a thoughtful opinion like your's isn't expressed more frequently by your colleagues. From the tone of it I wouldn't have realised that you were an Admin! Leaky Caldron 17:03, 13 October 2012 (UTC)

For what its worth I saw those other comments about the name and I believe those are invalid as well and I wish I would have taken the time to state as such on the RFA before it closed. I'm also sorry if I inferred you were the only one or that it was your only objection. If the software allows such names and since we do not have policies against them then we cannot and should not be holding it against users. If they are such a problem, which I am not convinced there is, then we need to change our policies to disallow and then we can stop the account from being created in the first place not when they editor tries to be an admin and then make it an issue.

Also in response to your comments A.B. I appreciate the thoughtful comment and I too was somewhat surprised at the notion you were an admin by that post. I do not agree though that an Admin need be a modal for all things Wiki. Its a set of tools and thats it. Any prestige we have given it is because we have put adminship on some pedestal and not because its some high achievement. These days it doesn't even have anything to do with ones knowledge or experience but is more of a popularity contest with those editors who are liked being promoted and those that deal in more controversial issues being passed over. No offense intended. AS for the notion that admins need be readily accessible I partially agree however of the approximately 700 active current admins over 70% Rarely use their admin tools and of the ones that are left its the same 20 or 30 that frequent the admin boards and actively do the admin related content work.

I deal a lot in areas which are admin related personally and I am getting tired of making work for other people because I have to go and ask some admin for help and then wait anywhere between several hours to a week for them to implement the changes I made, often times not even understanding fully the changes and wouldn't know if they were right or not. They do it because they trust that I did it right based on past experience of working together. Yet the greater community says I cannot be trusted to have the admin tools and I frankly find that insulting, frustrating and petty. Kumioko (talk) 19:49, 13 October 2012 (UTC)

If you had read the RFA you would also have noticed that as soon as the signature issue was raised one of the candidate's nominators stated that it had been raised pre-nomination. The candidate chose to do nothing about it. You seem to be suggesting that opposes not based on clear policy are somehow inferior and lead to issues at RFA ("oppose based on a username that Wikipedia technically allows is absurd and reflects the abuses and poor civility of the RFA process itself.") I'm afraid that I do not follow your rationale at all. Anyway, it ultimately had no bearing on the outcome, which was entirely due to the community making its feelings clear about the candidate's off-wiki activities. I have read your grievance about your own situation elsewhere. Are you suggesting that your recent RFA, closed 28/46/16, was unjust? Leaky Caldron 20:49, 13 October 2012 (UTC)

Yes I think in general opposes that are not based on policy should be limited and given less weight. I am also not overly concerned with his off wiki activities, I am concerned with his activities here on this Wiki and he has always seemed to be a positive contributor here. As for my RFA, yes, I do not think it was just. Most of the opposes were based on the fact that I was blocked, inappropriately by an admin that didn't bother to learn the full details of the situation and made an assumption based on a comment I made. They lept before they looked and set a course that ruined my reputation in Wiki and contributed to that failed RFA. Had I submitted it in December 2011 I probably would have passed with a large margin. I lost my temper and got very frustrated. But I have never and will never vandalize the project and other than some snide remarks I have always tried to be a positive benefit. Unfortunately there are a lot of folks with their own agendas and those agendas are not always in the best interest of the project like article ownership issues. Whats additionally unfortunate is that a lot of editors on here won't bother to learn the full truth of the situation and will just oppose because I got blocked. I knew the RFA process was broken long before I submitted (that's why I hadn't resubmitted since my initial submission in 2006) but that cemented in my opinion that the RFA process is broken and I have no faith in it. I would additionally state that anyone who thinks I would abuse the tools or intentionally do harm to the project is gravely mistaken. My contributions to the project reflect my belief in the project (not necessarily all the editors in it or their petty little battles) and I stand by my edit history. I haven't always been perfect but I have always tried to do whats best for the pedia even if that wasn't necessarily the popular thing to do at the time. Kumioko (talk) 21:09, 13 October 2012 (UTC)

Kumioko, I looked at your user page and your second RfA. Don't get discouraged -- as a prolific content contributor and article-starter, you are much more valuable to Wikipedia doing what you're doing. You're more valuable to Wikipedia than 90% of the admins on the project (myself included). Who comes to Wikipedia's site for the admins?? It's our content.

Observations:

• The majority of editors may or may not make good admins, based on their online temperaments. Adminship is a niche task here -- or it should be.
• There are two major sides to adminship -- content control and behavioural policing. The second is where the greatest potential for abuse is. If we unbundle the two aspects, the community might be more willing to take "risks" with content admins.
• Good "behavioural" admins should reduce friction for our editors, not cause it. Block vandals that so frustrate editors building articles, calm down warring editors, etc. If some think of admins as cops, then we need more Andy Griffith and less Dirty Harry.
• There needs to be an easier way to remove problematic admins. Leadership selection through crowdsourcing will never be perfect, even if it's the best of the alternatives (as I think it is).
• Admin attrition gets a lot of attention but I'm not sure it's that big a problem yet. I am much more interested in attracting and keeping good editors.

Thanks for creating all that content.

--A. B. ^{(talk • contribs • global count)} 22:35, 13 October 2012 (UTC)

I appreciate that but its hard not to get discouraged. If enough people tell you that you can't be trusted and or not wanted it starts to sour your attitude and although some have tried to persuade me that they do trust me that fact that I don't have the tools reflects otherwise. If I was trusted, then I would have passed my RFA its as simple as that. Especially as more and more content gets protected requiring admin intervention, as more and more admins leave and less are promoted to replace them and frankly as I see editors who have been here far less and know far less get the mop just because they keep to themselves in their little corner and don't meddle in contentious things. This tells editors not to do contentious tasks like work in WikiProjects, submit things for deletion, block vandals or a variety of other things which can and are contentious. Instead we promote those that keep a low profile and don't ruffle too many feathers. I agree that some aspects of Adminship can require a certain temperment that not all of us possess or want (I used too and did for years but none of it mattered) but there are too many tasks centered around the admin tools that need doing. Either we need more admins or we need to split some of those functions out into bundles which also seems unlikely. In the end Wikipedia is going to suffer because of a catch 22. Getting back to the point of this original thread though I still believe that its a set of tools like Twinkle or AWB. Nothing more, nothing less. We need to trust our editors and allow them to have the tools they need to do the job. Otherwise we need to change the job so those tools aren't needed. Kumioko (talk) 23:11, 13 October 2012 (UTC)

I recused myself from your RfA, so at least no one can blame me for what happened there. Don't get discouraged - take heart at Ironholds' RfAs (I think he probably holds the record). Back to the point: If some many people assume that our current admins can't be trusted to do a good job, what would be the metrics of confidence applied to those who would ask for the unbundled tools? And who would accord those tools and by what process? Kudpung กุดผึ้ง (talk) 04:41, 14 October 2012 (UTC)

In all honesty I wished more people would have voted on it either way. I will say this though, the relative low number of votes I received in contrast to the number of editors I work with and around says just as much about how I am looked upon as who did vote, regardless of how they voted. Kumioko (talk) 22:49, 14 October 2012 (UTC)

• Administrator quality has risen since 2006. One informal measure is the reduced activity at wikipediocracy and wikipediareview.

Decreased activity in these forums could be due to many reasons. It could e.g. be that many of these people had their own personal run-ins with the Wikipedia community and project, so most of them aren't in the discussion for the bigger picture. Others may lose interest because they recognize that outside forums are only used to vent, or that due to its structure, Wikipedia has a gigantic inertia against reform and will likely never substantially improve in overall quality. Many critics of Wikipedia's processes and community structure have reluctantly accepted that fact and don't even bother complaining anymore. Back in 2006 we still had high hopes for Wikipedia. --87.79.131.231 (talk) 15:09, 15 October 2012 (UTC)

I feel that admin quality has been dropping since - oh, a long time now. RFA requirements are now more seniority than skill based. Folks who have talked with me know that I've been saying this forever. They also know my arguments for why. I don't think admins have to be particularly skilled though, as long as they don't blow up the wiki. If just about everyone could get some appropriate flags for showing they can use them responsibly, it wouldn't end up being such a huge thing, right? I consider it to be like getting a drivers' license or pilot's license. A responsibility, sure, but it doesn't make you a ruling class. :-P

I don't think WP is unreformable. It just takes people who want to put time and work in mediation, process and teaching. These are all rather unglamorous pursuits in an already somewhat unglamorous project ;-).

I think the main thing I've seen in organizations that survive for 50-100 years or longer (and I've worked for some now! \o/ ) , is that they build in and maintain flexibility. That's what we need to aim for here.

Make it easy to get admin, easy to lose it if you really mess up. Make it easy to edit, but make it easy to stop editing if it doesn't work out between you and WP.

But most of all, make things fun. If it's a joyful experience, people will keep doing it forever. :-) --Kim Bruning (talk) 04:29, 21 October 2012 (UTC)

Hm. I believe the greatest problem is how to call people on their behavior. How does one "argue" against a side that has no argument and no case, that merely consists of bias, ideology and intellectual dishonesty? There is no mechanism for sorting this out, and consequently some of the most barefaced liars are deeply entrenched ("established") in Wikipedia's community. I have little to no hope that this will ever get better. The challenge we're facing is akin to outlawing Fox News. --87.79.226.156 (talk) 11:37, 21 October 2012 (UTC)

Anyone may call anyone on their behavior. As long as the good guys outnumber the bad guys, this tends to keep people honest in the long run. Are there particular people whom you feel should be called on their behavior, and haven't been so far? --Kim Bruning (talk) 15:40, 21 October 2012 (UTC)

• A related discussion is taking place at Wikipedia:Village pump (proposals)#New user right proposal.

Kudpung asked in the thread immediately above: "What would be the metrics of confidence applied to those who would ask for the unbundled tools? And who would accord those tools and by what process?" There are many ways that could happen. What follows is just a starting point, a scenario off the top of my head. There will be things I haven't thought of, and other ways of doing things that might be better, and better names than the ones I have used. And too much detail at this point would just be confusing.

• We could start by forming a project around the use of each tool. Some projects already exist that could be adapted for that purpose, such as WikiProject Deletion. A tool project would be the place where the community of users of a particular tool, as well as the community as a whole, can set guidelines governing the use of the tool, help new users and discipline remiss users. The tool projects would also have boards where users can apply for the use of a tool

• There might perhaps be a concept of something like a "user of good standing". A user of good standing could apply to a tool board for the use of a tool. Decisions would need to be made about what a "user of good standing" was, and individual tool projects would need to decide whether addition criteria must be met before a tool can be issued.

• A special project or board, separate from the individual tool projects, might oversee the mechanics of assigning and removing tools from users, and maintaining software that automated these tasks.

• Initially, all current admins would be assigned all the tools they currently have access to. That is, initially there would be no change at all in the way tools are assigned to admins. A given tool would be withdrawn from a user if, say, it hadn't been used for 12 months. That withdrawal could be automated by a bot that checks tool usage and automatically withdraws tools that qualify. If the user decides they still want the tool, that should be no big deal, they can just reapply. The current admins would lose their tools after a year if they didn't use them. If they wanted to retain all of them, just like they have them now, then all they have to do is use them, or reapply if they didn't use them.

• As the various tool projects find their feet, other editors in good standing could start applying for the tools.

• A given tool project would bring together the whole community of users of that tool. Because each project is responsible for monitoring standards against which the tool is being used, there will be much more awareness of each other and the ways the tool can be used than currently exists. Because of this focus, it should be possible to resolve disputes about the uses of the tool more effectively.

• I would see all the current tools being used in this way, except the ability to block editors in good standing. The ability to block vandals would be treated like the other tools, though that one would need some thought. It might perhaps amount to the ability to block editors who are not of good standing under guidelines that must be observed to ensure only vandals are blocked. Editor who fail to follow guidelines governing the use of a particular tool would be dealt with by the tool board, who would have the power to remove or suspend tool use.

• Tools might be removed for a period as a mild disciplinary measure, in which case the software would be notified to automatically restore the tools after the elapsed period.

More serious disciplinary measures involving editors of good standing would be entirely another matter. That could be carried out by a special board or boards, not that different from the current ANI, with empowered members perhaps elected by the community of editors in good standing along the lines of a stripped down RfA. Incidentally, a current admin who was elected this way would start out with all the tools admins currently use. The old term "admin" would no longer have a use, but could be applied to members of this board. The board would be tasked with the role of disciplining editors of good standing in areas not covered by the individual tool boards. They would have special powers, such as the power to block, and perhaps the power to remove the status "editors of good standing" for a period. A user without that status would not be able to use any tools. ArbCom would still have a role, partly to oversight the behaviour of this board. --Epipelagic (talk) 06:50, 14 October 2012 (UTC)

You've described an elaborate bureaucracy centered on individual tools. Apart from its sheer impracticality, this simply has no basis in the reality of why said tools exist and how they're used.

We don't need "blockers", "deleters" and "page protectors". We need users trusted to evaluate situations and act (or not act) appropriately.

The tools don't exist in isolation and mustn't be treated as though they do. (If a job calls for a hammer, a carpenter with only a screwdriver at his/her disposal is likely to turn it around and make do — with suboptimal results.)

To quote the above proposal, "a given tool would be withdrawn from a user if, say, it hadn't been used for 12 months." When a tool hasn't been used, that means that it hasn't been misused. How is that a valid reason to revoke it? Why should editors be punished for exhibiting such restraint? What harm would it prevent? Do you want to encourage users to find excuses to use the tools (just so they can retain them)? —David Levy 08:21, 14 October 2012 (UTC)

What "impracticality"? You're making that up. It's not an "elaborate bureaucracy" centered on individual tools, it's a series of small projects centered on individual tools. Often the tools will operate quite adequately as standalone tools. It might be that for certain purposes, users want two or three tools. So what's the problem? They would just apply for them. It's a modular approach. It's like a carpenter selecting the tools he wants from a board. You assemble the tools you need. Where is there "punishment" in removing a tool that has not been used for a long time? It's just a matter of limiting unnecessary proliferation. It's like a library asking for a book to be returned. The user only has to ask if they want it back. This is just a preliminary sketch, you can redraw it all sorts of ways. --Epipelagic (talk) 08:47, 14 October 2012 (UTC)

What "impracticality"?

For starters, there's the impracticality of establishing and maintaining a "block tool project", a "deletion tool project", a "page protection tool project", etc.

Then there's the matter of users having no clear place to turn when problems arise. Under the current system, they can simply seek out administrators and request assistance. They needn't determine what precise course of action is called for (e.g. whether they need a "blocker" or a "page protector"). They need only contact an administrator, whom the community trusts to evaluate the situation and act (or not act) appropriately.

And there's the non-consideration of administrative tasks not directly involving sysop tools (e.g. discussion closures).

You're making that up.

My opinion?

It's not an "elaborate bureaucracy" centered on individual tools, it's a series of small projects centered on individual tools.

That seems like an elaborate bureaucracy to me. Worse still, it isn't even logical. Again, the tools don't exist (and aren't used) in isolation. There's no valid reason to treat them as though they do.

It might be that for certain purposes, users want two or three tools. So what's the problem?

There isn't one. Just make trustworthy editors administrators and permit them to use whichever tools they need. They can be trusted to not abuse the others.

They would just apply for them. It's a modular approach. It's like a carpenter selecting the tools he wants from a board.

No, it's like a carpenter having to file a separate request (with the Hammer Commission, the Screwdriver Commission, the Saw Commission, etc.) for each of the tools with which he/she should simply be entrusted.

And when an unforeseen/emergency situation arises, let's hope that the right tools are in his/her box.

Where is there "punishment" in removing a tool that has not been used for a long time?

An editor's failure to use the tool would result in its non-voluntary revocation.

It's just a matter of limiting unnecessary proliferation.

What problem do you seek to solve? How is a tool's non-use harmful?

It's like a library asking for a book to be returned.

In what respect? A library book is lent for a specific period of time. The date on which it's due to be returned has no basis in whether the borrower's been reading it.

And the library possesses a finite quantity of each title, so a copy's return increases the book's availability to others. I honestly don't know what analogy you intended to draw.

The user only has to ask if they want it back.

He/she has to "reapply". —David Levy 09:39, 14 October 2012 (UTC)

Well setting out your reply so elaborately makes it look like you are countering with some substance. But I can't find it. It is easy to set up a central board where users can request assistance. Discussion closures could have their own project and be treated like a tool. Or they could be regarded as part of the responsibilities of the disciplinary group. Just because you say things are so doesn't make them so. Flatly saying it's not logical doesn't make it not logical. You need to back your statements with proper arguments and concrete examples. For example what is "Just make trustworthy editors administrators and permit them to use whichever tools they need. They can be trusted to not abuse the others." meant to convey? To me that sounds confused. You're making out problems exist which just aren't there. You only apply for the tools you want once. You don't apply every time you want to use them. In the second half of your reply, you seem upset about withdrawing the tools if they haven't been used for a long time. So let's just flag that suggestion, it's too trivial to argue about at this stage. --Epipelagic (talk) 10:09, 14 October 2012 (UTC)

I'm afraid I see this more the way David Levy does. Who would they apply to? That would become the new RfA - or a proliferation of new RfA venues. And as he's pointed out, you don't necessarily know in advance what "tools" you're going to need; that's not a perfect analogy, because you need more than one to perform various actions - for example, merging a history. A case could be made for unbundling the ability to edit a protected page (something that, as I pointed out in an earlier discussion on this page about inactive admins, is not logged so it's very hard to determine whether that ability is being used, but it is an admin-only right); however, because at least some of those pages/templates are protected for good reasons because mucking them up would do serious damage, there's a higher level of trust involved ... and judging whether to assign someone that right in isolation would inevitably start to look like an RfA. The likelihood of its becoming a mini-RfA is even clearer with the block button, which I suspect many people would like to see unbundled because they see themselves as being able to offer useful fast help against vandals. I think this would go in the wrong direction. We need adminship to be less of a fuss and bother, so that there will be more trustworthy people able to help out; and less like being crowned or awarded tenure so that it won't have such an aura of Power. The point of unbundling would be to achieve those aims, but multiple commissions and withdrawal of rights for not (visibly) exercising them would make it more of a fuss and more of a Big Deal. We're all trusted users, after all - that's what AGF is. Now that I've done some admin actions, I see that the "tools" are in fact quite interlocked. Certainly, deciding what to do requires me to have several. It's more like an auto transmission than a pegboard. (She says, no doubt once more demonstrating her poor scientific education.) So I now tend to see the problems with unbundling. (Plus, you never know what you will want or need to do down the road - just like in other areas of editing.) --Yngvadottir (talk) 12:21, 14 October 2012 (UTC)

I think it can be taken for granted that a number of admins are going to have little will to think about adapting things so alternative solutions can work. There are bound to be exceptions where this thing or that thing will not quite fit the general mold above, and might not work without some adaptation. Without a willingness for alternative approaches to work, it will be very easy to paint what appears to be a bleak picture. But that's all it is, the inevitably bleak picture that is so easy to paint if there is little willingness to make it work.

Yngvadottir, you ask "Who would they apply to?" I assume you mean who would an editor apply to if they wanted access to a tool. Did I not make that clear enough? A baseline acceptance applying to all editors would be that they must be editors in good standing. Beyond that, individual tool boards could add further qualifying requirements if they are appropriate. It would have nothing to do with a "a proliferation of new RfA venues".

Then you say editing a protected page is not logged. Yes, there are bound to be some wrinkles, and some rethinking may be needed to accommodate them. Is there any good reason why they can't be logged? If they can't be logged, then a modified solution needs to be found. That's clearly not difficult, because we've already worked that way for many years.

You talk as though it is simpler the way things are. Have you read Kudpung astonishing Advice for RfA candidates? The fully burnished shiny admin able to propel themselves through this extraordinary marathon of loops that Kudpung is erecting would be an unusual person indeed, but not someone who would necessarily make a good admin. This seems to be the way we are headed, going from bad to bizarre. The RfA process is not some infallible way of assessing the real calibre of a user. Look at some our worst admins as proof of how fallible RfAs are. Rolling a dice might do as well. You must remember that over half of all admins were appointed before 2008, when any child could get the bits for little more than the asking. For many years, hundreds of woefully underqualified admins have been holding the bits on Wikipedia, including the right to edit protected pages. Yet no real disasters have eventuated. Not because RfAs were magical rites of passage that guaranteed the trustworthiness of a user. Rather, because admins know the moment they use the bits inappropriately is the moment they will lose them. The same sanction applies to the unbundled bits.

I'm amazed that you can say "withdrawal of rights for not (visibly) exercising them would make it more of a fuss and more of a Big Deal". That part was the least of the proposals I set out above. David Levy kept going on about it too. It seems a big fear for some admins. So for now, lets forget about it. I said as much to David Levy. It is not necessary to keep things tidy. We can proliferate tools without limit if that reduces anxiety. You also say, "We're all trusted users, after all - that's what AGF is". Phew... that leaves me speechless. --Epipelagic (talk) 14:46, 14 October 2012 (UTC)

Beyond that, individual tool boards could add further qualifying requirements if they are appropriate. It would have nothing to do with a "a proliferation of new RfA venues".

On what do you base the assumption that these "individual tool boards" would institute standards laxer than those imposed at RfA? The possession of one tool requires no less trust (and as noted below, it actually would be more dangerous).

Again, the tools simply don't exist in isolation; they complement each other and should be used (or not used) as circumstances dictate. The idea of separate "boards" in charge of the deletion, blocking and page protection tools doesn't make sense.

You talk as though it is simpler the way things are.

It truly is. Certainly, RfA's culture has become far more complicated than it was (and far more complicated than it should be, in my view), but your proposed setup is intrinsically complex to the point of absurdity. And you've addressed objections by suggesting that more elements be added to fill the gaps.

You must remember that over half of all admins were appointed before 2008, when any child could get the bits for little more than the asking. For many years, hundreds of woefully underqualified admins have been holding the bits on Wikipedia, including the right to edit protected pages. Yet no real disasters have eventuated.

So why should the tools be unbundled? —David Levy 16:17, 14 October 2012 (UTC)

I hope you don't mind, but I've numbered your points for ease of answering.

1. The standards would not be laxer. Quite the reverse. A tool board will focus editors on the best practices using the tool. It will also facilitate users to become more familiar with each other, as the project has the responsibility of ensuring their members use the tool well. There could be programs to train new users in best practices. By contrast, the current admin system throws all the tools at an admin, regardless of whether they want to use them or not. An admin is presumed to be capable of using all the tools, not just the ones he or she elects to use because they want to use them. That's ridiculous. An RfA, as I said above, is not a good way of determining whether a user should have the tools. At the end of the day, the best way is to give them the tools, if there is no strong reason to not do that, and closely monitor how they use them. If you think that certain tools always go together as a group, then those tools should be grouped within a single tools project. I don't see any problem with that.

2. Well yes, there's a new layer of accountability surrounding the use of individual tools. That could be a very positive thing, with users setting and monitoring standards and training methods. The absurdities and horrors of the old RfA are gone. There might be a modified form of RfA to elect "admins" with the right to block experienced content builders, but those admin would not be expected to be the great renaissance figures of Kudpung's refurbished RfAs. The required skills, as with the tools, would became confined to what is specifically required. Admins with the right to block editors might be expected to have substantial content building experience and have shown they can be skillful facilitators and mediators.

3. If you look at the context, that was referring to the right to edit protected pages. Not do things like block experienced content editors. That has been a disaster area.

Most importantly, from my point of view, the tools should be unbundled so some dignity can be restored to content building. --Epipelagic (talk) 17:34, 14 October 2012 (UTC)

1. The standards would not be laxer. Quite the reverse.

My point is that there's no reason to expect that the standards would be less onerous than they are at RfA. Access to one of these tools in isolation requires no less trust than access to all of them does (and actually would be riskier), so it's likely that the same (or worse) demands would emerge.

2. A tool board will focus editors on the best practices using the tool.

There are only best practices for using the tools. Treating them as independent entities is analogous to focusing on users' use of individual keys on the keyboard (e.g. an "R board" advising editors on the best practices of R key use) instead of focusing on their editing as a whole.

3. By contrast, the current admin system throws all the tools at an admin, regardless of whether they want to use them or not.

And any trustworthy editor will simply not use the tools that he/she doesn't need.

As you noted above, "no real disasters have eventuated" under this system.

4. An admin is presumed to be capable of using all the tools, not just the ones he or she elects to use because they want to use them. That's ridiculous.

Admins are supposed to be trusted to not misuse any of the tools. The belief that they must have a demonstrable need for all of them is a relatively recent phenomenon (and one that I regard as harmful).

5. An RfA, as I said above, is not a good way of determining whether a user should have the tools. At the end of the day, the best way is to give them the tools, if there is no strong reason to not do that, and closely monitor how they use them.

That's much closer to an idea that I could support, provided that "tools" remains plural.

6. If you think that certain tools always go together as a group, then those tools should be grouped within a single tools project. I don't see any problem with that.

Okay. Let's call it "Requests for adminship". —David Levy 18:17, 14 October 2012 (UTC)

1. "Access to one of these tools in isolation... would be riskier": You have no basis for saying that whatsoever. You just made that up.

2. No, a tool is not at all like one key on a keyboard. If you want to use analogies, try and find one that actually is an analogy.

3. That statement, about no real disasters eventuating, referred to tool use only. There have been major disasters with the way seasoned content builders have been treated.

6. Where is there any logic in that? Where was there any suggestion that all the tools would be bundled together. The suggestion was that if certain tools went together, they could dealt with by one tools project. More important than the unbundling of the tools is the decoupling of the right to block experienced content editors.

You are misrepresenting the proposals very badly David. --Epipelagic (talk) 00:37, 15 October 2012 (UTC)

"Access to one of these tools in isolation... would be riskier": You have no basis for saying that whatsoever. You just made that up.

Can you please stop accusing me of that? If you don't understand the basis of something that I've written, request clarification instead of assuming bad faith. And if you still disagree with me, that's fine. But there's no need to question my sincerity.

I, along with others, have explained the argument behind that statement several times. Whether part of a package or assigned individually, one of these tools (e.g. blocking or page protection) has the same intrinsic potential to cause harm. In no way does the unavailability of other tools lessen the possible damage. It does, however, increase the likelihood of users making do with tool x when tool y is called for (i.e. misusing tool x) because they possess tool x and not tool y.

No, a tool is not at all like one key on a keyboard. If you want to use analogies, try and find one that actually is an analogy.

You appear to have misunderstood the analogy, the purpose of which wasn't to equate administrative tools with keyboard keys. The point relates to interdependency.

That statement, about no real disasters eventuating, referred to tool use only.

That's what I'm addressing. Administrators have been given the full suite of tools, and "no real disasters have eventuated" (even when some of the tools were unneeded).

There have been major disasters with the way seasoned content builders have been treated.

If you think that I'm disputing the fact that some administrators have grossly abused their positions, you're mistaken.

Where is there any logic in that? Where was there any suggestion that all the tools would be bundled together.

I'm suggesting it.

The suggestion was that if certain tools went together, they could dealt with by one tools project.

And I'm opining, as I have since my first response to this proposal, that all of the tools go together. So let's call the group "adminship" and the tools project "Requests for adminship".

You are misrepresenting the proposals very badly David.

You're misinterpreting my opinions as misrepresentations of your proposals. —David Levy 02:45, 15 October 2012 (UTC)

That 'Kudpung is erecting loops' is a totally disingenous claim. The loops were created by the community. All I did was to summarise them and back them up with links and footnotes. The irony is that 3 years ago I was so disenchanted at the behaviour of some admins that I began to research into how they were allowed to get the bit. Is the unbundling proposal therefore a process that would erode the tools from the current admins because those who pass what is claimed by many to be an obsessively high bar at RfA are still not to be trusted? One of the most oft heard rationales for opposition at RfCs to unbundle the tools is "Anyone who is to be trusted with any one of the tools should be mature and competent enough for them all.'" But adminship is not only about the tools themselves, it includes being able to conduct themselves in a civil and truthful manner during discussions and contentious issues. This unbundling proposal is not without merits, but in theory only. In practice it would require a huge multi-level bureaucracy to administer and there would be a stampede from the hat-collectors. It's time people realised that we're an encyclopedia and not a peculiar leadership game. Aye, there's the rub. Kudpung กุดผึ้ง (talk) 16:39, 14 October 2012 (UTC)

IIRC, you were the OMG candidate who appeared to be genuinely shocked at your nomination and who handled yourself at RFA so impressively that for the only time I changed from oppose to support. All of which sort of lends credence to something I have thought for a while. Pre-selection, adoption, Admin schools are all very well, but you cannot beat a genuine candidate with no skeletons rattling about which their ever so determined following of nominators and supports try gerrymander in, no matter what. If we want less drama at RFA we need more candidates like you and fewer hat collectors and highly recommended protégés. Leaky Caldron 12:42, 14 October 2012 (UTC)

Yup, the OMGs I got nominated for admin candidate :-) Thanks again for your confidence in me. But - and now I'm responding to Epipelagic above, too - some of our longtime admins are real internet heroes. And Kudpung and Worm, for example, are really working on solutions. Also the community doesn't really need more technical klutzes like me, per se :-) And I would argue that it does need at least a few more admins who very much know what they're doing ... like Rjanag from years past and Crisco 1492 and Scottywong from recently. I would agree though, we could use more RfAs like mine (... although with less socking). That was fun, on the whole :-D I've come out in the past for making it easier to remove adminship; that part of this proposal I can get behind ... except that it has to be part of making the whole thing less of a Big Deal. Otherwise it's going to be like removing tenure, and for good reason - as has been pointed out many times, we don't want admins to be afraid to do the right thing because it might make them open to grudge desysoppings. And the fundamental problem is that making adminship the prize at the top of the mountain attracts highly ambitious people, and turns off a lot who would probably be good at it. (Scottywong's changed the scoring on his admin candidate thingie and I'm now less than 3/4 of a likely candidate; I have someone in mind who scores well, but I hesitate to ask them whether they want to be nominated.) Sorry, Epipelagic, but we are all trusted users, at least till we show we can't be trusted. I'm not sure why that appals you so, but I may well be missing your point. I'm a great believer in making things simpler and less angsty, and I fear unbundling done the way you propose would be far more complex and in some ways even more angsty, but I'm just me :-) Yngvadottir (talk) 15:27, 14 October 2012 (UTC)

That scoring tool has no place in any selection process and should be deprecated and abandoned - quickly. It is unreliable, gamaeble and will set false expectations. Bad candidates will be proposed and good ones potentially denied. As for Admins being heroes, not in my book. There are good and bad ones and I'll keep my council on that. ;) Leaky Caldron 15:38, 14 October 2012 (UTC)

I do and have for a long time supported unbundling the tools. The RFA process is a nightmare and a joke. It causes more problems with its long beauracratic process than it solves. Multiple people in multiple venues in only the last couple weeks have stated as such and even referred to the process as defunct. Several have stated they had no desire to go through it. Yet here we are with multiple people (mostly admins) supporting that we keep it and fighting tooth and nail any suggestion to change it while at the same time agreeing its a bad process. We simply cannot continue on the current course of less and less admin promotions. Half every year since 2007. From 400 then to 20 now and on course for 10 next year if the trend continues while less and less editors new and old continue to edit, we have more and more trouble recruiting new editors and keeping them, more and more "admins" leave the site and more and more content gets "protected" causing more work for those admins that remain, causing a longer backlog at those venues that require admin action. We are really beyond what the perfect solution is at this point this unbundling needs to happen if the project is to survive. Its clear the RFA process is no longer viable and isn't going to cut it anymore. Kumioko (talk) 13:40, 14 October 2012 (UTC)

As you know, I agree that RfA is in a state of disrepair. But proposals such as this one simply aren't viable solutions.

The administrator rights package exists because it makes sense. The tools complement each other and become less useful and more harmful in isolation. The premise that they could be handed out individually with less bureaucracy has no basis in reality. The potential for misuse would only increase (because "page protectors" would be more likely to protect pages when blocking is called for, "blockers" would be more likely to block users when page protection is called for, etc.), so the community would have absolutely no reason to relax its standards. As I noted at the village pump and Yngvadottir noted above, we'd end up with multiple RfA-like processes. And that simply doesn't make sense, as anyone who can be trusted with one of these tools can be trusted with all of them.

You wrote above that some of us are "supporting that we keep [RfA] and fighting tooth and nail any suggestion to change it while at the same time agreeing it's a bad process". The process itself isn't bad. The culture surrounding it has gone astray. We want to fix (or at least improve) RfA by working to restore its nontoxic atmosphere. It won't be easy, but it's the sensible thing to do. —David Levy 16:17, 14 October 2012 (UTC)

You say "the tools complement each other and become less useful and more harmful in isolation." Here you go again, with more sweeping generalisations, just making things up. Then you say "the premise that they could be handed out individually with less bureaucracy has no basis in reality." How you manage to pluck that out of the vacuum? And then call the vacuum "reality". Unbundling largely resolves the current RfA problems, which is perhaps the worst bureaucratic nightmare Wikipedia has experienced. Nor is there any reason why "we'd end up with multiple RfA-like processes". You are just making these things up. If people were to accept uncritically what you keep asserting, on no grounds at all, it would seem the proposal wouldn't work. --Epipelagic (talk) 01:12, 15 October 2012 (UTC)

You say "the tools complement each other and become less useful and more harmful in isolation." Here you go again, with more sweeping generalisations, just making things up.

Can you please stop accusing me of that? If you don't understand the basis of something that I've written, request clarification instead of assuming bad faith. And if you still disagree with me, that's fine. But there's no need to question my sincerity.

I, along with others, have explained the argument behind that statement several times. Whether part of a package or assigned individually, one of these tools (e.g. blocking or page protection) has the same intrinsic potential to cause harm. In no way does the unavailability of other tools lessen the possible damage. It does, however, increase the likelihood of users making do with tool x when tool y is called for (i.e. misusing tool x) because they possess tool x and not tool y.

Then you say "the premise that they could be handed out individually with less bureaucracy has no basis in reality." How you manage to pluck that out of the vacuum?

I've explained that statement's basis too.

Your proposal is based upon the premise that unbundling somehow will cause the community to abandon the demands imposed at RfA (which we agree are onerous) and distribute the tools "without fuss". On what do you base this assumption?

Unbundling largely resolves the current RfA problems, which is perhaps the worst bureaucratic nightmare Wikipedia has experienced.

...in the scenario that you've envisioned, based on your belief that unbundling somehow would cause the community to behave reasonably again instead of transferring the "bureaucratic nightmare" to the new tool request fora.

Nor is there any reason why "we'd end up with multiple RfA-like processes".

Sure there is. The users currently populating RfA would move to the new fora. Why would they suddenly change their ways of thinking? —David Levy 02:45, 15 October 2012 (UTC)

The problem David is that we as Wikipedians have let the RFA process erode to such a point that any changes we make to it will take a long time to rebuild any level of trust from what it once was. It has steadily eroded since before 2007 and it will take probably at least that long to get it back. Its easier to gain fat than it is to lose it and the RFA process is a prime example of that. I also do not agree that unbundling the tools is going to be so bad as you describe. Yes we will have less people with the whole set of tools but more with the individual ones that are needed and it will be easier, as it is with filemover and rollbacker, to remove them if abused. Its nearly impossible to remove tools from an Admin, partially for good reason. The current RFA system also reflects that some of us users, myself included, will probably never have the tools and will probably always rely on others to implement our work. Which at some point will likely cause me to stop editing because more and more content will be protected (especially once the Protected changes module is implemented). At that point, there won't be any need for me or others like me because so much content will be protected only admins will be able to do anything. That is the future of Wikipedia as we are building it. Kumioko (talk) 16:34, 14 October 2012 (UTC)

The problem David is that we as Wikipedians have let the RFA process erode to such a point that any changes we make to it will take a long time to rebuild any level of trust from what it once was.

Perhaps so. But I've yet to see a better alternative proposed.

That's not to say that one can't exist. Maybe we should replace RfA with something else. But not this.

I also do not agree that unbundling the tools is going to be so bad as you describe.

Perhaps you could address my concerns in greater detail. Assuming that we can resume granting access to these tools in a non-onerous manner (via RfA or another means), in what way is a piecemeal approach intrinsically beneficial? How, in your view, is a user with one of these tools less likely to cause harm than a user with all of them is (and why should the community permit this more readily)?

The current RFA system also reflects that some of us users, myself included, will probably never have the tools and will probably always rely on others to implement our work.

If said users are trustworthy, they/you should have the tools. Otherwise, they/you shouldn't. Access to an incomplete toolbox doesn't require a lesser degree of trust; it only increases the likelihood of misuse, particularly in emergency situations.

Which at some point will likely cause me to stop editing because more and more content will be protected (especially once the Protected changes module is implemented). At that point, there won't be any need for me or others like me because so much content will be protected only admins will be able to do anything. That is the future of Wikipedia as we are building it.

It's one that we must strive to avoid. If too much protection is occurring, this needs to stop. —David Levy 17:17, 14 October 2012 (UTC)

Its already happening, too much content is protected already and the problem will be compounded ten fold once Protected changes comes online. In response to your other comments maybe the time has come for me to walk away from the project. I would never do any harm to the pedia and I think most editors know that but yet I am not trusted to have the tools and never will be because some editors feel I don't have the right temperment. So since temperment is an excuse to not give a productive editor the ability to do more for the project, then the tools need to be separated from the temperment or the editor needs to realize that they are not needed or wanted. I have yet to realize the latter so thus the former must be true! Kumioko (talk) 18:11, 14 October 2012 (UTC)

If an editor lacks the proper temperament to be trusted with the tools as a group, he/she also lacks the proper temperament to be trusted with them individually. (Note that I'm passing no judgement regarding your temperament or the fairness with which it's been gauged.) —David Levy 18:32, 14 October 2012 (UTC)

In which case the bar must remain high. Most editors could have the temperament to semi-protect an article, for example. The temperament / trust to remove an editor's ability to contribute is a different level for there the mistakes really do matter. Leaky Caldron 18:41, 14 October 2012 (UTC)

Any tool with the potential to provide an advantage in a dispute requires the utmost level of trust. And one must have the proper temperament to address the complaints that frequently arise in response to administrative actions (including protection/semi-protection). For example, a hypothetical editor with a dismissive attitude is poorly suited to the responsibility. —David Levy 19:17, 14 October 2012 (UTC)

A situation which is, unfortunately, all too common among those who have somehow managed to get a badge for life. Leaky Caldron 19:27, 14 October 2012 (UTC)

Agreed. —David Levy 19:46, 14 October 2012 (UTC)

Also, for what its worth, most of Snottywong's tools won't work for me. My 400, 000 edit count is too much for most of them and so are the hundreds of votes I have levied at the various for deletion/discussion boards. Even the Admin analysis tools craps out after about 20 minutes. Kumioko (talk) 16:06, 14 October 2012 (UTC)

1. It is easy to set up a central board where users can request assistance.

One already exists. And its functionality relies on the fact that administrators are equipped to deal with the various issues that arise; we have no "Wait, I'm only a deleter. Let me find a blocker." or "I'm not a page protector yet, so I'll just block these accounts for now." moments.

Editors also turn directly to administrators (via their talk pages and e-mail) for help. They shouldn't have to determine what tools are needed (and they might be mistaken).

2. Discussion closures could have their own project and be treated like a tool.

That simply doesn't make sense. Depending on the context, discussions can result in outcomes requiring one or more of the tools that you seek to unbundle. Again, these things don't exist in isolation (as you evidently imagine them to).

3.Or they could be regarded as part of the responsibilities of the disciplinary group.

Please define "the disciplinary group".

4. For example what is "Just make trustworthy editors administrators and permit them to use whichever tools they need. They can be trusted to not abuse the others." meant to convey?

I don't know how to state that in a plainer, more straightforward manner.

Trustworthy editors should simply be granted adminship. They needn't demonstrate an immediate need for every tool. We can trust them to use the tools properly (and not use the tools that they don't need).

Unbundling the tools doesn't reduce the risk of harm any more than unbundling a carpenter's tools does. As noted above, it actually increases the risk by encouraging the wrong tool's use when the right tool is unavailable. —David Levy 16:17, 14 October 2012 (UTC)

1. To repeat again, tool users would not normally apply for just one tool. Please stop persistently implying user would have only one tool. They would apply for a set of tools that matches the things they want to do on Wikipedia. Such a central board would work a lot better than current boards, because more users would have the tools.

2. To repeat again, tool users would not normally apply for just one tool.

3. The "disciplinary group" is the special board discussed in the final paragraph of the proposal.

4. When you make statements like that, I can only assume you have yet to read the proposal. --Epipelagic (talk) 01:37, 15 October 2012 (UTC)

To repeat again, tool users would not normally apply for just one tool. Please stop persistently implying user would have only one tool.

To repeat again, I'm not claiming that. I'm referring to scenarios in which users possess some (but not all) of the current administrative tool package.

They would apply for a set of tools that matches the things they want to do on Wikipedia.

Again, it's impossible to know which tools are needed until confronted with a situation. There aren't separate tasks for "blockers" and "page protectors". One must examine the circumstances and then decide what course of action is appropriate.

Such a central board would work a lot better than current boards, because more users would have the tools.

You base this assumption on the unsubstantiated premise that the community will distribute the unbundled tools "without fuss".

The "disciplinary group" is the special board discussed in the final paragraph of the proposal.

Why, in your view, should discussion closures (most of which are unrelated to discipline) fall to this group?

When you make statements like that, I can only assume you have yet to read the proposal.

I'm sorry that you regard my disagreement that your suggested setup would function as you envision as evidence that I haven't read the proposal. I assure you that I did so before replying. —David Levy 02:45, 15 October 2012 (UTC)

Unbundling is a poor idea. Administrators need to have all of the options at hand in order to act with the greatest objectivity. If a user had only one tool, that tool would likely be used in cases where another tool—unavailable to the user—was more appropriate: "To the man with a hammer, everything looks like a nail." Keep the tools together. Binksternet (talk) 17:05, 14 October 2012 (UTC)

This argument really doesn't make sense. If I had the right to edit a protected article or pull in more than 25000 articles to AWB how would that be useful for blocking a vandal? If I had the right to protect or unprotect a page how would that allow me to see protected reports like Unwatched articles? There are a lot of tools in the admin toolbox that are not and do not need to be admin only tools. Kumioko (talk) —Preceding undated comment added 18:11, 14 October 2012 (UTC)

Unbundling the tools creates inefficiencies. Now, instead of one person doing a job all at once, one person will do step 1 of the job, and then they have to go find someone who can do step 2, and then that person has to go find someone who can finish the job with step 3. A task that would have taken a conventional admin seconds could end up taking hours. Nothing that is gained by unbundling tools is offset by these increased inefficiencies. -Scottywong| spout _ 18:14, 14 October 2012 (UTC)

Kumioko, consider the possibility that I was able to block an IP but not semi-protect a page. Sometimes in the case of multiple or dynamic IPs, protecting the page is the best choice. Sometimes in the case of public IPs such as those in libraries and internet cafes, blocking the IP is harmful to the next, probably innocent person who uses the service.

On the other hand, consider if I were able to semi-protect pages but not to block IPs. If a single IP editor was running wild with vandalism, I could not simply stop him and end the trouble. Instead, I would have to keep protecting articles that he had hit, refreshing his contribution page to see where the next attack was. I would have to appeal to an administrator to block the IP. Binksternet (talk) 18:34, 14 October 2012 (UTC)

Points like these are irrelevant. The original proposal nowhere suggested that users would be allowed access to only one tool. They could have as many as they needed. --Epipelagic (talk) 20:55, 14 October 2012 (UTC)

Again, it's impossible to know which tools are needed until confronted with a situation. There aren't separate tasks for "blockers" and "page protectors".

Indeed, under your proposal, users would be able to file separate requests for the various individual tools. This would provide no benefit. Trustworthy users should simply be made administrators, not forced to apply to separate "boards" to assemble a complete set of tools. —David Levy 22:02, 14 October 2012 (UTC)

Oh dear, according to you, "forced to apply to separate 'boards' to assemble a complete set of tools". Forced? You would apply to the relevant boards. If you already had some relevant background and were a "editor in good standing", your application would generally just be "rubber stamped". The whole process might take a couple of hours for several tools. Compare that with undertaking a full blown RfA. The fantasies you have outlined above of undergoing mini RfAs for each tool are sheer nonsense. I agree there is no point unbundling if we are going to be stupid about it. --Epipelagic (talk) 01:49, 15 October 2012 (UTC)

You're describing what you want to occur. I'm opining that it's unrealistic. My position is based upon the attitudes currently exhibited in the community, which I don't believe would magically vanish upon your idea's implementation. —David Levy 02:45, 15 October 2012 (UTC)

A lot of the tools are interdependent. So flat unbundling, isn't just bad, but it just wouldn't work on the technical side. And that's another thing, many tools are "separate" tools only because they were coded that way, they could just as easily been bundled together in coding (PROTECT is a good example of that). So, while I understand requesting full unbundling sounds good on the surface, when you start to look at what we're dealing with, it really won't, and can't work. And that without even going into asking what the process for granting these unbundled tools would be. - jc37 18:48, 14 October 2012 (UTC)

See also this proposal which, AFAICT, is about as close as we've come to unbundling. I've been considering trying again. But I'm still working on trying to write clearer explanatory proposal text. - jc37 18:56, 14 October 2012 (UTC)

Vague, yet quasi authoritative technical statements like these have a chilling effect on proposals, because it is not clear to what extent they are true. Are you saying you have a panoramic insider view of the relevant software and other technical issues, and you know for sure that unbundling "can't work". If so to what extent is it that they "can't work"? Or is it that you have an incomplete insiders view, but you have heard it said that that there are issues? If there really are insurmountable technical problems that make unbundling impossible then that is an end to any proposals about unbundling, and it is long overdue for the group that collectively have the insider perspective to step forward and formally say what can and what can't be done. Otherwise, these arguments just go round in circles.--Epipelagic (talk) 20:55, 14 October 2012 (UTC)

AFAIK, complete unbundling of every individual tool is simply not technically possible. In some cases, the coding involved requires that the editor has another tool.

I've spent a LONG time researching the tools. I've made MANY proposals. And have been in MANY discussions related to this.

That said, I'm human, and am welcome to being informed of where I am mistaken.

But just because you don't like the answers you're getting, doesn't make them any less accurate. - jc37 21:02, 14 October 2012 (UTC)

Of course I don't like the answer. I would prefer more flexibility in the system. Naturally, when people keep just saying that the proposal it is not possible, I want to know why. So far most of the arguments have been directed at positions that are not in the proposal (which is huge waste of time). But your technical issue is different, the first counter than may carry real weight, so it definitely needs sorting out. So are you in a position to state with accuracy just what can be done from a technical point of view and what can't be done? And I don't just mean with respect to the current implementation of the toolset (though that would be a start), but also what can be done if the toolset were to be recoded. --Epipelagic (talk) 21:38, 14 October 2012 (UTC)

I'm not a dev. I've not seen the code. I'm relying on what I've read, written by those who have.

As for re-coding, good luck with that. There is already a way in code to allow individual user-rights. (Essentially what you call "unbundling" is really just creating a user-right package which only contains one tool.) But that doesn't change what the user-rights do, or what they require to work. Beyond that, from what I've read, I highly doubt any dev resources would be spent on recoding the way user-rights work unless there was an amazingly huge, overwhelming consensus. This latter being just my understanding, of course. - jc37 22:49, 14 October 2012 (UTC)

I struck part, because I just remembered that I've been shown several sections of code involving this. - jc37 22:53, 14 October 2012 (UTC)

You are right that some tools are dependant upon others. Protection and Unprotection would need to go together as would Delete and undeletion, Blocking and unblocking. So your correct in that sense. But there are several such as Editing protected pages or Even Editing the interface that shouldn't and needn't be admin exclusive. Also increasing the wingspan of admin functions by allowing niche tasks to be done by the folks that deal with them isn't a bad thing anymore than being a brain surgeon is a bad thing being separate from a general practice doctor. You are also correct that in some cases it might bring upon some inefficiencies in that I may have the tool to protect the page someone vandalized but not block them or vice versa but since we will have a lot more people with each of those separate groups of tools, there will be more people that can do these things independantly without having to wait on admin intervention. I really don't expect this to pass and certainly I don't expect the majority of Admin's to accept this but its clear the RFA process isn't and doesn't work. We should not wait until there are few admins left or the workload becomes so overwhelming to Admins that no one wants to be one. Kumioko (talk) 20:02, 14 October 2012 (UTC)

You are also correct that in some cases it might bring upon some inefficiencies in that I may have the tool to protect the page someone vandalized but not block them or vice versa

Not just inefficiencies. When only a block is warranted, "page protectors" would approach the situation with the wrong tool in hand. (The same goes for "blockers" when only page protection is warranted.)

Again, these tools require the same level of trust, so there's nothing to be gained by providing access to one and not the other. It would only increase the likelihood of misuse.

but since we will have a lot more people with each of those separate groups of tools

For the reasons discussed above, this isn't a safe assumption. I see no reason to believe that the community would hand out the individual tools with less reluctance than it does the current administrative package. —David Levy 20:30, 14 October 2012 (UTC)

Then unfortunately the system will continue to degrade as it has been until there aren't enough admins or editors willing to be...as is already happening. The RFA process is hopelessly lost at this point and any ideas to fix it are dismissed. Many of us senior editors that have the technical and procedural knowledge and the desire to do more can't pass RFA or don't want to try can't get any of the tools because they need to be kept together. I just do not see anything good coming from an already bad situation. Kumioko (talk) 20:56, 14 October 2012 (UTC)

Again, I'm not indiscriminately opposing any and all reform (including the possibility of replacing RfA with something different). I oppose this particular proposal. —David Levy 22:02, 14 October 2012 (UTC)

@ DL - You know, for years now, I keep hearing the If they're to be trusted with some, then they should be trusted with all argument. And I'm sorry, I just don't buy it. Please pardon the phrasing, but I think that's simple lazy "citizenship" of sorts. ("I don't want to be bothered to actually look at the tools, it's easier just to "vote" on entrusting someone with a bundle.") And that "laziness" just results in this huge mass of tools.

I don't disagree with the block vs protect arguement, but that's a simple fix.

But we already have separate tools for checkuser, and we obviously don't trust all admins to this tool. So if there are tools we can trust certain editors, but not all admins, why can't there be some tools from the (current) admin package which we might trust some editors, but not all editors? - jc37 21:10, 14 October 2012 (UTC)

There are. They've already been unbundled. And while I oppose unbundling the ability to edit protected pages, it doesn't strike me as an illogical idea. Replacing RfA with separate request processes for each individual tool does. —David Levy 22:02, 14 October 2012 (UTC)

There are others which could be unbundled, I think. That aside, I agree with you that the process for granting is and has been a stumbling block when discussing unbundling. - jc37 22:18, 14 October 2012 (UTC)

David, you continue to talk as if unbundling meant users would have access to one tool only. Of course that wouldn't make sense, and no one proposed that. Your statement "I see no reason to believe that the community would hand out the individual tools with less reluctance than it does the current administrative package" is based on what? If the community is going to be that silly, then you are right, and nothing can work. The whole point of the proposal is to allow access to experienced users without fuss. That's why, in the proposal above, I went to length to explain how the process could be monitored to minimise problems. You are arguing against proposals that were never made, tilting at windmills that aren't even there. Please argue against the proposals, and not against positions that don't exist. --Epipelagic (talk) 21:14, 14 October 2012 (UTC)

David, you continue to talk as if unbundling meant users would have access to one tool only.

I'm not claiming that. I'm addressing the fact that some users would have access to one tool (or some quantity below that of the current administrative package). In my view, this would provide absolutely no benefits. Only detriments.

Your statement "I see no reason to believe that the community would hand out the individual tools with less reluctance than it does the current administrative package" is based on what?

I've explained my reasoning repeatedly. I'm referring to the level of trust required, which isn't magically reduced when the tools are broken up. (As noted previously, it would only increase the likelihood of misuse.)

The whole point of the proposal is to allow access to experienced users without fuss.

And I, along with others, am attempting to explain why it would fail to accomplish that goal. —David Levy 22:02, 14 October 2012 (UTC)

I just want to clarify the misperception that we would need to create a new venue for these rolls if unbundled and implemented. The venue already exists at Wikipedia:Requests for permissions where users currently request the Rollbacker, Filemover and Edit filter manager rights. Its already in operation and it works. There is no reason to assume that it would not work if additional roles are unbundled. Kumioko (talk) 22:58, 14 October 2012 (UTC)

Whether that process "works" is debateable. It may work in the granting, but follow-up is uncommon. And I don't know that I, at least, would be comfortable with other tools being given out in that way. - jc37 23:05, 14 October 2012 (UTC)

Regardless, that isn't what's proposed above. —David Levy 23:25, 14 October 2012 (UTC)

Comment on this thread: While current admins should be involved, since they have relevant perspectives, they should not be allowed to dominate proposals like this, since they are a minority group and they have too much investment in the status quo. An effort should be made to engage more content developers on issues like this, since it is the content developers who will achieve less demeaning working conditions if proposals such as this one succeed. However, my guess is most have lost heart, and will not engage. In particular, admins who were appointed prior to 2008, and especially those that started out as schoolboy admins and have never made any substantial content contributions to Wikipedia, should not be allowed to dominate discussions like this, and should not be allowed to derail them with persistent one sided misrepresentations. --Epipelagic (talk) 02:20, 15 October 2012 (UTC)

I'm a veteran content editor who thinks unbundling is not a good idea, that it would be inefficient, that it is one of the less practical answers to a problem with other solutions. Binksternet (talk) 02:43, 15 October 2012 (UTC)

I've participated in this discussion in good faith, with the community's best interests at heart. At no point have I deemed myself superior to other editors, nor has any other administrator commenting here.

As I've stated on numerous occasions, I agree that RfA is broken and want very much to address the problems (whether that means fixing RfA or replacing it with something that works better).

I oppose this proposal not because I believe that it would make the tools easier to obtain (which I agree is a high priority), but because I believe that it would make matters worse.

You're entitled to disagree. I only ask that you do so respectfully. Your above characterization is uncalled-for. —David Levy 02:45, 15 October 2012 (UTC)

Epipelagic has done very little throughout these discussions but make disrespectful and offensive generalizations and characterizations, and I for one am intensely tired of it. — Hex (❝?!❞) 17:55, 16 October 2012 (UTC)

Hex, I understand you get "intensely" annoyed. That is clear from the incoherent and crude characterisations you keep making about me. I don't reply to them because they don't address substantive issues, or address, apart from here, anything I actually said. This latest outburst of yours follows on from, and was perhaps triggered by the last post I made here, which may be offensive to you, but is objective and specific and is not a generalisation. --Epipelagic (talk) 20:08, 16 October 2012 (UTC)

Roflcopter. So far, you've caused an invocation of Godwin's Law, compared the current situation with our administrators to the military dictatorship in Burma, referred repeatedly to (unspecified) younger adminstrators as "children", and far more besides. What substantive issues do these comments address? And I'm the one making "crude characterisations", apparently. You're a hypocrite. And by the way, I said tired, not annoyed. It takes a lot more than talk page blowhards to annoy me. — Hex (❝?!❞) 07:50, 17 October 2012 (UTC)

You're talking drivel. --Epipelagic (talk) 08:44, 17 October 2012 (UTC)

Nice comeback. — Hex (❝?!❞) 14:02, 17 October 2012 (UTC)

Thank you. It's good we've found something we agree on. --Epipelagic (talk) 22:18, 17 October 2012 (UTC)

This starts the review that we really need. But the separation that is really needed is that the resolution of difficult situations needs someone more qualified that an average admin.. And of course, the current broken RFA system of criteria "never did anything that somebody didn't like" needs to be fixed. North8000 (talk) 11:47, 31 October 2012 (UTC)

Unbundling the tools won't fix anything

What is it about adminship that makes community standards for RfA so high? Among other things:

1. Admins close and implement deletion discussions (which includes implementing CSD and prod tags).
2. Admins deal with vandalism by blocking users, protecting pages, and deleting pages.
3. Admit it or not, adminship is a high-profile position of trust. So bad behavior on the part of an admin reflects more poorly on Wikipedia than bad behavior on the part of an editor without "special hats" or with only low-profile "special hats".

Unbundling won't help with any of these.

1. Any bundle that includes the ability to delete is going to have the same problems as RfA; just look at how many RfA opposes are based on poor records at CSD or AfD, resulting in a lack of trust that the candidate will be able to properly use the delete button. Arbitrary policy restrictions aren't going to be trusted by most RfA voters, and arbitrary technical restrictions (were they to be somehow added to MediaWiki) are unlikely to placate enough people to matter and still leave the unbundled tool relevant. And note also that WMF requires that any bundle that includes the ability to undelete or view deleted revisions must have an RfA-like approval process with the same criteria for promotion, at which point you're into "why bother unbundling this?" territory.
2. Any bundle that is intended to deal with vandalism will either fail (as many have done in the past) on Maslow's hammer issues or will have to include all three of blocking, protecting, and deleting. Besides the issues related to delete tool mentioned above, a bundle including the block tool is very likely to also have the same problems as RfA as it seems there are few in the community who would trust someone with the ability to block without trusting them with the other tools in the admin package. And as above, arbitrary policy restrictions aren't going to be trusted by most RfA voters, and arbitrary technical restrictions (were they to be somehow added to MediaWiki) are even less likely to placate enough people as ill-considered blocks of new editors are considered a very serious problem by many in the community.
3. The "high-profile position of trust" issue is probably the most likely to be improved by unbundling, but I personally think that unbundling would just be creating more "high-profile positions of trust" without seriously reducing the RfA opposes based on this issue. IMO, renaming admins to "janitors" would do more to reduce the high-profileness than unbundling would, and still the fact that these "janitors" had the ability to delete new users' pages and block them would probably do little to reduce the concerns raised about personality and temperament in several recent RfAs.

So yeah, unbundling can work for some things, like unbundling rollback or the current "editprotected" unbundling proposal, that are seldom if ever the grounds for an RfA failing. But it's not going to fix RfA.

Unfortunately, I don't have any solution to offer for the "RfA problem", which is why I've tended to stay out of these discussions. Making it easier to remove problem admins would probably cause the bar to lower, except that it would probably cause other problems by making it even more difficult for admins to deal with controversial matters (unless we had "tenured admins" with the same sysop bit but not open to community recall so they could deal with the controversial stuff without fear, but that would probably have a worse RfA process and be opposed for WP:PEREN#Hierarchical structures). Waiting until there are so few admins that things start falling apart and the community must lower standards will probably work, but it's an awful "solution". "Clerking" of RfA and more enforcement of civility would reduce the negativity that scares off some candidates and might somewhat reduce the pile-on effect, but would do little to reduce the generally high standards for adminship. Anomie⚔ 19:33, 14 October 2012 (UTC)

What anomie descibes above pretty much matches my experience with these things.

Though I think the arguement is more about the interdependence of block/protect (an editor with one, should have the other, since situations vary). Delete isn't really much of a behaviour-related tool (it deals with the effects of vandalism, and not so much with the prevention of it), though anomie is quite right that it may be seen as a "big deal" to some editors, content editors in particular.

It's due to these points (and others) why I tried both this and this. But unless/until we get past the "votes" of "if we trust them with some then we should trust them with all" or "I oppose this because I have my own idea, and I'm afraid if this passes, mine won't pass", then these have little chance of achieving consensus. - jc37 22:00, 14 October 2012 (UTC)

I took a look through a few recent failed RfAs and noted that a fair number of opposes were along the lines of "This user wants to work in deletion-related areas, but has a poor CSD record and few XFD contributions". So yes, deletion seems to be a big deal for a significant portion of the community. Has anyone recently (or ever) done a study on the different categories of oppose reasons, and how many contributed to the failure of recent RfAs? Anomie⚔ 23:04, 14 October 2012 (UTC)

For me it would depend on the tool frankly. I think that the following things could be done relatively harmlessly through that: Protecting/unprotecting articles, Editing protected pages, Editing the interface (As its own rule similar to Edit filter manager), pulling in more than 25000 articles to AWB, viewing protected database reports like Unwatched pages. Many of those things are not to me inherently administrative. Others would likely need a more significant process like Blocking/Unblocking and deleting/undeleting content. Kumioko (talk) 23:11, 14 October 2012 (UTC)

Those who suggest that in the case of impasse the WMF should design and impose a new system or set of regulations by fiat, may have missed the statement by the Foundation only 3 months ago (with their bold text) in response to what was possible one of the best, most coherent, and well presented of the many proposals. While they will not interfere with with the community's decisions, they will insist that the same scrutiny and standards at RfA be applied to any requests for an unbundled tool. As unbundling requires software tweaks that only the WMF staff can do, they have the power to reject any proposals in the same way as they have rejected major community consensus in the recent past. Common sense tells me that there is little to be gained by further examining the possibility of unbundling due to :

• The additional bureaucracy it will need
• The stampede for tools by the hat-collectors.
• The many mini-RfA style requests for the tools that will generate more heat than light.
• WP:PERM works because probably 90% of the requests are denied. Interesting is however, that there are only two or three admins who are prepared to work there, and the process is hampered by overly enthusiastic editors trying to turn it into a mini RfA-style discussion.
• Any assumptions that individual tools require less competency, maturity, and trust to obtain than the complete set, and can thus be accorded without fuss.
• The unique (and unilateral) power of paid staff to accept or reject software modifications just because they don't like them.
• The many previous attempts to propose unbundling the tools that have failed.
• The failure of the community to discuss changes dispassionately and objectively, beyond their own personal wants and needs.
• The failure of discussion participants to read threads from the top, and take into account research and discussion has already taken place and is available for consultation.

That list may have a chilling effect, but at the end of the day, IMHO, the only thing that needs to be addressed is the behaviour of users on RfAs. And on that note I will reiterate again for those who missed my slightly earlier post above, that the allegations that I have a personal agenda to set the the bar high and abolish oppose votes from RfA altogether are ludicrous and and are not conducive to healthy discourse.

Kudpung กุดผึ้ง (talk) 00:52, 15 October 2012 (UTC)

I think you are mixing apples and oranges here Kudpung. The link you provided was for the creation of a moderator role. Additionally, several tools have been separated from the admin bit before which are Edit filter manager, Rollbacker and Filemover, which all use a separate vetting system than RFA so the precedent has already been set for splitting tools out without adding to the beauracracy. Kumioko (talk) 01:13, 15 October 2012 (UTC)

The tools that have been split already, and the ones that are being asked for now, are the apples and oranges. The Foundation has already stated that any tool to be unbundled in the future will require the same degree of scrutiny of the candidates as for RfA. And that is not what happens at PERM for rollbacker and reviewer, etc. Kudpung กุดผึ้ง (talk) 11:30, 15 October 2012 (UTC)

Kudpung, as you know, I've not been active for a while, so I may have missed this, but the last time I checked it was only deletion/undeletion which the Foundation wanted scrutiny for. Has anything changed since then? Worm^TT(talk) 11:43, 15 October 2012 (UTC)

I think it was a legal issue related to allowing people to see deleted content - iirc, the WMF's legal bods reckoned nobody can be allowed to see deleted content without a selection process akin to RfA -- Boing! said Zebedee (talk) 12:05, 15 October 2012 (UTC)

That's what I thought (last confirmation was August 2011) but unbundling block (or even unbundling block of users with less than X edits) shouldn't be a problem... it's something I'd support even. Worm^TT(talk) 12:12, 15 October 2012 (UTC)

Blocking probably is the most contentious administrative action, so it's highly unlikely that the community would grant access to the tool without an RfA-like process or something comparably stringent (which would defeat the purpose of unbundling it).

Regardless, I would strongly oppose any proposal to create a separate permission to block users with fewer than x edits. Yes, that would cover a vast majority of vandals, but it also would include well-intentioned newcomers, whose edits are easily mistaken for vandalism. If anything, greater care is required to avoid making the wrong call and driving away new editors.

There's also the issue of interdependency (particularly with page protection), as discussed above. —David Levy 12:55, 15 October 2012 (UTC)

I agree with David. I started the proposal to village pump to increase productivity to the 'pedia by experienced non-admins handle edit requests of fully protected pages and perform needed maintenance of their own. My motif, because the number of admins are shrinking, they should focus more of time in contentious areas. This proposal is an attempt to relieve the admins of a job that shouldn't be just limited to admins.—cyberpower ^Chat_Offline 13:05, 15 October 2012 (UTC)

(edit conflict)(edit conflict)(edit conflict) The statement was here. Without reading between the lines (which i don't usually do) it seemed clear enough to me at the time. It was around that time though, following that statement, that I gave up following the debate because I knew the proposal would not reach consensus. You're welcome to read it and do let me know if I was wrong. Kudpung กุดผึ้ง (talk) 12:32, 15 October 2012 (UTC)

For the record, the Foundation was making that statement only as regarding the delete/undelete right. We have not taken a position on whether other rights would need RFA-like scrutiny. Philippe Beaudette, Wikimedia Foundation (talk) 15:09, 27 October 2012 (UTC)

I agree that all admin tools require the same degree of trust. However, the skill required for different admin tasks does differ between tasks. Editing protected pages and moving require the user to be able to judge the presence of consensus while blocking and protecting require the user to be able to correctly judge bad faith. Trust is mostly a criterium to keep bad faith users away from admin tools, but I do not believe most mistakes (including those made by admins) are caused by bad faith but rather by misjudgement.

Therefore, having protected page editors who are less adept at judging bad faith may not cause problems, while vandal protection users do not require good article review and consensus detection skills. As the skills to judge the appropriateness of the tool does differ between tasks, specialisation of admins may not be a bad thing. This will however still require packages of tools and not a complete debundling and I do not know how the community will judge for these specific tasks.

I also do not think these specialised users should replace the current generalist admins so it is only useful if the criteria will indeed be more lenient for the specialists (and I assume RevDel will cause the same concerns from the WMF as undelete, so both would be reserved for the generalists). The system would indeed increase interdependence between specialists, which is not necessarily a problem if the generalist admins do the less routine tasks.

It may not solve the problems, but a hybrid system, with a milder debundling than proposed above is probably not harmful. Therefore, I think the total expectation would be positive and we could at least try.

That being said: as I understand a software change is necessary, shouldn't this discussion be on the mediawiki or meta wiki? Wouldn't a change in the user-rights profiles affect all projects and not just en:wikipedia? PinkShinyRose (talk) 01:38, 28 October 2012 (UTC)

Discussing the change here (in en:wiki) is appropriate. Some changes that are requested here then become available both to all projects and to all wikis, but if it is implemented in a manner that only enables this wiki to do something, then none others are affected. The wiki software has many capabilities that are not used by some wiki's and by some projects. So discussing it here is fine. Apteva (talk) 13:38, 28 October 2012 (UTC)

Just to add to that, the available user groups, the rights assigned to them and the names the groups are given is fully configurable for each wiki. There are very few technical restrictions on what can be done. QuiteUnusual ^TalkQu 20:12, 28 October 2012 (UTC)