Wikipedia talk:Requests for adminship/Archive 248

I will just note here that I believe things w.r.t us having the necessary freedom of process have gotten to the point where blocks or some other intervention against those that attack oppose voters may have to be considered. The level of vitriol for simply voicing one's opinion in accordance with the process has become excessive, and we need an urgent reversal of this. Mob rule is not an acceptable way to conduct an RfA, and I think we urgently need to demonstrate that RfA as an open process is salvageable and will not be needing replacement with a secret voting system. Samsara 10:31, 13 October 2017 (UTC)

I agree, there is far too much being said to that one oppose voter on MLG's RfA. The point has already been made that the opinion is not popular, in fact, the point was already made implicitly by the tally of 200something/1/0 when the comment was published. All the hounding is just turning it into a mare in red. – filelakeshoe (t / c)  10:36, 13 October 2017 (UTC)

As somebody who has been there and done that, all I can say it's just a standard part of discussion. As Paul Graham put it, "Agreeing tends to motivate people less than disagreeing. And when you agree there's less to say. You could expand on something the author said, but he has probably already explored the most interesting implications. When you disagree you're entering territory he may not have explored." So an "oppose" is more likely to attract discussion about differing views than "support", which are merely endorsing or regurgitating what the nomination has said. You get the odd badgering of support voters on obvious NOTNOW cases. Obviously I have a dog in one of the fights currently running at RfA so I wouldn't personally get involved, but in my experience the best option is to shunt discussion off to the talk page, and if necessary, hat the discussion with a comment that paraphrases "what has any of this got to do with deciding that 'x' would be a good administrator?" Ritchie333 ^{(talk) (cont)} 10:39, 13 October 2017 (UTC)

Ritchie is correct. RfA is a discussion, not an election, so disagreements should be discussed. That said, if ten people have already raised the same criticism of a certain !vote, one does have to consider whether they really need to add another response. Imho, some more clerking by crats might be in order when it comes to such discussions, removing off-topic discussion to the talk page. Comments such as "baseless and shameful", "bad faith", "sad I supported your RfA" etc. might reflect on someone's feelings but they don't add anything to the discussion and are better added on the talk page. Regards SoWhy 10:46, 13 October 2017 (UTC)

Except that it is actually an election, with numeric cut-off points for pass and fail, and even 'crats acting as sort of electoral college for grey-area cases. RfA is both an election and a discussion, and no amount of Wikipedian distaste for vote counting is going to change that.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  20:57, 13 October 2017 (UTC)

As a wikipedian of 10+ years - and someone who's passed RFA I would have to disagree. RFA has been a discussion as far as I can recall. If you make a weak argument on either the support or oppose side, you are fair game for a counter-argument. SQL^{Query me!} 04:06, 14 October 2017 (UTC)

Thus the "and" in "RfA is both an election and a discussion".  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  17:27, 14 October 2017 (UTC)

There's a rather important distinction you're missing between badgering a random oppose and condemning an oppose that's rather overtly based on sexism. We would never see an oppose based on someone primarily writing articles on men. ~ Rob₁₃^Talk 11:25, 13 October 2017 (UTC)

The only sexism that has been displayed in this whole sorry affair is the assumption that User:Xxanthippe is male. It seems that that editor has been pretty scrupulous about not disclosing whether she is a woman or he is a man, but the only person I have heard of with that name (without the double "X") was female. 86.17.222.157 (talk) 12:40, 13 October 2017 (UTC)

Feel free to point me to where I assumed a gender. I did not. ~ Rob₁₃^Talk 12:44, 13 October 2017 (UTC)

You did not, but others did. 86.17.222.157 (talk) 12:58, 13 October 2017 (UTC)

• (edit conflict)It is important to universally establish a few things here:
  • What are the correct grounds to strike any vote?
  • Should explicit consensus exist to strike a vote?
  • What are the grounds in which discussion of a vote is acceptable? (one might think here, WP:CIVIL and such, but I think that may have failed us in this case)
  • Should the above be a binding guideline or policy, rather than the informal advice given in WP:RFAV?
• The problem lies in that we have varying opinions and ideas related to this. It's all a little bit chaotic. The Megalibrarygirl oppose was struck, unstruck, struck again, and then unstruck, with explicit instructions by Cyberpower678 to not strike the vote again. A vote which has been overlooked here was Paulmcdonald's vote on TonyBallioni's oppose, which pressured him to move to neutral. He may be wrong, IMHO, but that was not nice. My name continues to not be dave (talk) 11:27, 13 October 2017 (UTC)

• "sad I supported your RfA" - I said that, oh yes. But not before having thoroughly reviewed their history. And why did I do that? Because I just couldn't believe at first that an admin could be so mean spirited. It was a deliberate attempt to provoke drama (he said so), and drama he got. And peer pressure did what it was supposed to. Kudpung กุดผึ้ง (talk) 11:52, 13 October 2017 (UTC)

• I fully agree with the sentiments in the subheading being discussed and second the full letter and spirit of filelakeshoe's eloquent admonishments against such belligerence. Please check all strawmen at the door and refrain from usurping this discussion by such means. I see nothing in Xxanthippe's oppose rationale remotely close to the blatant personal attacks being volleyed about for the sake of comparison.

I'll go a step further off the beaten path and disagree with notions that subjectively chosen commentary ought be refactored, as published, and segregated away from the RfA to its talk page. In my opinion the whole RfA process would improve many fold if instead we followed the Arbcom model and prohibited threaded discussions outright. If a threaded discussion was needed, I would concur that the RfA talk page would be the best place for it to sprawl, and it should then be published to the talk page, not moved there; so as I believe.--John Cline (talk) 12:44, 13 October 2017 (UTC)

• FYI, I'm indicating towards an RfC on this matter, or some clear resolution. We can't go on like this. My name continues to not be dave (talk) 12:50, 13 October 2017 (UTC)
• The question is "what is excessive". I do believe that ridicule is the only correct response to ridiculous votes ("less than 100000 edits, less than 10 FAs, fails my criteria"), but I usually voice that in the support section. Outright misinformation ("candidate has 2000 deleted edits, so doesn't understand WP:GNG") must also be called out. But what we really should do is focus on the candidate, not the voters, so discussion about votes is often offtopic and should be moved to the talk page. Especially if it gets in any way personal. —Kusma (t·c) 13:19, 13 October 2017 (UTC)
• I agree that we do need to clearly establish when a vote can be struck, and by whom. In future we should try to avoid the back-and-forth striking and unstriking we saw here.--Pawnkingthree (talk) 13:27, 13 October 2017 (UTC)

• @Pawnkingthree: When get home, later on tonight, I will form some draft ideas. Others, may of course, add to the questions that need to be answered, or provide ideas relating to the questions I posed above. And I am meant to be on enforced wikibreak!😀 My name continues to not be dave (talk) 14:28, 13 October 2017 (UTC)

• Well, the conversation took sort of detour. Coming back to the original subject: i think presenting oppose to any vote should be welcomed. But attacking that vote/voter should be discouraged. If the vote is blatantly stupid, or obviously with grudge, then the policy of striking it out can be thought about. @My name is not dave and My name continues to not be dave: i think either only crats or admins should be allowed to do that. In case of blatant vandalism, anybody (already) can revert it. —usernamekiran(talk) 16:47, 13 October 2017 (UTC)

• PS: The phenomenon I am referring to happened with me in Ansh666's RfA. Mz7, and Ritchie opposed my vote in very civil manner. 16:50, 13 October 2017 (UTC)

• To be fair, my RfA was a bit different. Other than that one random "NO!" vote which ended up being struck, the opposes were largely understandable (if not reasonable; depends on who you ask) and mostly civil (again depending on who you ask), and it wasn't headed for unanimous support with no major concerns. In those circumstances, people felt less need to defend my honor, so to speak, and allowed for more reasonable discourse. Of course, Xxanthippe also hasn't responded to the furor over their oppose; staying and clarifying (or doubling down I guess) would probably have led to less heated discussion. Drive-by opposition, especially of such a controversial nature, should be discouraged.. ansh666 18:44, 13 October 2017 (UTC)

• Thanks for bringing this up, the situation was getting quite ridiculous. Saturnalia0 (talk) 06:41, 14 October 2017 (UTC)

Okay, I'm ready to give my full two pence:

What are the correct grounds to strike any vote?

What may be constituted as appropriate grounds to strike or unindent a vote -- which construes the vote not being counted in the vote counter at the top of the RfA, are:

• Any vote which unambiguously does not adhere to the relevant behavioural guidelines, such as WP:CIVIL/WP:NPA etc.,
• Any vote that is purely trolling or vandalism,
• Any voting where the same user has voted twice, or there has been evidence to establish that sockpuppetry has been involved. Discretion should be given as whether the first vote should stand.
• An oppose vote, where evidence has been asked for the voter's rationale, and it has not been given (within what period of time)??
• An oppose vote, which clearly misrepresents or misinterprets the candidate, broadly construed.
• Any vote that has a rationale which is patent nonsense, or not comprehensible.

Should explicit consensus exist to strike a vote?

• Consensus to strike a vote should be recommended practice when the vote is controversial. Otherwise, the vote can be removed by a 'crat, or, if vandalism, anyone.

What are the grounds in which discussion of a vote is acceptable?

• This has a few options to it that I have. The first, as recommended above, is to disallow threaded discussion of any vote. Discussion of a vote can be placed in the General comments section, or on the RfA's talk page. The second option is allow threaded discussion, but to prevent pile-ons or the Justice and Retribution team turning up with many pitchforks and burning sticks. Of course, discussion can be allowed that is around consensus for removing a vote.

Should the above be a binding guideline or policy?

• The current instructions, contained in WP:RFAV and WP:AAAD, are guidance only. They deliberately are that; the harshness of replies to certain opposes is new. There should be binding recommendations as to when votes should be removed, some of which are unwritten convention (vandalism and sock puppetry), and some not.

Yours, My name continues to not be dave (talk) 19:06, 13 October 2017 (UTC)

• As I see it, there are roughly two kinds of situations where an oppose !vote becomes a candidate for striking or unindenting. One is where it is just plain wrong, by leaps and bounds. When that is the case, and there are no additional problems, it seems to me that it should be left in place, but refuted by replies. Unfortunately, there can be way too many replies, but that can be dealt with by moving the discussion to the talk page. But instead of striking it, just rely on the Crats to recognize it for what it is. That's what they get paid the big bucks for. :) The other situation is where the !vote actually violates existing policy. If the violation of policy is such that it disrupts the RfA process, then strike it (or when necessary oversight it). If there is an oppose !vote that offends many editors but does not violate existing policy, then that's reason for a discussion about whether a new policy needs to be adopted, but not a valid reason to strike it. We don't need to make it more complicated than that. --Tryptofish (talk) 20:44, 13 October 2017 (UTC)

• I strongly oppose the striking of votes for any reason (If an edit is pure vandalism it can be reverted). People will support and oppose for their own reasons, if their reasons are good they will convince others, if their reasons are bad they will not. The striking of votes necessarily causes drama, and almost always for no gain: if an editor already has 200+ is support, why is it critically important that they have 0 instead of 1 or 2 opposes? Antrocent (♫♬) 20:51, 13 October 2017 (UTC)

  I know where you are coming from, but my RfA received nearly 200 supports, and I still didn't get the T shirt. Striking a few votes may well have made a difference to the outcome. Hawkeye7 (discuss) 22:23, 13 October 2017 (UTC)

• @Tryptofish: and @Antrocent: I'm quite happy with 'bad' votes to never be struck. If this shall be the unwritten convention, then that should be the case. It's more that people are fighting whether votes should be excluded or not, before the final count takes place. My name continues to not be dave (talk) 20:56, 13 October 2017 (UTC)

• /sees all of the "grownups" in the room and raises his hand/ I too am more than okay with not striking 'bad' votes. In the case of an absolute blowout, the one vote won't make an overall difference. In the case of a "close call", other factors will come into play (*WP:NotNow, etc*), so the one vote still won't make a difference. As much as some of these RfA's seem to turn into popularity contests, I seriously doubt that a candidate will see this "blemish" as some reflection on their ability to use the tools the job comes with. Will it be disappointing to see a ###-1 "score", maybe, but if you're that concerned with the final tally, then maybe you're looking at it for the wrong reason. In the situation that is driving this topic, I don't quite understand the editor in their reasoning, but at the end of the day, that's their opinion, and nothing I say or do will change that fact. Respect their (*maybe misguided*) vote, and the process will continue to be healthy, because all voices are being equally heard. Start striking votes that don't quite fit the expected norm, and you're now opening up the process to all kinds of drama that isn't needed in the community, in my opinion. - NsTaGaTr ^(Talk) 21:46, 13 October 2017 (UTC)

• This is my major gripe with the issue. It needs to be confirmed, however, that this should or should not be case. I shall continue no further, however, and I will start up an RfC on this matter, which shall confirm, going forward, whether the community finds it acceptable to strike votes that are not blatant vandalism or personal attacks. My name continues to not be dave (talk) 22:08, 13 October 2017 (UTC)

• We've all seen oppose vote reasons from the ludicrous ("Wikipedia is in maintenance mode, so we don't need admins anymore", "I'm trying to rack up 100 oppose votes", "I always oppose fourth or higher nominations") to the bizarre ("the candidate urinates in public"). But I'm not in favour of striking votes. If votes violate WP:NPA or WP:SOCK then they should be removed, not stricken. The crux of the problem is that oppose votes are intrinsically worth twice as much as support votes. Hawkeye7 (discuss) 22:40, 13 October 2017 (UTC)

• There would be less of a need to police "oppose" votes if they weren't given such weight in the universal RfX tracker (and /Recent). A single-digit oppose count is given a disproportionate amount of visibility with its listing opposite a three-digit support figure, triggering an immune response. This could be resolved by replacing the support/oppose/neutral count with a simple support percentage and total count. (not watching, please {{ping}}) czar 05:26, 14 October 2017 (UTC)

• The only acceptable case, as I see it, is if someone is voting twice (no opposition there I guess?) or if there is crystal clear, undeniable evidence of sock puppetry. Saturnalia0 (talk) 06:41, 14 October 2017 (UTC)

...the RfC has been moved here, per suggestion by Kudpung. My name continues to not be dave (talk) 07:59, 14 October 2017 (UTC)

 You are invited to join the discussion at Template talk:Editnotices/Group/Wikipedia:Requests for adminship#Include_instructions_for_using_Template:Rfan_in_the_RfA_Editnotice. Ahecht (TALK
PAGE) 13:48, 16 October 2017 (UTC)

This has been bugging me since a long time. "!vote" means negative/oppose vote. And "+vote" means positive/support. There is also an essay about this terminology but I cant find it now. It is evident that most of the editors think "!vote" = "vote". Well its not. Also, the use of ! as a prefix to vote, has its roots in computer programming. —usernamekiran(talk) 10:04, 14 October 2017 (UTC)

You're only partially right. Yes, the ! operator comes from computer programming - it's the negation operator, so !vote just means "not a vote" (i.e. not a straight poll), rather it's a consensus-building discussion where arguments are weighed, not just numbers. No, it has absolutely no relation to whether a "vote" is negative/oppose or positive/support at all. WP:!VOTE is what you're looking for. -- Begoon 10:15, 14 October 2017 (UTC)

The terminology/slang we have of "!vote" is supposed to imply a "nonbinding vote" or "soft vote" and not a "no" vote. It's meant to remind us that these pages are not treated as straight up numerical tallys. For example, back in the day, the place where articles were deleted was called "Votes for Deletion" or VfD. It was changed to "Articles for Deletion" in order to downplay anyone taking it as a straight "vote". -- Fuzheado | Talk 16:02, 14 October 2017 (UTC)

oh. So the ! operator still means negation, but used here in a different context. ie, "!vote" doesnt mean "oppose vote" but it means "this is not a vote, this is a personal opinion". Thanks Begoon, and Fuzheado. —usernamekiran(talk) 17:45, 14 October 2017 (UTC)

Typically it's used in two manners: either to say "I know this isn't supposed to be a vote but I'm framing my comment as a support/oppose vote anyway", or "I know this isn't supposed to be a vote, but we all know that the closer is going to treat the supports/opposes as a straw poll". It's an affectation that is only accessible to programmers, and personally I think it would be better to use "discussion" if you want to emphasize that this isn't supposed to be a vote, or "viewpoint" (or something similar) if you're expressing your support/oppose view. isaacl (talk) 01:14, 15 October 2017 (UTC)

Yep, I tend to use "comment" and "discussion" instead of "(!)vote". ansh666 02:06, 15 October 2017 (UTC)

The whole "!vote" thing is horrible jargon that should have never become common in the first place. It's serves as intimidating exclusive lingo that obscures the RfA process rather than making it more transparent. As Ansh666 points out, there's better ways to phrase it. We never vote on Wikipedia so calling our input a "comment" or a "support" or "opposite" etc is just fine. Jason Quinn (talk) 08:26, 15 October 2017 (UTC)

It is a vote though; it's only pretended not to be a vote because, as you say, "we never vote" here. So we call a vote something else. — fortunavelut luna(Currently not receiving (most) pings, sorry) 13:18, 15 October 2017 (UTC)

No, it really is not a vote. That it often functions as a vote is different from actually being a vote. If there are 10 well thought-out supports and 50 lousy "what found" type of opposes. Guess what? That person is still well supported as far as the closing bureaucrat cares. That's why people in RfA need to stop making a big deal out of nothing over ridiculous or even offensive opposes pop up. The closer knows how to treat them. Jason Quinn (talk) 19:01, 16 October 2017 (UTC)

Ignoring the extreme case where the opposes are obviously bad votes (e.g. "I oppose because I was paid to do so"), then the straw poll nature of RfA has been essentially codified with the explicit creation of a discretionary range in the support percentage. In theory, the bureaucrats could still fail an RfA where the straw poll result is above the discretionary range, but they'd have to have ironclad justifications. Which is why we get all of this back-and-forth arguing, because changing the minds of individual commenters can actually affect the result. If we really want to weigh the relative strengths and weaknesses of a candidate, then we should stop wasting time criticizing specific criteria, and spend more time discussing how the candidate's pros and cons should be considered in relation to each other. isaacl (talk) 19:27, 16 October 2017 (UTC)

I think this is mostly true, tbh. The few cases I can remember of RfAs being decided by something other than vote-counting were all very controversial - meaning, they were all below 70% support but still passed. And they were all a very long time ago too, around the time when "!vote" got stuck in the Wikipedian sociolect. By contrast I can't remember any RfA being above 80% support and failing.

And yes, they are discussions as well, but in most cases what decides the outcome is de facto the number of people in the support and oppose columns. – filelakeshoe (t / c)  20:05, 16 October 2017 (UTC)

Yep. But various people are going to continue being denialist about this indefinitely.  :-/  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  10:03, 17 October 2017 (UTC)

Does a candidate who withdraws when it's clear that their RFA is not going to succeed, rather than letting it run the full week to the inevitable failure, have a better chance next time, or does it make no difference? I cut my first RFA short when it turned clearly negative, I feel it was to my advantage when I returned for my second (successful) RFA, but what do the stats say? Roger (Dodger67) (talk) 08:48, 18 October 2017 (UTC)

• I have seen candidates who withdraw their candidacy when an RfA clearly will not succeed enjoy kudos of small favor in their subsequent RfA for having done so; and there's nothing wrong with positive recognition of such a decision. Unfortunately, I have also seen candidates endure a backlash of negativity for not withdrawing, in such a circumstance, and that recognition inevitably rest upon wrongful conclusions derived by assumptions of bad faith; in my opinion.--John Cline (talk) 12:22, 18 October 2017 (UTC)

  Note to self: Use the phrase "kudos of small favor" in casual conversation at least once this week... Roger (Dodger67) (talk) 13:19, 19 October 2017 (UTC)

  That doesn't appear to be grammatical, though (in two different ways). I'm not sure if the intent is "kudos for a small favor" or "kudos, in the form of being favored a bit". Kudos means 'thanks, gratitude, acknowledgement', or 'honors, praise, acclaim', depending on context, but in either case "kudos of" doesn't really parse in English.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  03:08, 20 October 2017 (UTC)

• I do think if a candidate withdraws early when their RfA is clearly not going to succeed, it's a sign of good judgement and respect for consensus. However I don't necessarily think the reverse is true. RfA is an election, and strange things tend to happen as elections proceed (ahem). Waiting it out is maybe just hopefulness, or maybe the candidate appreciates the feedback. Editors who attack a candidate for not withdrawing reveal more about themselves than about the candidate. Ivanvector (^Talk/_Edits) 12:27, 18 October 2017 (UTC)
• The key to that question is "when it's clear". I think RfA candidates typically quit sooner than they need to, perhaps out of shame. I wouldn't pull out unless the numbers were below the discretionary range after the first couple days. Any candidate running should know where they stack up. ORCP is good for that, too. Between the candidate and the nominator, they should know where the potential issues are. The only people that are surprised by the aggregate's reactions are editors lacking self-awareness and honesty. Headbomb's fourth RfA failed after it went the distance, as did his third RfA. GAB's first RfA was withdrawn after a couple days and the second was successful. Anarchyte's first RfA was withdrawn after five days and the second was successful. Clpo13's first was withdrawn after five days and the second was successful;Same with Primefac's first. However, Cyberpower678's first went the distance and closed no-consensus in 'crat chat before his second passed. If you look at that recent sample, the answer would seem to indicate discretion is the better part of valor. Chris Troutman (talk) 15:07, 18 October 2017 (UTC)

  That confirms my gut feel. Roger (Dodger67) (talk) 13:19, 19 October 2017 (UTC)

• Imho, it depends on how you handle either way, not which way you choose. Personally, I would see neither withdrawal nor letting it run as negative, as long as the candidate does show that they know that the request will fail and either now see it as a way to grow as an editor or as a waste of everyone's time that should be stopped sooner rather than later. But even a clearly failing RfX (like my recent RfB) can be a source of useful feedback for the candidate, provided they are willing and able to follow it. Likewise, a withdrawal can make sense if more feedback is clearly unlikely to happen and all new !votes will just be rehashing old ones. In the end, it boils down how the candidate has handled the failure itself, not the specific way they did it. Regards SoWhy 17:13, 18 October 2017 (UTC)

  I agree when the opposes turn into pile-ons without any new feedback it's time to pull the plug. Roger (Dodger67) (talk) 13:19, 19 October 2017 (UTC)

• The context is everything. As SoWhy says, either path can reflect positively on a candidate. It depends on the way a candidate goes about it, and their motivations. I think if there is significant opposition, it generally reflects better on a candidate to simply bow out gracefully, with a promise to address the issues raised and return in a few months for another try. That demonstrates maturity and a willingness to accommodate community concerns. On the other hand, if a candidate acknowledges their RfA is going to fail but requests that it be left open so that they can receive the feedback, that demonstrates the same thing and can reflect just as positively on the candidate. Swarm ♠ 18:12, 18 October 2017 (UTC)
• My withdrawal from my first RFA was probably motivated by embarrasment more than anything else, when the opposes turned into simple pile-ons, there was no further useful feedback. I even cringed when I read parts of it now, several years and a successful RFA later. My withdrawal statement was an attempt at a "graceful exit" even though I would have preferred to just slip out the back door as quietly as possible. So when I see a candidate stay with an obvious "crash and burn" all the way to the end I wonder if they are impervious to embarrassment, which, imho, is a rather undesirable trait for an admin. Roger (Dodger67) (talk) 18:42, 18 October 2017 (UTC)

There always seems to be a lot of talk about how RfA is broken, and it needs replacing etc etc, but I wanted to get some hard data to see if this was really the case. The most obvious thing to pick up on is a record of recently failed RfAs, and see if I can discover why they failed and if there was anything that could be done to prevent that. I had a look and made some brief notes:

Looking at all of those, there seems to be a common theme to the reasons each one failed. What I also discovered is the first couple of opposes are rarely picked up on by other people; the more substantial opposition tends to come later. The actual opposition that makes an RfA fail tends to on the whole be civil and constructive; I don't think Yash! got a fair deal, and Godsy was right on the knife-edge of support, but of all the others I think the result was probably correct and matched what I normally think the RfA bar is.

People say that too many people who could be admins don't want to run. I've had a look through emails over the past year and found the following comments:

• "I do not wish to take on this role at present"
• "I decided not to proceed any further in the process at that moment. I had a lot of real-life commitments to attend to, and things were still very much in flux."
• "Thanks for the vote of confidence, but I honestly think it's beyond me. In a lot of ways, it is like glorified babysitting and really, really stressful."

There doesn't seem to be much complaining about RfA itself at all - most of the discussion is me either deciding a candidate is not suitable or them saying they do not have time per above.

So what is the issue exactly? Ritchie333 ^{(talk) (cont)} 11:06, 19 October 2017 (UTC)

Reading Godsy's RfA, I don't think treating InfoWars as a source was the reason for it to fail. Behavioral concerns seem to be the main reason here (which was expressed in the crat chat as well). That said, I think the analysis shows that reasonable, civil, clueful editors have nothing to fear as the recent successful RfAs demonstrate (MLG, Ansh666, Cullen328, GAB, Anarchyte, etc. (Tony most likely as well)) and there are plenty of candidates that fit that bill. The comments you mention show less a problem with RfA and more one with adminship as well but unfortunately no change to RfA can change the characterization of adminship as "glorified babysitting" and "really, really stressful". No amount of RfA reform can change that... Regards SoWhy 12:01, 19 October 2017 (UTC)

• I'd say the pendulum has swung the other way and that the main problem currently is that people get frustrated when their opposition !votes are not listened to. For example, there may be 150 votes in favour of someone with rationales such as: 'Why not?' Then there are 20 or so oppose votes with in-depth genuine concerns for why the candidate is not suitable for the responsibility of the extra tools. At the end of the 7 day period, the bureaucrat comes along and closes the RfA as successful which subsequently makes what should not be a simple vote into what is effectively; a simple vote. However, on the other side of the coin, there are still individuals out there who just want to oppose for the sake of opposing and will find any old rationale for doing so. I think it's a good thing that RfA is more critical now than it used to be, on balance. Weeding out those who are hat collecting, those with unsuitable temperaments and those who would quite plainly make veteran content producing editors quit with their administrative actions is a critical responsibility of the RfA process. There are simple solutions to the issue as a whole that would mitigate the previous concerns and soften the process up significantly but implementing them requires convincing people of the need for change. Many people might agree that there should be change but if just 1 millimetre of a suggestion for reform is unpalatable, it ends up derailing the idea towards no-consensus. Solve that and you solve RfA. -=Troop=- (talk) 12:11, 19 October 2017 (UTC)

• My RfA is/was pretty drama free. I didn't mind the opposes in general, and the only thing that really bothered me was the implication that I somehow exhibited a pro-Catholic/mildly LGBT-unfriendly attitude in the way I handled new pages. I didn't get into it both because I didn't want to increase the drama that would come with those specific examples, and also I felt engaging on that specific topic would only lend the (IMO) false implication credibility. Of course, I was lucky enough not to have to deal with the opposes and ensuing drama that happened on the MLG RfA, and mine was relatively drama-free (the arguments on the talk page IMO were more about the concept of the draft space than me, so I didn't mind). TonyBallioni (talk) 12:42, 19 October 2017 (UTC)
• The problem with RfA is not always the results but rather the way in which people oppose. For instance, I would consider my RfA an example of why RfA is broken, in that multiple editors spent the week claiming I was a sock. That was a rather hellish week for me, and my activity dropped off for a bit afterwards because I wasn't the most motivated to do anything on the project. When people who would clearly be net positives as admins look at that, sometimes they decide just not to run. I know of at least one case at the moment where someone who would be a real asset as an admin is uninterested in running due to the RfA environment. ~ Rob₁₃^Talk 12:59, 19 October 2017 (UTC)
• My impression is that the (intentionally?) mean-spirited inquisitions of a few years ago have toned down a lot recently. I suspect it may somehow be connected to the significantly higher number of !votes in recent/current RFAs. When there were only a few dozen participants the inquisition fires burned hotter. My first failed RFA in 2014 had only 74 participants, the second successful one in January this year had 145. Currently the norm seems to be in the 200 to 300 participants range and the discussions seem far less acrimonious. Roger (Dodger67) (talk) 13:31, 19 October 2017 (UTC)

On RfAs like Wikipedia:Requests for adminship/Rickyc123, the discussion certainly seems to be more towards "moral support, thanks for your interest, have a go in the future" rather than chewing the candidate up and spitting them out. Indeed, if I had to pick the most snarkiest oppose in that RfA, it would probably be my own. Ritchie333 ^{(talk) (cont)} 14:11, 19 October 2017 (UTC)

(edit conflict) This is one of my arguments for why ORCP is more toxic than RfA itself and should be marked as historical. Just putting that plug in here. TonyBallioni (talk) 14:13, 19 October 2017 (UTC)

TonyBallioni I wonder if ORCP might actually be the reason why RFAs are less traumatic these days than before? Perhaps ORCP is drawing the fire away from RFA, with a significant proportion of unqualified candidates getting "roasted" there and so don't come to RFA at all. Roger (Dodger67) (talk) 20:35, 19 October 2017 (UTC)

Dodger67, I think it does that but I also think it scares a lot of good candidates who could easily pass. I know of at least one candidate who I think has an excellent chance of passing who was treated to a useless ORCP that could be used against them in an RfA. I also know that the advice that had been given is that you should aim for a 8/10 or higher average to have a shot at RfA. That is of course, ridiculous advice that makes people who get an average of 60% chance of passing think they shouldn't stand. I think more RfAs from candidates that range the spectrum is important. While everyone should want a 100% RfA, we need more people who are willing to stand on the chance that they will pass with 75%: not everyone is going to be perfect and get 99% or 100% support and that is okay, and we need to start letting potential candidates know that. Thus endeth my soapbox. TonyBallioni (talk) 02:56, 20 October 2017 (UTC)

• In more recent RFA’s oppose votes have been subjected to a higher degree of scrutiny for a variety of reasons both good and bad (calling out bias, disagreeing with policy interpretation, disagreements over activity in certain areas or lack thereof, I want to see this RFA pass with 0 opposes). That a well-qualified candidate usually has a number of people willing to defend them when opposes appear is a good thing for the candidate and eases their burden. There also appears to be an increasing trend towards “I think I might oppose but let me ask a question about my concern first” which helps the candidate further defend themselves in a visible area and not look like they are badgering the oppose votes. ZettaComposer (talk) 14:16, 19 October 2017 (UTC)
• When I ran a couple of years ago, I made the admittedly-stupid-in-retrospect mistake of letting a couple of editors here whom I'd had positive interactions with know about it, though in a glib "Hey look at this!" manner (I have an odd sense of humor sometimes). This was subsequently and perhaps predictably depicted in my RfA as canvassing, and led to a snowballing of opposition to the point that I withdrew my nomination. It chafed more than a little that some if not most of the opposers seemed to think my stupid mistake was instead an intentional attempt to sway the vote. Of course, noting WP:AGF at the time wouldn't have been practical or prudent. I do sometimes wonder how things might have gone for me if that had been a non-issue, and I might consider running again, but there's no way I'd nominate myself, and for better or worse nobody's come forward to suggest I run again. I also admit that while I think I would be an asset as an admin, I'm not entirely sure that anything I'm most strongly interested in doing here really requires the tools. I'd be curious to have a conversation about that with anyone who'd be interested in nominating me, but that's not an explicit request for such a discussion. DonIago (talk) 14:33, 19 October 2017 (UTC)
• For whatever it is worth, the numbers suggest RFA may be having a bit of a turnaround. Total and successful RFAs are already more than last year, and the rate of success is the highest in over a decade. That doesn't speak to the subjective experience of being a candidate, but statistically a plausibly qualified candidate (leaving aside the NOTNOWs) has been more likely to pass than to fail at RFA this year. --RL0919 (talk) 19:21, 19 October 2017 (UTC)

• There's a lot of ways to look at the data. Plus, there's lots of effects on the data as well. The statistical data set is also very small, so it's quite hard to draw conclusions from it. January of this year was quite the statistical blip (9 successful RfAs). If January had been a 'normal' type of month, the number of successful RfAs per month this year would be at the lowest level ever. Over the last 8 months complete since then, we've had a record number of 0 successful RfA months (3), and the worst February - September run of RfAs since RfA began (8 successful, vs. last year's low record of 10). I don't mean to be a downer, or to say that the RfA sky is falling. Rather, any statistical variations have to be taken with a very large grain of salt. --Hammersoft (talk) 20:19, 19 October 2017 (UTC)

• To answer the question in brief, the problem is a series of cultural and structural issues surrounding adminship and the RfA process. On the structural side, the standards have increased from 4000 edits / 6 months in 2007 to 10000 edits / 2 years today. Candidates are judged on one-off or rare instances of mistakes, rather than evaluated on their overall competency. On the cultural side, we demand a "need" for the tools, when we actually mean "potential use" - the stronger language helps lead to an increase in standards. Adminship has also become a status symbol, and admins are held to higher standards of behaviour than non-admins. All of this combines to create an environment where the tools are a sufficiently big deal, and where the process to get them is sufficiently difficult, as to prevent candidates from wanting to step forward.

As to why you get such vague answers when asking people why they don't want to be an admin, there is also a truly comical culture around here that you can't want to be an admin. I'm sorry to burst any bubbles here, but I would estimate that 99% of people who are admins wanted to be one. Maybe there is the odd one that had their account hacked and put up to RfA in 2004 that passed. But the notion that someone is around for a decade, has edited in all admin-related areas, and then just happens to stumble across RfA and accidentally submit their candidacy is ridiculous. Admins wanted to be admins or they wouldn't have volunteered for it, and there's nothing wrong with that. If you're concerned with people being power hungry, there are other ways to see that - hat collecting and always trying to win debates are two easy ones. -- Ajraddatz (talk) 19:50, 19 October 2017 (UTC)

I'm sorry to burst any bubbles here, but I would estimate that 99% of people who are admins wanted to be one. I'd take that bet. I was one that had to be "dragged kicking and screaming" to RfA, 2 1/2 years ago. When people first started suggesting it to me I responded "no, no, never!" It took a good 8 or 9 months, and massively flattering offers from several prominent admins, before I finally got up the nerve to run. I know the same is true of at least one other current admin whom I recruited a few months into my tenure - he initially refused to consider it before saying a month later "well, maybe..." - and several other people that I have urged to run that have so far refused. I think the reluctant admin, who gets talked into adminship rather than aspiring to it, is not as uncommon as you think. Don't get me wrong - I now enjoy contributing as an admin - but I think there are a lot of good admins who had to be persuaded to run. --MelanieN (talk) 21:58, 19 October 2017 (UTC)

I think this is an issue of wording, much like the word "need". We've twisted the word to mean something it doesn't. Whether others persuaded you to run doesn't change the fact that, after they suggested it, you wanted to perform the administrative duties. Nobody in any volunteer role is there unless they want to do it. I edit on Wikipedia because I want to; I'm not paid or coerced to be here. It's not a bad word. We've co-opted it to mean something nefarious, that the person is secretly planning to gain advanced permissions so they can do naughty things with them. But that's not what "want" means. What we really mean by "not wanting to be an admin" is that people should want to be an admin for the right reasons - helping people, doing boring maintenance tasks, protecting the quality of the encyclopedia. We currently associate "wanting to be an admin" with being power-hungry and not here to contribute to the project - just playing the MMORPG. But that's not what the word means, and if we can move away from treating any desire to help out in admin areas as bad, that will go a long way towards a more healthy culture surrounding advanced permissions.

I'm curious why you were reluctant to run. I think it probably has something to do with a) the standards expected of RfA candidates, and more generally, b) how admins are viewed and treated in the community. Adminship has become a very big deal now. If an admin makes a mistake, they are dragged over the coals for it. Adminship is very exclusive, because the vast majority of admins passed RfA years ago. But imagine how it would be if all you had to do to become an admin was email a bureaucrat once you got 1000 edits and had used a sub-admin permission without issue for 3 months (not a serious proposal, just a thought exercise). Imagine how it would be if admins were treated like every other editor, just with a few extra buttons, in the same way that someone with rollback or page mover rights is treated. Would the same stresses prevent you or others from volunteering yourselves? -- Ajraddatz (talk) 22:38, 19 October 2017 (UTC)

I think that an example might help get my point across here. Imagine that you are looking to volunteer at a local food bank, helping to sort cans. You are talking with the volunteer coordinator, and they ask if you want to do it. You say "nope, I have never wanted to do this". Do you see how strange that would be in the real world? The organization would probably direct you elsewhere, because they want people who want to be there. It is still possible to want to help at a food bank for the wrong reasons, like if you steal some of the food you are there to help sort. But the issue isn't with wanting to help, it's with why you want to help. Same deal with adminship here. When we associate any desire to help out in admin areas with having bad reasons to want to help, we develop a vocabulary that is separate from reality. And one that, I would argue, contributes to the unhealthy culture surrounding advanced permissions in general. -- Ajraddatz (talk) 23:00, 19 October 2017 (UTC)

If that's what you mean by "want" - that they were willing to take it on for whatever reason - then I wonder how the other 1% wound up as admins? 0;-D In my case (reverting back to the subject title here) it wasn't because "RfA is a horrible and broken process." It wasn't that I was afraid a bunch of nasty people would tear me to shreds. It was more a matter of self-doubt - fear that I might not the grade, might not be worthy. Personality, not process. As it turned out, my personal experience with RfA was not horrible or broken at all - to my surprise I enjoyed it. And back to the even bigger question here: one way to get qualified people to run is for us as individuals to make more of an effort to identify them, approach them, encourage them, and maybe eventually talk them into it. Personally I quietly evaluate a dozen or more possible candidates a month, although most of them wind up in my "not suitable" basket for one reason or another. --MelanieN (talk) 23:20, 19 October 2017 (UTC)

The other 1% probably got it way back when all you had to do was email Jimbo or whatever, and didn't really know what it was for. :D Then again I wasn't around for that time so I don't know if that's accurate or not. ansh666 23:30, 19 October 2017 (UTC)

I made an obviously bad attempt at humour, suggesting that the other 1% were accounts that had been compromised and taken to RfA without the original owner's consent. But you've got my point - what I mean by "want" is what the word actually means, a desire to do something for whatever reason. When we change the meaning of words here, it impacts how we think about the issues. When "want" means "want for nefarious reasons", then any want - good or bad - becomes bad. When "need" is used instead of "potential use", then anyone who would only use the sysop tools twice a day no longer meets that criteria, despite clearly having a potential use for the tools. When "notable" is used instead of "inclusion criteria", we attach a normative judgement to the quality of the subject of an article, when we are just trying to set a standard for what should be included or not.

Thanks for sharing your own reasons for not running. Your suggestion of identifying and contacting potential candidates is great. But I fear it is very complacent to how we treat advanced permissions here. The reason that Wikipedia was so revolutionary is because the most important right on the project - edit - was given to everyone. This allows people to make mistakes, because everyone else can fix them. Obviously administrator rights should be a bit more restricted than that, but any admin action* can be reverted in a similar period of time. So why do we have such a higher standard for admins? Now, I'm not suggesting that there shouldn't be a higher standard, but I wonder what the right balance is. We're on a website where everything can be changed and undone, but we're fixated on promoting admins who will never make mistakes. I would propose that moving towards a better balance, between expecting competence but not perfection, would help people who fear that they might not be worthy to step up to the plate. -- Ajraddatz (talk) 23:39, 19 October 2017 (UTC)

I first became aware of adminship around three years before I actually ran; in my case I'd had advanced privileges on other projects and sites all the way up to full root access on the server and had just "been there and done that" one too many times. I eventually concluded that since there were so many existing admins and that no admin is obliged to use their tools, that I wasn't going to end up in the same situation I've been in the past where I seem to be running the whole show in my spare time. Ritchie333 ^{(talk) (cont)} 10:48, 20 October 2017 (UTC)

As long as it remains a popularity contest it will remain broken. It really should be based more on objective measures of competence. This would remove the "you're going to fail because you pissed off an admin who has a bunch of friends and they're all going to bloc vote against you" problem, and also weed out "best behavior" near-noobs who don't have any enemies but also have insufficient experience to do the work properly. But, such a reform will likely not happen any time in the next decade.

What is more likely is that more tools will be unbundled. At some point we should get to where the only admin stuff remaining admin-only is "dangerous" things that involve deletion, blocking, and privacy/security matters. Some deletion tasks, maybe even all of them, could be offloaded, if the application process required sufficient clue. There are plenty of non-admins who know CSD better than the average admin, and plenty of admins make CSD mistakes, so the adminship bar is actually too low when it comes to CSD, while it is too high for many other tools (which is why we keep unbundling things like WP:Template editor, WP:Page mover, and WP:File mover). Another one that could be unbundled would be temporary page protection. It doesn't take a genius to figure that stuff out.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  03:21, 20 October 2017 (UTC)

I've been of the opinion that we've reached the limit of what we can safely unbundle, but you do have a point, SMcCandlish. The anomaly is that all the minor rights carry a clearly set entry threshold while adminship has none (bar the arbitrary criteria exercised by the voters) but candidates come under heavy scrutiny. Any of the unbundlable rights on the lines you suggest might not need an RfA style process, but they would require a high level of proven clue and trust. It would still leave RfA however, as that one place where users can be as nasty as they like - to both the candidate and each other - with almost total impunity. Kudpung กุดผึ้ง (talk) 06:19, 20 October 2017 (UTC)

But "we've reached the limit of what we can safely unbundle" is what most editors say, until we unbundle something else. The very fact the the requirements for getting them are technically higher (demonstrable experience, need, and clue) but socially lower (not a popularity contest) is why it works despite howls of naysaying and predictions of doom every time a new unbundling is proposed. I just outlined two additional things we can safely unbundle, at least partially and step-wise. Every time we do this, all the "sky will be falling" predictions turn out not to be true, and it works out just fine. The bare fact of the matter is that most long-term, active editors who do not have a history of blocks and topic bans can – if they have the requisite skills – actually be trusted with most admin-level activities, or they would not have survived here. They're certainly more trustable, as a practical matter, at these things than editors with 6–18 months experience and lots of smiles but little in the way of applicable experience and community "institutional memory" who nevertheless "earnestly" (desperately) want admin hats to wear. This is why, e.g., I voted for Headbomb despite his being a bit irascible (like me >;-) – he has a great deal of experience in the areas he wants to be an admin to be more effective at, and no interest at all in the "drama" processes where his patience and temperament might be strained or be a strain to others. Meanwhile, we have various near-noob editors who want to be admins and who have not pissed off anyone, but who have no idea WTF they are doing.

A radically alternative approach to adminship would be something like 100+ WP:NACs with 95%+ rate of not being overturned, and no [legit] blocks or bans within the last two years = automatically an admin. Make it an administrative judgement meritocracy instead of a popularity contest. Something sensible like this will probably also never happen, because the RfA crowd are addicted to candidates' sweetness, with little regard to competence. This is why I support the unbundling approach – just make the admin position less and less relevant, and increase the number of people actually able to clear out backlogs without needing golden magical wizard hat to do it, when a crumpled Template Editor tool bag and some grubby Page Mover overalls, etc., will get the job done.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  07:10, 20 October 2017 (UTC)

"It would still leave RfA however, as that one place where users can be as nasty as they like - to both the candidate and each other - with almost total impunity." But my research from the past year shows that in general, that's not the case - at least not to the candidate. Indeed, most of the biggest slanging matches at RfAs appear to come from experienced participants slugging it out with each other (and yes, I'm guilty of this) which generally have little effect on the candidate. Perhaps if we all let it go a bit (and I'm including myself in this) then that would be the most obvious way to improve the atmosphere. If somebody puts up a daft oppose that makes no sense, well as the old saying goes; "never argue with an idiot, they will just bring you down to their level and beat you with experience". Ritchie333 ^{(talk) (cont)} 10:48, 20 October 2017 (UTC)

I think RfA is better than it was, and I think the scrutiny of opposes is right and proper, especially when by those that almost always oppose. Sure, there are supports with no real rationale as well, but I believe that the watchlist notifications are making it more inclusive and democratic. I was surprised how easily my RfA went, after watching the drama for several years and being reticent to expose myself to the negativity. Peacemaker67 (click to talk to me) 11:30, 20 October 2017 (UTC)

@SMcCandlish: most of the recent unbundling has had the benefit of 2 factors: support by the community, and ease of technical implementation. The suggested items you have above don't enjoy the second factor - as they would require both new software programming and the decision to incorporate that to WMF projects. With community support creating a new group "deletors" that can "delete pages" is much easier then "delete pages in namespace x, that are less then y days old, with less then z contributors"; likewise assinging "change protection levels" is much easier then "change protection levels, only for articles, only for protection times <x time, etc, etc". — xaosflux ^Talk 13:25, 20 October 2017 (UTC)

That's what rules of usage are for. All sorts of things on WP are technically possible but not permissible, and the system is generally just fine. Not every restriction we might want is necessarily a technical limitation.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  00:41, 21 October 2017 (UTC)

Yup, administrative controls - I've suggested that as an option before, usually shot down as "just be an admin". These are still SLOWLY getting broken out, the most recent was EFH. — xaosflux ^Talk 01:18, 21 October 2017 (UTC)

Unbundling is admittedly the best and most practical solution to the unhealthy culture surrounding RfA and adminship in general. But I'm still not a fan; you describe adminship as the golden magic wizard hat, but it doesn't need to be. Being an admin has exactly as much "social status" here as we collectively give it. Adminship is put up on a huge pedestal, and further unbundling only makes the group more exclusive and unattainable - something which I think is contributing to its social status. The unfortunate truth here is that, as you say, for any experience editor the admin tools are very easy to use. The most difficult part on enwiki is interpreting the maze of policies and guidelines written by a small fraction of the community over years as they relate to using the tools. So why is it such a big deal? (Obvious answers: no easy way to remove bad admins, standards so ridiculously high at RfA that anyone wanting to help out in admin areas needs to "campaign" or do all the right things for at least a year, an expectation of perfection in admins despite us being on a site where everything can be undone)

I would be in favour of complete unbundling. Adminship can't have some ridiculous social status if it doesn't exist. Some devolved groups, namely those using delete and block, would probably require RfA-style requests. But we could establish standards for those, and have merit-based evaluation rather than the current type of free-for-all. -- Ajraddatz (talk) 18:20, 21 October 2017 (UTC)

• The problem with RfA is not just how horrid it is for those taking part, it's the far greater number of people who wouldn't even consider going near it, as a result. Andy Dingley (talk) 11:49, 20 October 2017 (UTC)

I still maintain that there's an aspect of not enough nominators actively looking for candidates. In the past year, of 37 RfAs - 26 of which had a nominator - 19 were made by the same 5 nominators (Ritchie333, me, MusikAnimal, and Kudpung, and There'sNoTime). When I made an effort to search out candidates, ask them about running, and nominate them, I was able to nominate 5 users in the space of a month, and I've got plenty of notes about other editors I'm either still talking to or am planning to ask later. I wonder what RfA would look like if, rather than having a daily thread where we write thousands of words about whether or how RfA is broken, we got out there and found editors who could do good work with the tools and just nominated them. Sam Walton (talk) 12:58, 20 October 2017 (UTC)

The question asked in the heading of this section is "Is RfA a horrible and broken process?" Horrible sometimes, but i'm not sure about really broken. In the last 100 RFAs, (say), how many of those new admins have since demonstrated they should not have been appointed, and how many who missed out have subsequently demonstrated that they should have been successful? I don't think the process is as "broken" as many people claim because it seems to be producing the right results. The unfriendly/unhealthy aspect of the process only exists because we allow it to exist. . Moriori (talk) 02:30, 22 October 2017 (UTC)

To somewhat echo was Andy Dingley said a couple of posts above, RfA producing admins we later need to desysop doesn't really seem to be the main problem. The serious breakage is that for every person who passes RfA (and, yes, most of those who pass should pass) we likely have dozens or more who will not even try because the process is too horrific and the outcome too uncertain because it's all subjective and personality-based. I.e., as long as RfA is a popularity contest and the primary concern is, basically, "has the candidate ever pissed off anyone who has friends", then we're always going to be short of admins we actually should have. Our community mistake is is making this a political process based on charisma rather than a meritocratic one based on competence.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  03:22, 23 October 2017 (UTC)

Lots of good points here, as someone who went through the RfA gauntlet and eventually withdrew at the start of the year I find myself mostly agreeing with those saying that RfA generally does get the right result, though sometimes not in the kindest of ways. Looking back, I probably wasn't quite ready in January - but some of the negative comments did perhaps result in me losing my enthusiasm for editing for a while afterwards. That being said, the vast majority of those comments were constructive, it was just the views of a few that I took somewhat to heart. I think the process itself isn't that broken, and if a minority of users rephrased how they put across their views on candidates, it'd be a far nicer place to be generally, and perhaps encourage more people to run. OcarinaOfTime (talk) 09:07, 22 October 2017 (UTC)

I'm probably going to step in it here, but...

As I noted in the above thread, when I ran for adminship previously I made the stupid (especially in retrospect) mistake of telling a handful of editors about it, because I got carried away in the moment. This effectively derailed my RfA, as concerns of canvassing came to dominate the discussion.

This bothered me a bit at the time, and I've come to realize it still bothers me a bit, so I wanted to try to verbalize my concerns. If nobody wants to speak to them, I'm cool with that. At least I'll have gotten it off my chest. I'm really not looking for an argument here, and I'm certainly not looking for any sort of apology or such.

• The way RfAs work seems to be at odds with most "election" systems I'm aware of. When someone runs for public office, it's typically widely-publicized. But the RfA system seems to encourage nominees to keep their candidacy secretive, on the grounds that a nominee could bias the vote. Even when an RfA is in progress, the announcement states that there's one or more RfAs in progress, but not who is running (that's likely a system limitation, and not one I'm especially concerned with). My point here is: do we really think that a candidate could, through canvassing, get so many editors to participate in an RfA that it would significantly bias the result? As I said, in my case I notified a handful of editors. Certainly if I'd notified, say, every editor I'd spoken to in the past year, that would be a different matter. If this situation is really a bit of a trap, just to see whether a candidate is aware of canvassing policies in general, then I guess that's something else, but it seems like something that could be handled by asking the nominator to speak to it rather than Opposing purely for that reason (unless we want to assume that no Admin has ever experienced their own lapse in judgment). I'd also note that because I was nominated by another editor, anyone watching my Talk page would already have seen that I had an RfA developing, which arguably would make my contacting editors redundant, though granted I have no way of knowing who actually does watch my Talk page.

• Even assuming I was attempting to contact other editors to bias the RfA (I would note that the actual messages I left at the time didn't contain any wording actually soliciting opinions, but I suppose that could be considered to be implied), there seems to be a failure to assume good faith here, if we believe that any editors I approached (and potentially editors who would see their Talk pages) would then speak well of me at the RfA solely because I approached them. Especially now, when RfAs are publicized, I don't think we have any clear way of determining what motivates an editor to speak to an RfA. Put another way, "canvassing" attempts may reflect poorly on the canvasser, but I have trouble with the idea that they actually have any net effect on the RfA unless we assumed that the canvassed editors also cannot be relied upon to exercise good judgment.

TL;DR - yes, if you're involved in an RfA, approaching other editors about it may indicate a lapse in judgment, but I'm not convinced that it can really shape an RfA except in the most grievous of cases, especially given that now ongoing RfAs are announced in any case (even if names are withheld). If an editor has been approached for an RfA than results may already be biased, as those watching the editor's Talk page will be aware of the situation.

Thank you for your time! DonIago (talk) 13:48, 20 October 2017 (UTC)

Hello Doniago. Just to put the perspective, the guideline exists to protect Rfa candidates as well – to prevent involved candidates from appealing to like-minded editors to oppose an Rfa (clearly, such attempts would be enough to tank Rfas of long-time editors who've had their fair share of disagreements in their tenure). Lourdes 14:06, 20 October 2017 (UTC)

Just to be clear, "approaching other editors about it may indicate a lapse in judgment" isn't the half of it. You'll be crucified as "canvassing" even if you did nothing but notify a couple of editors who opposed you last time, in a beyond-good-faith attempt to make sure your detractors get their say. I made this mistake years ago in my second RfA (when I was actually ready, willing, and competent, unlike in my first), and basically got lit on fire for it. I've never bothered to try RfA again, since I later realized I don't want the shackles that come with the "job".  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  12:08, 23 October 2017 (UTC)

• In a sense, yes, RfA is secretive. Another way to think of it; in some jurisdictions it is illegal to campaign within X distance of the polling booths (Ok, RfA is not an election, but the analogy holds a bit for this case, bear with me). All of Wikipedia is within any X distance of the 'polling booth' of RfA. Thus, some people may be quite sensitive to any hint of campaigning for that reason alone. I think what is more common though is that RfA voters look for any misstep to hand a negative vote upon.I read your notifications [1]. The tone of your notifications did not violate any policy. But, there's a couple of underlying things here. For one, how were these people selected for notification? Random? Prior positive contacts? The assumption people will make is that these contacts are people who will likely support your RfA. Certainly it is unlikely someone would contact someone whom they think will likely oppose them. In your RfA 3 of the 5 that contacted you responded, and all of them supported your RfA. I.e., these people were chosen for their likelihood to support you. Regardless of the wording of your notifications, it is self evident these notifications were canvassing. I.e., it violated the canvassing policy by way of WP:VOTESTACK, which is also part of that policy. There's a nice grid on that policy page at Wikipedia:Canvassing#Inappropriate_notification. Where your notifications failed was at the "audience" component. What is important to understand here is that people can not know your intent aforehand. What they see is your notifications. Is it a failure of WP:AGF that people assume bad faith in these notifications? Not really; again, who would contact people who would likely oppose them? It looks like canvassing on the face of it because of audience selection. A casual check [2] shows for example that you're one of the top contributors to the talk page of one of the people you notified. I.e., it's not random. So, with some mild digging the canvassing seems confirmed. The appearance of impropriety weighs heavy. That's true in real life, and it's true at RfA. Another aspect of this is that while contacting five people, plus inadvertently notifying the people who are reading those talk pages, there's another impact; dominoes. People are swayed by the presence of support from other users. The status of your RfA was 3-0 with the first edit by one of the people you canvassed [3]. 10-2 at the second [4], and then 12-4 right before the first notification of the votestacking (edit where votestacking was made known). Then it tanked. If the people you notified had not edited the RfA up to that point, it would have been 9-4; 69% vs. 80%. I.e., it would have fallen into the discretionary range, and possibly not pass. The support votes influence downstream people. Canvassing can influence the outcome, even if it is not readily apparent. People are rightfully concerned about this, and respond with the only tool they have at their disposal; not supporting the RfA. I seem to recall one RfA that tanked because a candidate placed a notice on their user/talk pages indicating they had an active RfA. I don't know or think that RfA demands candidates to be perfect, but any missteps that happen just prior to or during an RfA are going to come up. --Hammersoft (talk) 14:43, 20 October 2017 (UTC)
  • This isn't the first and I fear won't be the last RFA where this has been an issue. But I think that RFAs where the candidate is accused of advertising are rare and the advice to candidates looks good to me. The relevant advice is at Wikipedia:Guide_to_requests_for_adminship#What_RfA_contributors_look_for_and_hope_not_to_see ""Advertising" your RfA: Some editors do not like to see an RfA "advertised" by the nominee on other people's talk pages or on IRC. RfA is not a political campaign. The intent is to develop consensus. Impartial evaluation of a candidate is the goal, rather than measuring their popularity. Canvassing is generally looked down upon. Consider using {{RFA-notice}} on your userpage, which is a more neutral way to communicate your RfA to other users." I think the advice is still correct today. ϢereSpielChequers 14:06, 23 October 2017 (UTC)
    • I don't have any dispute with that, and I (hopefully made clear I) wasn't intending to contest how my own RfA evolved, even if it was a bit disappointing. I just wonder whether the advice is really effective. It seems to me that in this day and age, anyone who really was trying to bias their RfA via canvassing wouldn't do it nearly as transparently as I did. Maybe I'm giving an editor who would do such a thing too much credit? Either way, I guess there's no harm in leaving the advice there. I guess the core of my "grievance" is that it seems more like something a nominee should be trouted for versus having their entire RfA go down in flames with potentially concomitant damage. Anyway, thanks everyone for your thoughts on the matter. Happy to talk about it more, equally happy to move on to other things. DonIago (talk) 20:09, 23 October 2017 (UTC)

What if every editor with more than ten years and 50,000 edits, and an acceptable behaviour record of course, were simply given the mop, no inquisition, no debate. Here, take this mop, it's your punishment for being a good dedicated veteran editor. How many admins would we gain? How much damage would they do? (This is not a real proposal, it's just a "what if" thought experiment.) Roger (Dodger67) (talk) 08:05, 10 November 2017 (UTC)

Yo Dodger67, something of the sort at WP:FREQUENT, specifically this. Thing is, you'd ed up with characters like me :p on a serious note, would it affect for even-more-advanced permissions like oversioght, checkuser etc., which I think actually require a vigorous vetting procedure (i.e. what we see at RfA); if RfA wasn't vigorous, would that have legal implications? — fortunavelut luna 08:14, 10 November 2017 (UTC)

The problem is right there in your question: what is an "acceptable behaviour record" and how do you determine it without the inquisition? Once you know that, you can check WP:WBE and see how many people you gain. I don't expect they'd do a lot of harm. —Kusma (t·c) 10:04, 10 November 2017 (UTC)

plus 1 re difficulty of agreeing an acceptable behaviour record. I'd add that someone who does 49,900 edits in their first year and then averages 1 a month may not be an ideal candidate even if their only block is over 8 years old and they have edited in the last two years. More importantly though, adminship isn't suitable for every experienced editor. Some people don't want to risk the delete button falling into the hands of a deletionist, and others don't want the block button going to a member of the civility policy etc etc. Also people who've been here ten years aren't really the most important target group. The wikigenerations who we underrepresent among the admins are particularly those who started editing between January 2009 and the end of 2015. ϢereSpielChequers 14:08, 10 November 2017 (UTC)

• I endorse any attempt at reforming RfA voting in one way or another. I would support your idea. I've had a lot of 'what if' ideas like... What if everyone who voted at RfA were required to simply vote: "Support - per my criteria.' or "Oppose - doesn't meet my criteria." With the link actually linking to their set of criteria that they would be mandated to have on one of their user pages in order to vote. If people wanted to make an exception to their criteria then they'd have to give a reason in their !vote as to why they are deviating from their own criteria. I think this would reduce the 'Why not?' support votes and the extreme nitpick oppose rationales that are seen at many RfA's. It would also assist Bureaucrats in assessing consensus because it would be easier to see whose opinions are valid based on the sensibility of the criteria set. The process has been stagnant for quite a few years though so I would expect opposition on any change because people believe the status quo to be the best albeit imperfect solution. -=Troop=- (talk) 13:35, 10 November 2017 (UTC)

  Trooper1005 The problem of moving from a discussion to a straight up vote is that you lose the discussion. You lose the strongest point of RFA, the ability of those who actually assess candidates rather than question them to find the diffs that sometimes derail an RFA from someone who shouldn't pass. You also lose the dialogue that sometimes happens when people oppose on mistaken assumptions. As for the idea that the process has become stagnant and any change is better than the status quo, I agree that the debate is stagnant, you only have to look up one section to see that any change will be opposed - even by people who admit it will make the project better at nil or negligible cost! If you are thinking of RFA reform in general I'd be interested in your views on User:WereSpielChequers/RFA reform where I've tried to list all the main reform ideas and most of the minor ones. ϢereSpielChequers 14:13, 10 November 2017 (UTC)

• It would likely be completely harmless, as long as one thing changed: ANI or AN had a the power to desysop, not just ArbCom. ArbCom moves too slowly to deal with suddenly thousands of new admins, some percentage-small but head-count-high number of whom would be boneheads, jackasses, or PoV-pushers.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  17:05, 10 November 2017 (UTC)

You'd like desysops handled by ANI? Ok - that could be "fun". -- Begoon 17:26, 10 November 2017 (UTC)

We'd lose them as soon as we made em. It can be enough of a mincing machine as it is, let alone one that's on a conveyor belt! — fortunavelut luna 17:43, 10 November 2017 (UTC)

Well quite.SMcCandlish isn't wrong about the "boneheads, jackasses, or PoV-pushers". -- Begoon 18:01, 10 November 2017 (UTC)

It's wouldn't work well for today's adminship system, but in an alternate-reality Wikipedia, where virtually everyone became an admin if they stuck around, it would be the only practical way. It's how we handle everything else. E.g., every editor (in our reality too) is entitled to edit, and to edit when they want, and to edit what topics they want, and to talk with whatever other editors they want, and in whatever wording they choose. When this goes badly awry, ANI (or another venue, like AE) can impose blocks, community bans, topic bans (and move bans, etc.), interaction bans, and restrictions like prohibitions against incivility/attacks with escalating sanctions. In this Alt-WP, one's auto-granted admin bit would be just one of numerous editorial privileges one takes for granted and is left alone to use – until one becomes a pain in the community's backside.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  19:19, 10 November 2017 (UTC)

Yeah, I take your point, but Iri's observations about whether WMF could even allow this seem crucial. Plus, it seems it could potentially make people like you and me sysops of alternative-reality wiki, depending on the exact "parameters", and I'm certainly opposed to me. -- Begoon 13:18, 11 November 2017 (UTC)

Yo Begoon... Suport + Opose = Suppose... ;) — fortunavelut luna 13:48, 11 November 2017 (UTC)

At my age I mostly prefer repose. -- Begoon 13:53, 11 November 2017 (UTC)

• Continuing on this hypothetical, would I get the mop if and only if I ran AWB for a week? power~enwiki (π, ν) 17:53, 10 November 2017 (UTC)

  If you made 40000 AWB edits in a week you'd probably get a block :D — xaosflux ^Talk 17:59, 10 November 2017 (UTC)

  Maybe two weeks? I've seen talk-page categorization runs of about that size and time. My point is that "edit count" is a metric that is easily-gamed, and "editing tenure" only slightly less so. power~enwiki (π, ν) 18:03, 10 November 2017 (UTC)

  Separately, I'm not sure I agree with the "Wikipedia needs more admins" line. The chronic backlogs at WP:AfC and WP:GAN are far worse than any of the admin queues; and I haven't seen any severe AIV backlogs recently (while "quick" for most things on Wikipedia is 1-3 days, even an hour wait for obvious cases at AIV is painful). power~enwiki (π, ν) 18:03, 10 November 2017 (UTC)

• I'll be the party pooper who makes the obligatory comment whenever this is proposed; it's not a discussion that it's worth having, since for legal reasons under the California law under which the WMF operates there needs to be an approval process of some kind for the admin bit, otherwise the WMF will lose §230 immunity and there are literally no circumstances in which they would allow that to happen. ("The Wikimedia Foundation will not endorse this suggestion and will not implement it in the unlikely event that it should reach consensus. For legal reasons, we require RFA or an RFA-identical process for access to certain tools (deleted revisions among them)." if you want chapter-and-verse.) You and I have relatively limited assets, and aren't worth the effort of suing; a company with an annual revenue of $80 million plus is far richer pickings, and the WMF would immediately be buried in lawsuits. RFA is with us to stay come what may; the only issue up for discussion is what form the "community review" takes. ‑ Iridescent 18:13, 10 November 2017 (UTC)

• Also, @Dodger67 who decides what constitutes an acceptable behaviour record? You? A star-chamber which is either populated by elections with all the hassle and bad feeling that entails, or a self-appointing group which will by definition be filled with the blowhards and fuckwits who like to hang round on drama boards bitching about everyone they ever encountered who didn't agree with them on something at some point in the past? Or the entire community, based on a process in which the prospective admin explains why they think they'd benefit from the admin bit and anyone who cares to offer an opinion then gets to support or oppose over a period of a week, at the end of which one of the 'crats weighs both the numbers and the arguments and decides whether there's a consensus? ‑ Iridescent 18:21, 10 November 2017 (UTC)

• 50,000 edits don't even guarantee that someone won't be banned, let alone that they'll make a decent admin. Darius Dhlomo, probably Wikipedia's most prolific plagiarist, had more than three times that when he was indeffed for massive copyright violations. Granted, he only had five years' tenure rather than ten, but IMO that doesn't constitute much of a safeguard. The only thing potentially preventing adminship from getting into very unsuitable hands under this is the "acceptable behaviour record" standard, and for that to fulfill this function it'll have to be something which looks very like RfA. Hut 8.5 19:25, 10 November 2017 (UTC)

Which basically brings us back to square one: RfA is already mostly a "acceptable behaviour record" test, so the only thing that would change is that competency would no longer be checked for. I don't know if that's really a good idea though... Regards SoWhy 12:00, 11 November 2017 (UTC)

The big issue is that RfA is more a "completely spotless behavior record" test. It's easy to find one example of an editor making a mistake among a general pattern of thoughtful, useful, and civil contributions. Everyone has at least one diff they aren't proud of, perhaps on a bad day or when they rose to the bait. All it takes is that one diff to oppose an RfA, and then it gets picked up and magnified by others who don't bother even looking at the record to see all the good a candidate has done. ~ Rob₁₃^Talk 04:05, 12 November 2017 (UTC)

• To be honest, I think that for RfA reform, we need to enact reform in other places. If we can really change the feeling of the drama boards, then we should be able to change how editors think. But, I do think that there is one reform we could implement. Specifically, it would be nice if comments (not votes) could be removed if the crats agree that it is inappropriate. RileyBugz^会話_投稿記録 14:20, 13 November 2017 (UTC)
  • RileyBugz, they already can. 'crats are just loathe to do it. TonyBallioni (talk) 23:41, 14 November 2017 (UTC)
• Great idea but for me I'd prefer all RFAs to go on the record, Any editor can go rogue and start blocking anyone and everyone ... ofcourse this can happen anyway but the chances are very very slim, I do agree RFA does need reforming but personally I don't think giving the mop freely is the way to achieve this. –Davey2010^Talk 23:03, 13 November 2017 (UTC)
• Giving the mop to people with a decade of tenure and 50,000 edits is almost what RfA does these days anyway. Real reform would involve a) recognizing that most of our current admins were admitted with under 10,000 edits and less than a year of editing experience, and b) stop clinging to inflated numerical standards that so obviously don't reflect whether an individual would use the sysop tools well. We need to stop systematically excluding newer users, not reinforce that by formalizing the "only people who have been here forever can be admins" rule. However, there is one idea presented here that I really like - RfA as a process solely concerned with evaluating the behavioural suitability of a candidate for adminship. That's a decent proposal for reform right there. Decide ahead of time what we want the numerical standards to be and fix those so they stop increasing. Then, RfA would just be about evaluating the candidate's behaviour. That could be a significant step in the right direction, if appropriate numerical standards were set. -- Ajraddatz (talk) 23:18, 13 November 2017 (UTC)
  • Ajr, while I have some of the loosest RfA standards out there, I do have a rough numerical number in my head, and that is because I expect to be able to see how you interact with others before getting the bit. To be blunt, the tools are easy to learn (its basically advanced Twinkle), but what I actually care about is that someone isn't a jerk and doesn't go power hungry or rush into drama once they get the bit. If I don't have something to go off of, I'm not comfortable supporting. To me, this is a much larger concern than the desysop rate. TonyBallioni (talk) 23:37, 14 November 2017 (UTC)
    • Yes, there definitely needs to be some numerical standard. I would generally peg the absolute minimum here at 2,000 edits and 6 months of experience - I don't think anyone could learn our maze of core policies and guidelines in less time. The issue is that people look at random arbitrary statistics (how many articles created, percent edit summaries, voting record at AfD) without actually looking at how the editor interacts with others or understands our policies and guidelines.

Imagine the following system instead. We agree as a community that any candidate for adminship must have 2,000 manual edits and have been here for 6 months. We then set a series of competencies that we expect to see in admins: interpersonal skills, accountability, and the ability to find and understand our community policies. Instead of the current question free-for-all full of gotcha questions, we could have 1-2 questions for each of these competencies. "Show us an example of how you maintained a cool head during a tough situation." "Here is a scenario of a new user making a certain type of edits, is this vandalism? Please justify your answer based on existing community policies and guidelines." These questions could be tailored to the candidate and the areas they want to work in, or be totally general. Or we could have a set of pre-made questions depending on the field they want to be active in. If vote reasons were then required to be related to the candidate's meeting or not meeting of the competencies, rather than whatever the opposition or supporters feel like, then we might have a functional process. This isn't a new idea at all, but I think moving in this direction is the best way we could fix what is currently broken. -- Ajraddatz (talk) 01:33, 15 November 2017 (UTC)

Regarding "the field they want to be active in" - I think these are good types of questions - if you want to be involved in AfD - are you ready to close AfD's and delete things is perfect -- however I also consider a broad will this candidate take the time to learn how to properly (both from a community standards pov and a technical pov) use any other admin function before they start using it in the future - because most candidates don't vow to never use the other tools - so yes you could be the best afd closer, but if you are likely to just slam changes in to the abusefilter without clue then you're not getting my admin support. — xaosflux ^Talk 01:53, 15 November 2017 (UTC)

That's definitely a good point. If we had two questions on the "finding and applying policy" competency, the other could be a scenario-based question in an area the candidate is unfamiliar with. Not a gotcha question, like asking about some incredibly obscure procedure like when to use Special:MergeHistory versus delete-move-restore, but something that would require the candidate to do a little research into an unknown field. My experience with advanced permissions is that these competencies are pretty generalizable. If an admin can find and properly apply policy in one area, then there's a good chance they'll be able to do so in others. But this does require them to have some general project knowledge, and there is no harm in testing that rather than assuming. -- Ajraddatz (talk) 02:21, 15 November 2017 (UTC)

I have a whole library of these sorts of questions, but there are some current admins who are clueless about our policies and procedure who really, really don't like them. Hawkeye7 (discuss) 04:46, 15 November 2017 (UTC)

I'm not really that interested in candidates who can look up the relevant policy during an RFA. I'm more interested in whether they follow policy under normal editing conditions. ϢereSpielChequers 09:20, 15 November 2017 (UTC)

My concern about the Q&A section is that it probably distracts people from assessing the candidate. I'd prefer a rule against questions that aren't supported by a dif from the candidates editing, that way at least the questioners will have reviewed at least some of the candidates edits. ϢereSpielChequers 08:55, 15 November 2017 (UTC)

Fair point as well. That's why I would want to limit the questions to one or two per competency. Done correctly, questions should assist with a review of a user's contributions, not detract from it. But questions should certainly not be a replacement for actually looking through a candidate's contributions. -- Ajraddatz (talk) 18:58, 15 November 2017 (UTC)

If that were per competency in the area where they have indicated an interest then fine, but it should be possible to support them with diffs. Competency questions in areas where the candidate hasn't been active and doesn't intend to work are a bit of a waste of time. All the candidate has to do to ensure it won't alter the RFA is to start their answer "I haven't been involved in this area and have no intention of getting involved in it let alone using the tools there, but "[puts policy in own words]". Even if you get the policy a bit wrong it isn't likely to sink an RFA if you make a policy mistake in an unfamiliar area. ϢereSpielChequers 19:55, 15 November 2017 (UTC)

• The OP's "what if" is a non starter. We would never have the capacity to track, trace, and revert all the errors they would be making. RfA is also one way for the candidate to obtain the required level of trust for the job. Or do we want 1000s more admins using their tools for paid editing? Kudpung กุดผึ้ง (talk) 02:35, 27 November 2017 (UTC)

I agree. The thought of hundreds of unscreened people being given the unrestrained power to block other users and delete articles is absolutely terrifying. To me the most important comment people make at RfA discussions is "I trust them with the tools." There are a lot of people who have the suggested tenure and number of edits for whom I could not say that. --MelanieN (talk) 18:57, 27 November 2017 (UTC)