Wikipedia talk:Requests for adminship

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search
RfA candidate S O N S% Ending (UTC) Time left Dups? Report
RfB candidate S O N S% Ending (UTC) Time left Dups? Report

No RfXs since 23:01, 2 December 2019 (UTC).—cyberbot ITalk to my owner:Online

Current time: 18:36:43, 10 December 2019 (UTC)
Purge this page

Recently closed RfXs (update)
Candidate Type Result Date of close Tally
S O N %
Dreamy Jazz RfA Successful 2 Dec 2019 163 12 8 93
EvergreenFir RfA Successful 15 Nov 2019 252 42 5 86
ToBeFree RfA Successful 10 Nov 2019 182 3 3 98
GRuban RfA Withdrawn 9 Nov 2019 101 62 15 62
Girth Summit RfA Successful 26 Oct 2019 202 2 2 99
Kees08 RfA Successful 14 Oct 2019 144 6 6 96
Greenman RfA Unsuccessful 12 Oct 2019 104 66 9 61

Questions at RfA[edit]

I started a short lived thread above last month. I said the issue surrounding RfA is a perennial one, and while it is clearly one of the reasons why potential candidates won't come forward as was pointed out to me once again today, to save digging up the topic all over again, (new) regular subscribers to this talk page might just want to spare just 5 minutes and read this thread from nearly nine years ago! Kudpung กุดผึ้ง (talk) 12:34, 24 October 2019 (UTC)

Kudpung: which thread? Several address questions. –xenotalk 12:52, 24 October 2019 (UTC)
Sorry, my bad. It starts here and then through the threads that follow on about questions. It could have been written yesterday - plus ça change, plus c'est la même chose. Kudpung กุดผึ้ง (talk) 13:01, 24 October 2019 (UTC)
@Kudpung: not sure if I'm missing part of your question? The thread you linked to above is to the topic "Limit questions to one per user" from 2011. That topic has been addressed much more recently, in the 2015 RfC which decided (with a !vote of 82 to 24) on a "limit of 2 questions" per editor. Is there a new discussion post that RfC that needs following up on? — xaosflux Talk 14:30, 24 October 2019 (UTC)
  • I don't feel the volume of questions has been egregiously bad recently. And while we've had the occasional bad question, as a countervailing positive, questions seem to be moving more towards reviewing and providing reasoning for past actions rather than bizarre hypothetical tests. I'd be interested to know where questions are in a list of reasons why candidates don't apply. General hostility (rare for 1st time candidates) and the general discomfort of opposes based on character or competence (somewhat the sine qua non of RfA, alas) seem much more likely, but I fully concede I don't have a big data sample. Nosebagbear (talk) 14:20, 24 October 2019 (UTC)
I have a theory that the number of questions is more a function of time than anything to do with the individual candidate. Double or treble the number of RFAs in a month and the number of questions per candidate will fall sharply, especially those questions that aren't specifically relevant to the candidate. If I'm correct the number of questions is a symptom of the lack of RFAs rather than a cause. Fix the problems that deter lots of qualified candidates from running, and the problem will recede. ϢereSpielChequers 14:37, 24 October 2019 (UTC)
  • The idea is to read to the end of the page I linked to. It's got a lot more to do with the overall issue of questions than success of Biblioworm's December 2015 RfC that limited the questions to two per user. It addresses the entire philosophy of the questions and the ill spiritedness ( or even blatant trolling) that often lies behind them. Kudpung กุดผึ้ง (talk) 14:39, 24 October 2019 (UTC)
The trolling cannot be blatant or it would be obvious. ——SerialNumber54129 14:52, 24 October 2019 (UTC)
The trolling is obvious, Serial Number 54129, maybe you just don't see it. Kudpung กุดผึ้ง (talk) 06:10, 25 October 2019 (UTC)
I only wish that were true. ——SerialNumber54129 09:06, 25 October 2019 (UTC)
  • (edit conflict) To me it's not clear at all that the questions are what holds people back. After I passed I made a few serious pitches to people to consider RfA. One of them was Girth and obviously he said yes. The rest who declined all mentioned elements of their editing that they thought would prevent them from passing - that is draw too many opposes. None mentioned questions. To the extent questions came up, it came up with Girth around the timing of his RfA. He needed to do it at a point when he knew he'd have time to respond. That is a hurdle but feels like a logistical one rather than a foundational one that stops people from running. I'd love if some social scientist (ping Piotrus as someone who has studied sysop on Wikipedia before) would consider a study to find out about the experiences of those who go through RfA and those who are interested and don't (ORCP) or who people suggest should run and don't. Absent that we're all just using our experiences which might or might not add-up to an accurate picture. Best, Barkeep49 (talk) 14:54, 24 October 2019 (UTC)
This would be a fantastic thing to actually have more some more rigorous data on - I'd very happily support any such attempt Nosebagbear (talk) 15:28, 24 October 2019 (UTC)
Nosebagbear, sorry, but that's not possible. Canvassing for candidates obviously takes place off Wiki and convention stipulates that we cannot reveal names and emails. Kudpung กุดผึ้ง (talk) 16:39, 24 October 2019 (UTC)
While obviously that is accurate, I've seen a fair few bits of canvassing done on-wiki (including responses on why that can't/won't run), though obviously the number through email might be so high that its absence makes the rest of data too incomplete for benefit. Nosebagbear (talk) 21:07, 24 October 2019 (UTC)
There is more than one way to skin this particular cat. If someone wanted to do research that included people who had been asked off wiki and declined, you could either post a link to a survey here or ask active nominators to send such a link to people who they had approached. Finding out how many people had declined such a nomination or getting a statistically representative sample of them might not be possible, but giving a subset of those people a link to a research survey is fairly easy. Alternatively one could do research to a whole swathe of editors who haven't run but who have been active in recent years and have some other similarities to successful RFA candidates. One of the questions could be whether they have been approached by a possible nominator. ϢereSpielChequers 12:47, 25 October 2019 (UTC)
  • @WereSpielChequers, Nosebagbear, Serial Number 54129, and Barkeep49:, I do very strongly suggest first reading the archive I linked to before jumping to any conclusions. That's why it was pointed out to me earlier today and why I posted it here - for the benefit mainly of those who don't have the institutional memory, and to avoid perennial repetition in the ever vague hope that some day something will be done about it. That said, I can assure you all as a former long time talent scout that the major reasons I was given by dozens of potential candidates of the right calibre for not running were the questions and general hostility. There was a time when an RfA talk page contained only a copy-paste of the candidate's edit count and nothing else. The talk page nowadays is nothing less than shameful and is a sad new trend that reflects the increase in trolling. Let's hope nobody is called Richard. Kudpung กุดผึ้ง (talk) 15:06, 24 October 2019 (UTC)
    Kudpung, 2011 was obviously a banner year for new sysops compared to now (there nearly were as many successful RfAs when that discussion started as there were all of last year), but was also a huge drop-off from the preceding year and was part of a trend of a substantial % drop in new sysops year-over-year stretching back to 2007. I agree with you that we should be having more than 20 or so sysops a year (the general number since 2014) but I don't think we can say, with evidence, that it's questions keeping people who would pass away from going. It might be. We just can't say that it is. Best, Barkeep49 (talk) 15:16, 24 October 2019 (UTC)
Barkeep49, I wouldn't say 2011 was a banner year. If anything it was the year signalled the start of what was to be a rapid decline. The point is however, that the behaviour on RfA was already rotten to the core. What has changed is that the majority of the (now rare) RfA today conclude with a large consensus to 'promote', but only because we nip the non starters in the bud. Kudpung กุดผึ้ง (talk) 16:33, 24 October 2019 (UTC)
Kudpung, I agree that 2011 wasn't really a banner year - only a banner year compared to now. Arguably no year was a banner year after 2007. My point, however, is that 2011 was a different environment than today. I certainly had read that discussion and think it important. But I still respectfully, especially given the experience you mention below, disagree. Best, Barkeep49 (talk) Barkeep49 (talk) 16:39, 24 October 2019 (UTC)
  • @Kudpung: - I did read the archive set, including all of "How big a problem is the question section?" and the interesting "So... Are the optional questions optional?". I'm still inclined to agree with Barkeep49 in terms of what I've seen for the small group I've discussed it with - which is why I'm interested in your experience based on a much larger sample group. In terms of iffy questions, I would say that it just needs continuation of a point raised in the "optional" questions thread - candidates can't really deny answering a question, so it's up to everyone else to indicate that it shouldn't be answered. That's what happened to a very early question in mine (which was more struck in terms of "damned if I/he does, damned if he doesn't/irrelevance grounds) and also would have happened in the case of a question (as opposed to opposes) on naming basis. Nosebagbear (talk) 15:28, 24 October 2019 (UTC)
  • This discussion isn't going the way I had hoped at all, ist it? Rather than challenge the empirical experience of those who really do know, such as MelanieN, Ritchie333, and myself who for several years worked as a team to find victims for RfA, and the mass of research into voters and questions that was done at WP:RFA2011, I would have thought that once having read the threads I linked to and finding that the situation has worsened to the point of getting out of hand, someone could come up with a key phrase that could be the incentive for a new round of reforms. When Jimbo described RfA as a 'horrible and broken process' he was not wrong, even if RfA does actually do what it is supposed to. But it should be able to do that without all the silly questions, disingenuous oppose votes, bickering and in-fighting. Kudpung กุดผึ้ง (talk) 16:23, 24 October 2019 (UTC)
    Kudpung, I'm familiar with the arguments that have run at RFA over the years, and like you I have spoken to many, many potential candidates over the years, including ten who became admins. I'm used to people turning me down because they have skeletons in closets, because they fear that RFA is a brutal process that would make for a difficult week, because they don't need the tools, and because they don't fancy being at the community's beck and call. If anyone is specifically fussed about the questions, it really isn't as common a grief as any of the ones that I mentioned. Note to any watching potential candidates, some of the reasons I mentioned are true for some people, and at least one of the reasons is bogus - I get maybe one or two requests for admin action a month. ϢereSpielChequers 16:34, 24 October 2019 (UTC)
WSC, I'm not saying that they were all specifically fussed by the thought of the questions, but it was one of the many reasons they came up with for not running. The point I'm making here is that the questions are only one strand in the evil of RfA and it's one that could begin to be addressed. 'Baby steps' - you told me that yourself many, many years ago when you were unwittingly my mentor ;) Kudpung กุดผึ้ง (talk) 16:49, 24 October 2019 (UTC)
Agreed re baby steps, but to my mind the next babystep would be to target that watchlist notice to editors who are st least edit confirmed. There are some like yourself who would like to go further, but I'm not seeing anyone disagreeing with that change. ϢereSpielChequers 18:09, 26 October 2019 (UTC)
  • I'd rather have stupid questions than stupid oppose votes. At least in the former case we get to see how the candidate will respond, whereas in the latter case the convention is for them to stay out of it. – bradv🍁 16:37, 24 October 2019 (UTC)
    I apologize if my drive-by comment misses the point of this thread. But I do think this is broadly applicable to the idea of placing further limits on questions at RfA. The ability of the candidate to answer questions (and the wisdom to know which ones to ignore) is a make-or-break factor in many RfAs, and may be the best measure we have of whether the candidate is suitable. – bradv🍁 17:22, 24 October 2019 (UTC)
    One of my concerns about RFA is the overreliance of many voters on the question section. In RFAs we have plenty of time to assess a candidate's edits, look at how they communicate with others and especially how they perform in admin related areas. Do their deletion tags and arguments show they are ready for the deletion button? When they warn others and file AIV reports do they show themselves ready for the block button? Sadly most RFA voters skimp on this process, but a well argued and dif supported oppose or a dif supported question can sway a large proportion of those editors who, to be frank, don't research the candidate themselves and rely on others to do that and show their results in a question or vote. ϢereSpielChequers 17:37, 24 October 2019 (UTC)
    @WereSpielChequers:, while that's accurate, the problem is that if you didn't have questions, people would still be swayed by the few opposes who did do that research. With the questions, they get a better chance to defend/explain themselves. Nosebagbear (talk) 21:10, 24 October 2019 (UTC)
    Yes, responding to oppose votes is a trickier issue than responding to questions. I don't see anyone arguing for the abolition of the question section, but there are several, including myself and Kudpung who want it reformed. The question is how does one reform it and in what direction? I know the direction that I would like it to go in - "dif supported question"s that are the result of a questioner looking at the candidate's edits and wanting to clarify something or highlight an issue in a way that the candidate is expected to respond to. I'm less sure as to how one achieves that other than by example and discussion here. Others have different concerns re the question section, some of which I may well share. As for many RFA voters not doing research, we aren't short of RFA voters at present - in fact some record turnouts have recently been achieved. I want more voters to trawl through the candidate's contributions. But there is also an important role for voters who look at evidence and rationales provided by others, and decide whether it is germane and convincing. That role is not made easier by cluttering the question section with questions that could equally have been asked of any candidate. ϢereSpielChequers 07:34, 25 October 2019 (UTC)

@Barkeep49: Tnx for the ping, this will be a quick reply as a have to run to work. A research study would be valuable, my gut feeling is that this has to do with the combination of 'we get fewer editors' leading to 'most candidates have much longer histories than 10 years ago' than in turns is coupled with mud sticks (also see mini-essays of mine there 'why perfect admins don't exist' and 'mud sticks again'), which means that as time goes, new candidates find our expectations higher and their record easier to nitpick. And there is a common stereotype that RfA are super stressful, lot of people decline likely because they don't want to spend a ~week answering variations of did you stop beating your wife... --Piotr Konieczny aka Prokonsul Piotrus| reply here 03:58, 25 October 2019 (UTC)

  • Notwithstanding the comments by Nosebagbear and WereSpielChequers who make some valid observations, this is not a proposal to dispense with user questions entirely, but many of the questions are irrelevant even if it is claimed they are necessary to see how a candidate reacts under stress and stupidity. New, and up-to-date data as suggested by Piotrus is indeed required along the lines of this research where we documented and categorised certain types of over 100 inappropriate questions out of over 700, and examined the performance of 234 question posers over the 1 year sample period. Not that I believe any more recent data will come up with significantly different results.
I also strongly recommend reading the talk page of that research due to its poignant relevancy nearly 9 years later - it will stimulate further commenting here. Objective comments made there by users such as Mkativerata, Tofutwitch11, Swarm, Worm That Turned, and Scottywong, who are still around, could be the basis of new discussion instead of the perennial repetition here at WT:RfA without action over the ensuing 8 years.
The reforms made nearly 4 years ago touched on the issue of questions, but only their number. Where that reform project failed however, IMO, was in allowing the advertising of current RfAs. This has not improved RfA at all, neither increasing the number of new candidates nor changing the behaviour of the participants. To the contrary, the number of voters has more than doubled, and along with it, the drama which now overspills onto the comments section and the talk page - both places that were never or rarely used previously.
User questions and user votes both contribute equally to maintaining RfA as the one place where users can be as disingenuous and nasty as they like with almost total impunity. This should be a wake-up call either for new changes or for more, systematic, no-nonsense effective clerking but not while the core of regular, reasonable participants remains diluted by hundreds of drive-by, one-time, and often very inexperienced voters, and maybe the occasional troll. Kudpung กุดผึ้ง (talk) 06:08, 25 October 2019 (UTC)
Of some relevance may also be my study on why editors retire (for free mirror use Library Genesis), the key finding is that between a quarter to half of the editors, IIRC, retire because of stress / incivility / fighting. People don't enjoy being put on the grill, so to speak, and the RfA has the reputation of the biggest grill on Wikipedia outside being at ArbCom or such. It is hardly surprising that many people refuse to volunteer for such a stressful procedure. Why add to the stress in life and volunteering just so you can be allowed to help with the mop and bucket? :> --Piotr Konieczny aka Prokonsul Piotrus| reply here 06:42, 25 October 2019 (UTC)
Kudpung. How do you suggest we define and deal with "disingenuous and nasty" behaviour at RFA? Do you want to revive the clerking idea? ϢereSpielChequers 07:34, 25 October 2019 (UTC)
WereSpielChequers, the catalogue of questions here could be a start. Obviously since the community voted for the brilliant idea of publishing RfA so that every wannabe back office worker and troll can participate at RfA, it's unlikely that they will all become potty trained in Wikipedia behaviour, so clerking seems to be the solution. Of course, there is another idea: elect a thousand or so editors with a clean behaviour record, reasonable experience, and long enough tenure to be a RfA pool of regular voters. Sounds nice, but today's kind of community would never wear it. 09:19, 25 October 2019 (UTC)~~
Thanks Kudpung. One area where you and I don't fully agree is over the advertising of RFAs. I welcome the higher turnout, partly because I was hoping that it would lead to more candidates coming forward after a period as voters. But I would be happy to target the ads a bit more tightly, and I accept there is a price to pay in that most RFA voters will take a few RFA votes before they are really effective RFA voters, and even more before they are ready to run at RFA. I'm not convinced that we could find candidates or voters to elect a thousand RFA voters, let alone keep that renewed over time. But I do think it would be an uncontentious improvement to target the watchlist notice to editors who are at least extended confirmed. That would require a software change, and it isn't a rule excluding those who aren't yet extended confirmed from voting at RFA, but it would target the ads at people who are more likely to find them relevant, and be more likely to be ready to be effective as RFA voters. ϢereSpielChequers 11:06, 25 October 2019 (UTC)
I understand where you are coming from WereSpielChequers, but after nearly 4 years of the reform nothing has changed in this respect other than the increase by a factor 2 of the votes and the drama. Restricting the watchlist is one solution, but perhaps simply putting every RfA under ECA would do the same. Kudpung กุดผึ้ง (talk) 03:04, 26 October 2019 (UTC)
@Kudpung:, what fraction of undesirable questions would you guess are from non-EC editors? As a note, while we could both restrict the watchlist and page access, it would be very poor form to protect the pages without limiting who had it showing up on their watchlist. Nosebagbear (talk) 14:55, 26 October 2019 (UTC)
  • I'd say there is a false premise in this thread that "the problem with RfA are the questions". Whilst there were issues with questions back in 2011, were people would each be given lots of questions, or trick questions, or unfair questions, all to get one vote - this has been largely taken care of by limiting the number of questions that can be posed and by allowing the 'crats (and other editors) to effectively clerk the the RfA. Kudpung, of the last 80 or so RfAs (i.e. the past 3 years), how many can you actually point to with problematic questions?
    The issue I see at RfA, is that people just don't want to do it. It's become quite a high standard to pass, people will oppose for all sorts of reasons. I've said before that RfA needs a PR facelift (that's something we could do by removing a lot of the scary bumf), but the community needs confidence that when an admin is no longer suitable then they will no longer be an admin. Far more important than creating admins is improving the system for removing them. but that's another question for another day :) WormTT(talk) 15:13, 26 October 2019 (UTC)
I haven't been suggesting that the questions are the problem at RfA. They are one of several problems.I just chose to start a thread discussing this particular problem because it's visible and hard to deny. Kudpung กุดผึ้ง (talk) 15:35, 26 October 2019 (UTC)
Dear Kudpung, I would not like a situation where we were advertising RFAs to people through watchlists but then protecting the page against them editing it. I see that as very different to only encouraging people to vote once they are extended confirmed. What I haven't done is gone through the voters at a recent RFA to see what proportion were not yet extended confirmed, it may be that people at the start of their wiki careers ignore the RFA ads. BTW re Worm that turned, I'm not convinced that perceived difficulty of desysopping is related to RFA, other of course than the problem that many of the suggested reverse RFA systems would enable spammers and others and deter sensible people from running at RFA. I certainly don't buy the argument that making it easier to desysop Admns who Arbcom won't desysopp would lower the bar at RFA. Though there is a problem that almost any attempt to reform RFA risks being diverted into a discussion about desysopping. ϢereSpielChequers 18:04, 26 October 2019 (UTC)
I've looked through the three most recent (2 successful, 1 unsuccessful) and by my count none of the users asking questions were below extended confirmed. Wug·a·po·des​ 03:47, 27 October 2019 (UTC)
Thanks Wugapodes, and I have just refreshed my memory of the unsuccessful one and you're correct, though in that one the questions were almost all relevant to the RFA. So restricting the watchlist notice to extended confirmed editors will make little or no difference to the question section, though it might reduce the number of voters. Even if it turned out that almost all editors were ignoring the watchlist notice until they were extended confirmed I still think we should target the watchlist notice more narrowly as that would indicate it isn't relevant to new editors. ϢereSpielChequers 07:49, 27 October 2019 (UTC)

This, and similar recurring threads from editors over many years, are RfA Unicorns. There is no valid reason to curtail the overall numbers of questions (absolutely no evidence that questions inhibit candidates coming forward). Further, there is no need to limit the voting franchise and there is especially no justification in curtailing watchlist notices to encourage as wide a community participation as possible. Functionaries cannot be selected on a limited franchise. The handful of pointless / trolling !votes are readily handled by excursion to the Talk Page - an excellent initiative. To me, it is a case of an individual holding up a piece of their work from 9 years ago as somehow totemic and fully relevant to RfA when the rest of the world has moved on. Good candidates will come forward, be scrutinised and be successful. At the other end, Adminship tenure needs to be reviewed (I appreciate not the topic of this thread so I will no elaborate here). There is no "anti admin brigade". Leaky caldron (talk) 10:03, 27 October 2019 (UTC)

Exactly as Leaky cauldron says. I don't remember why we decided to place a limit on the quantity of questions a user may ask, but the measure was a solution seeking problems. While it uses a hybrid format, the core of RfA is debate – between a candidate and the community – and I think an effective RfA process has three features. Candidates should be approaching a broad sample of the community, receiving a healthy degree of scrutiny, and having that process of scrutiny be visible to voters. Measures that frustrate any of those three features must be treated more critically than they are being. AGK ■ 13:47, 27 October 2019 (UTC)
I disagree that the core of requests for administrative privileges is debate. I believe the essential question being asked is if the community trusts the candidate to assume the responsibility of holding administrative privileges. It's not necessary for there to be a sequence of questions discussing opposing viewpoints for this question to be answered. However even within the context of a debate, in the real world, debate questions are limited by available time and so often moderated in some form. English Wikipedia chose to enact a mild form of self-moderation by limiting the number of questions each person could ask. That still leaves the range of topics wide open while encouraging a broader swath of subjects to be broached. isaacl (talk) 19:10, 27 October 2019 (UTC)
Too wide, much too wide, Isaacl, and so broad as to encompass the very irrelevant and oft nonsense questions that are the root of the issues I belive should be addressed. The high number of questions is a fairly recent trend and they are often posed by users: "Ah, anyone can pose a question at RfA - goody, I'll think of something to ask". Kudpung กุดผึ้ง (talk) 01:57, 6 November 2019 (UTC)
My point was that RfA is not a debate. I think considering it as such is more conducive to the type of irrelevant questions you decry than my point of view that discussion (which doesn't have to involve questions to the candidate) should be focused on if the candidate is trusted by the community. isaacl (talk) 02:34, 6 November 2019 (UTC)
Well, Isaacl, instead of talking about it, start an RfC with your idea - I might support it. Kudpung กุดผึ้ง (talk) 16:55, 10 November 2019 (UTC)
Not sure what idea you're referring to—I was responding to AGK's comment that limiting the number of questions that can be asked by a single commenter was a solution seeking a problem, and that the core of RfA is debate. I disagreed that the core is debate, and supported the question limit. Regarding focusing on if candidates are trusted by the community, I don't sense that the community is any more ready to accept ways to manage discussion at RfA than it has been in the past. (At the last big RfC, my proposal to focus on pros and cons of the candidate didn't gain much support.) Taking real world examples, at candidate town halls, questions are selected by a moderator. In one-on-one interviews, an interviewer is chosen to devise appropriate questions. For better or worse, the portion of the English Wikipedia community that likes to discuss these matters is very reluctant to delegate authority to anyone and wants to try to make decisions en masse as much as possible. (Witness its continual questioning of any group that has been empowered with a level of authority.) Until it comes to terms with the fact that consensus, particularly in the form of a large, unmoderated discussion, doesn't scale upwards to a large group making effective decisions, we're stuck with the consequences. isaacl (talk) 20:08, 10 November 2019 (UTC)
We could save the community much time if we just asked Bishonen and Iridescent, instead of RfA. A candidate they both support will not brake the wiki. --Gerda Arendt (talk) 20:14, 10 November 2019 (UTC)
Yes, it's an idea that has been floated before: have an appointment committee, or a nomination committee. Although I have no evidence, my wild-ass guess is if the entire Wikipedia community could somehow be polled, it might agree with this. But amongst the portion of people who like to participate in discussions on English Wikipedia administration, this would mean ceding their personal influence on the matter, and so far they've been reluctant to do so. isaacl (talk) 20:21, 10 November 2019 (UTC)

Oh yes, it's been floated several times before and most recently again by myself. Here's one idea - just an idea:

(arbitrary break)[edit]

Voters
  • A user right to be requested at PERM.
  • Requirement: 1000 edits/12 months. Clean block log.
  • Grandfather all users who have voted on at east 5 RfA in the past 2 years, and not caused any drama. Clean block log for last 12 months
  • 'RfA Voter' bundled with Sysop.

Any user with 90/500 can comment but not vote. Such non-voter comments to go on the talk page

RfA Election Commission/Coordinators
  • Bureaucrats or appointed by voters from list of voters at a yearly election
  • 1 year term. 1 term only
  • Abstain from voting.
  • Duties: clerking, verification of validity of votes, moderating questions

Before this could be proposed, it would be very wise for someone who knows how, to run an up to date voter profile of the kind Scottywong did in 2011. Kudpung กุดผึ้ง (talk) 02:31, 11 November 2019 (UTC)

It's a fine proposal, albeit not an appointment committee as proposed by Gerda, or a nominating committee. isaacl (talk) 05:05, 11 November 2019 (UTC)
As I said, Isaacl, it's one idea - just an idea, but quite serious, not flippant. Kudpung กุดผึ้ง (talk) 04:23, 12 November 2019 (UTC)
Yes, agreed. isaacl (talk) 04:33, 12 November 2019 (UTC)
  • For those not grandfathered in, was it intended to suggest a non-time limited block log? And what is "not caused any drama" supposed to mean - some people state anyone who actively chooses to participate on the AN/ANI boards are needlessly creating drama, but most would say that might be a helpful experience set for !voting on an RfA. If it just means not (recently) blocked, then it seems duplicative. Nosebagbear (talk) 11:29, 12 November 2019 (UTC)
  • I don't have words for how disturbing I find even the thought of this proposal. Some animals are more equal than others.--Cube lurker (talk) 14:27, 12 November 2019 (UTC)
Can you give us an explanation of why you think it is disturbing, a simple proposal that could improve the process, particularly in light of the alternate proposal that is loss of control of this function entirely? scope_creepTalk 16:20, 12 November 2019 (UTC)
One should never have to beg for the right to vote. I disagree with people every time but never will I support stripping peoples right to vote as the solution. "Clean block log" and "no troublemaking" "not causing drama" are clear words of voter suppression. Good candidates get through. Bad ones fail. Just because someones bad candidate failed is not an excuse to try to create an elite voting class. The fact that I have to explain this is double disturbing.--Cube lurker (talk) 16:29, 12 November 2019 (UTC)
Well, Cube lurker, it's better than any of the suggestions you have come up with ;) What I find most disturbing is despite all the grand words, nobody has the guts to make a suggestion. That's what it's all about on Wikipedia, isn't it? No ideas but heckling from the sidelines. If you don't like the ideas, you could at least make an effort to put it more nicely, we're supposed to be on the same side.Kudpung กุดผึ้ง (talk) 20:20, 14 November 2019 (UTC)
I did put my disgust nicely. I rewrote it many times to keep my true feelings to myself and remain unblocked.--Cube lurker (talk) 21:38, 14 November 2019 (UTC)
New User rights, commissioners & coordinators. Seriously? This looks like the perfect recipe of WP:BLOAT. Leaky caldron (talk) 11:48, 16 November 2019 (UTC)

Rising success rate[edit]

Currently there have been 26 RFAs this year so far. Of these, 18 were successful, for a success rate of 69%. If this holds for the rest of the year, it will be the highest annual RFA success rate since at least 2004 (when 79% of RFAs were successful). IntoThinAir (talk) 17:35, 12 November 2019 (UTC)

To what extent is this partially a result of WP:RFA's extended confirmed protection? It was in place in 2018 too, but curious to see how previous year's percentages hold up if you take out sub-ECP user self-nominations. Sam Walton (talk) 17:41, 12 November 2019 (UTC)
  • IMO the requirements have grown more stringent, but there's also better resources for being a good admin available. The voters are also more educated. I think we've had some good candidates this year, and even some of the failed ones I think could have grown into the job and may pass in the future. I believe there's a natural ebb and flow to most things, including RfA. You (IntoThinAir) don't ask a specific question, so I'm not sure what more you're looking for other than discussion. I'm also unsure how extended confirmed protection plays a part. — Ched (talk) 17:54, 12 November 2019 (UTC)
  • Honestly I was just trying to call attention to this fact, since RFA seems like it is in need of good news with all the talk about how it needs to be reformed, etc. It seems like the reasons for the higher success rate are a combination of less experienced/qualified potential candidates getting scared off, and newer users being prevented from nominating themselves because of extended confirmed protection. IntoThinAir (talk) 18:13, 12 November 2019 (UTC)
  • I think many of the people prevented by ECP previously would have had their RFA reverted and deleted, so it would not have appeared in the stats either. But certainly more successes at RFA are good news :) —Kusma (t·c) 21:15, 12 November 2019 (UTC)
  • Definitely - we've not had a clearly unsuccessful (as in, no major support) RfA since June - 5 of the 8 unsuccessfuls/withdrawn had over 50% support, so they clearly weren't unjustified nominations. I think there's been more chasing of good potential candidates this year - and in the last 5 months, ORCP has been much more active. Our success rate last year was lower (though obviously on a depressingly small group size), so there are more-succeed features as well. For example, we (touch wood) seem to have slowed the demand for ever higher editcount etc. Having reviewed the last 3 years RfAs as well, I'd also say that RfAs have become more accepting of errors made so long as they're accompanied by an apology/acceptance. Nosebagbear (talk) 11:56, 13 November 2019 (UTC)
  • Do we know the minimum number we should be aiming for, say on a per montly basis, so we can perhaps have a target. I think 2 is insufficient to stem the losses. scope_creepTalk 14:33, 13 November 2019 (UTC)
  • Basically, you want the mirror of this to break even. That chart isn't quite accurate, as someone who resigns and changes their mind a month later will still show as a desysop for that month, but except for the Framageddon spike in June that's not a statistically significant number. The bottom line (desysops minus restorations) is the significant one; with an average net loss of around 50 per year we'd need 4 or 5 RFAs per month to break even, although bear in mind that most of those being desysopped are paper admins only who haven't edited for a year. ‑ Iridescent 14:42, 13 November 2019 (UTC)
  • (edit conflict; some of what I say is the same is Iridescent) Over the last eight years, we've been averaging a net loss (that includes elections) of about 42 administrators per year (based on WP:DBM). In the same time period, we've been averaging about 21 passing RfAs per year (based on WP:RBM). To reach net zero, we'd need to roughly double our monthly average of about 1.8. That said, this shows the pool of active administrators has remained more or less static for quite some time, having only lost 5% of active administrators in the last 500 days. Going back even further to see from what point to now we've lost 10% of the active admins, that takes us back 4 years to November 2015 when the active admin corps hovered around 560. We're at 502 today. Bottom line; we'd need to have a successful RfA every week to maintain current numbers. That's not going to happen. Eventually, we'll have to address failures in our processes due to a lack of administrators, but it's very hard to forecast when that day will come. But, come it will. --Hammersoft (talk) 14:52, 13 November 2019 (UTC)
  • Hammersoft, DYK is very close to failing due to lack of admin help, IMO. --valereee (talk) 17:32, 13 November 2019 (UTC)
  • Boing! said Zebedee, I know a lot of people don't like it. I like it a lot because I find it helps improve my new creations. But whatever your opinion on DYK, having that process fail isn't good for Wikipedia. If we don't want DYK, the answer is to get rid of it, not to allow it to implode for lack of admin help. --valereee (talk) 19:21, 14 November 2019 (UTC)
  • Valereee, Yes, that's fair enough - and my comment was a not particularly helpful distraction, I admit. Boing! said Zebedee (talk) 20:50, 14 November 2019 (UTC)
  • 60 odd a year, although I think it closer to 70, although it varies quite considerably. Eight years before it goes south and the WMF will intervene. Is it possible to reverse the process, not the RFA per se? Change the ad-hoc process into a planned process. Is possible to plan the number of admins we need every year. Start from the premise, e.g. next year we plan to have 67 administrators elected this year. So we need 67 next year, what do we need to do to ensure 67 are elected? scope_creepTalk 15:39, 13 November 2019 (UTC)
  • The relevant bit is active admins, as noted by @Hammersoft:. We're losing about 1.2 active admins a month, going off the last 4 years. In that timespan we've accepted about 65 admins (1.35/month). So we need 2.55 admins per month to arrest the halt - 31 admins/year. That should be the goal we accept at the beginning of the year, because I think it is actually doable without radical changes to the system. As to admin time allocation - that's a separate matter. Nosebagbear (talk) 22:05, 13 November 2019 (UTC)
  • What do folks think about a project to collaborate on actively nominating editors for adminship? I've long advocated that one of the solutions to our eternal admin problem is that we just need to nominate more people. The problem, of course, is that finding the right editors to nominate takes some effort. I'm interested in the idea of collating some resources and guides on finding editors and going through the nomination process. The project could include milestones to meeting the above figures, which I think are quite achievable with some coordinated nominating effort. Sam Walton (talk) 23:16, 13 November 2019 (UTC)
    • Oh, looks like Wikipedia:WikiProject Admin Nominators was a thing that existed briefly. Worth reviving? Sam Walton (talk) 23:20, 13 November 2019 (UTC)
      • Without the direct potential-candidate discussion, in my opinion. Sam Walton (talk) 23:26, 13 November 2019 (UTC)
        • According to @Hammersoft: and @Nosebagbear: it seems be one RFA every 11 days that is needed. The project WikiProject Admin Nominators seems to be have the correct name but I can't figure out what it did. The remit we need is to keep the pipeline going and present a suitable candidate every 11 days. Work could be done upfront to find the person, evaluate them, and ask if they want to do it, then queue them. scope_creepTalk 01:13, 14 November 2019 (UTC)
          FWIW I have long though a cohort model would provide some safety in numbers and could help convince some people to take the RfA plunge. But man one RfA every 11 days sounds awfully daunting. Best, Barkeep49 (talk) 02:02, 14 November 2019 (UTC)
          Indeed, it is a mountain to climb. It's not to say if there was more folk available but certainly for 2-3 years to get the numbers up, as a sustained effort. You would get very good at finding the good and bad quickly. Work aspect would perhaps streamline the number of voters. The guides to finding editors and how to approach the process must be part of it, to smooth the way so to speak so it as as trouble free and more so to be as stress free for the candidate as possible. scope_creepTalk 03:00, 14 November 2019 (UTC)
          I would suggest starting new requests on the same day of the week, to establish a regular cadence where participants can plan their involvement. Thus either weekly or every other week, depending on how full the queue is. isaacl (talk) 03:12, 14 November 2019 (UTC)
          That wouldn't help a (busy IRL) potential candidate looking in their diary for a 7-day period when they could do an RFA. DexDor (talk) 06:33, 14 November 2019 (UTC)
          Specific-day levels of coordination probably aren't necessary, but I do think that with some guides, milestones, and a sustained effort by multiple editors the numbers above are quite achievable. I'll spend some time poking around that WikiProject and see if it's worth reviving as the place to coordinate this. Sam Walton (talk) 08:51, 14 November 2019 (UTC)
          If there's enough interest, I suggest just going ahead with restarting the project, since the name is suitable and the goals broadly align. (The previous incarnation stalled when it decided it didn't want to discuss specific candidates on-wiki; however it seems you have other tasks in mind anyway.) isaacl (talk) 17:53, 14 November 2019 (UTC)
          True enough, but it can be scheduled whenever the candidate finds an appropriate opening. Candidates can specify on which days they will be responding to inquiries. isaacl (talk) 15:21, 14 November 2019 (UTC)
  • @Barkeep49: - I've generally thought that your cohort idea struggled to be practical, since at the moment it was difficult to find (say) 3 candidates who were free for the same week. Obviously, if we were able to find an increased rate, that might become easier and thus more practical - potentially a virtuous cycle. I'm against fixed days for launching, and to be careful about setting up a circumstance where we "fail" if we don't encourage a new candidate every 11 days. That risks all manner of negatives. If we do decide to go down a wikiproject approach, asking all the current members on the "willing to nominate" list would be definitely worth doing. Nosebagbear (talk) 09:53, 14 November 2019 (UTC)
    If a window is missed, so be it—it's not a big deal. Scheduled dates are motivating, though, and a regular cadence would help make requests for administrative privileges be treated like a ordinary event instead of a special occurrence. isaacl (talk) 15:27, 14 November 2019 (UTC)
    I think isaac's thoughts that rather than a group taking the pressure off a constant stream could do the same. The reason I mildly favor the cohort is that I think it could prevent some of the tempests in a teapots from happening rather than longer discussion about more substantial concerns. Some editor would oppose for some silly reason, the sillyness might or might not be commented on, and everyone would move on. No big deal and something the candidate could brush off. What can happen now is that someone opposes for silly reasons and huge discussion breaks out and this might cause more stress for the candidate. But I don't know. I do think putting together a cohort is easier than putting together a constant stream. Best, Barkeep49 (talk) 15:50, 14 November 2019 (UTC)
    If requests for administrative privileges are a regular occurrence, then I think commenters will be more likely to move on to the next one. You had your say; others will agree with you or not, and whether or not this candidate is approved, another candidate is coming up right around the bend. It should help with the "while a medium deal, not a huge deal" point of view. isaacl (talk) 15:59, 14 November 2019 (UTC)
    I didn't realize I had a set amount of say. I would have explained my thinking at more length originally and been deprived of the chance to say that your method might be more effective :). Best, Barkeep49 (talk) 16:21, 14 November 2019 (UTC)
    I apologize for imprecisely wording my comment; I was not referring to your statement. By "you" I meant the commenters on each RfA. If there were regularly scheduled requests, then each commenter would weigh in and perhaps feel less need to try to argue with others, as there is always another candidate coming right up. With many candidates to comment on, there is an opportunity cost to focusing on one. isaacl (talk) 17:44, 14 November 2019 (UTC)
  • I'm skeptical of the sky is falling concept. But setting that aside I mean this with full sincerity. One thing that will make RFA worse is to try pushing unapprovable candidates into the mix to meet a quota. If every week you put up 1 good candidate and 2 bad, you don't end up with any more admins than if you just put up the 1 good. You just self fulfill the view that RFA is 'mean'.--Cube lurker (talk) 14:46, 14 November 2019 (UTC)
    I think as much as possible, keeping it to one nomination in a window would be preferable. Of course, people can take the plunge whenever they want, but if they want to co-ordinate with others, then having one request active at a time will spread out the work load for commenters, thus keeping a regular rhythm going. isaacl (talk) 15:34, 14 November 2019 (UTC)
  • Heh. I'd rather see half a dozen nominations at a time. Those that are going to fail anyway are going to fail. Those that are going to pass with flying colours are going to pass. It's the ones that are in that "not a 100% shoo-in" range that benefit greatly from having multiple RFAs running at once. Candidates should start their candidacy when they're ready, not in accord with some arbitrary timetable that serves no purpose. (If you want to vote in RFAs, you're going to do it whether there's one candidate a month or one a day.) The reason we're having more success now is that people who have been well-qualified for a long time are finally realizing that the way to get admins is for *them* to run. They're starting to see that there is some fraying at the edges that requires admin tools to fix, and they're getting that they could do those things themselves rather than hope that some admin who isn't too focused on something else will come along and help them. Most of the admins who've come through in the past year have been qualified to run for some time. Now the rest of us need to get into that mindset: we need to get that we need more admins just to spread the work, to pick up slack when someone is away for a while and so on. Valereee gives a really good example above. So keep being respectful and kind to the candidates that will be fine as admins, and we might power our way out of the current leadership void. Risker (talk) 19:47, 14 November 2019 (UTC)
    If we're trying to encourage discussion of candidate characteristics, rather than just voting on gut feel, then I disagree that having one request per day is going to garner the same amount of participation as one per month. I agree with candidates running when they're ready, but the nice thing is that readiness doesn't go stale, assuming the candidate maintains some degree of activity (and they should post-RfA anyway, so presumably they will). So I'd rather see six requests spread out a bit, rather than all six at once and then nothing for weeks and weeks. isaacl (talk) 20:00, 14 November 2019 (UTC)
    I broadly agree with Risker. I believe that anyone who has never run before and who, unprompted, has multiple respected sysops telling them they should run for RfA stands an excellent chance of winning. I think there are some number of people who ran once who if they had never run would be a shoo-in now. The community's tolerance for a repeated RfAs is a bit of a mystery to me. If every candidate ran 6/12/18 months later as some at their failed RfAs were advised we could expand the sysop pool with people who've already been in the waters once. best, Barkeep49 (talk) 23:22, 14 November 2019 (UTC)
I do sometimes go back through failed RfAs and see if there's anyone who fancies a second go. Both Primefac and Enterprisey did well out of this, and GRuban 2 is probably more "when" than "if". Of other candidates who should give it a go, the one I'd really like is CaroleHenson, as I kind of screwed up their RfA and they've never quite been in a position to run again. If I'd left it 6-12 months and run it then, it might have passed and she'd be a good admin. Shoulda woulda coulda.
Part of the problem with second RfAs is that you need to show that the issue that caused the first one to fail is now adequately resolved. If it's too many bad CSD or AfD calls, that's easily demonstrable as resolved. For attitude and civility, it's harder. Ritchie333 (talk) (cont) 23:57, 14 November 2019 (UTC)

Right to Pause?[edit]

The right two paws. --Tryptofish
The paws that refreshes -EEng

Editing Wikipedia is a hobby. RFA is a 7 day process. Some of us, myself included, sometimes or often have a week or more when we can be around quite a bit. But there are plenty of active Wikipedians who I'm sure would make great admins, but who are rarely around for seven consecutive days. Would it help if we enabled candidates to program an RFA around their availability? Take the hypothetical active Wikipedian who has custody of their children three days a week and only ever edits on the other four. We don't need to know why that Wikipedian almost never edits Sunday, Monday and Tuesday, but would we be OK with an RFA that from the start was scheduled to run for 72 hours, have a 72 hour intermission, and then run for another 96 hours? ϢereSpielChequers 22:15, 14 November 2019 (UTC)

I think this problem stems from the assumption that candidates must answer questions or respond to concerns. I'm sure if Floq had been so inclined, they could have simply said "I'm not answering any questions full stop, see my contributions and have at it", and then disappeared for 7 days, coming back to collect the bit or the commiserations. In theory, if everyone voting needed to search contributions to make a decision, and support or opposition was rebuked by uninvolved third parties, there would be no need to worry about a candidate's time as they wouldn't need to do anything. Ritchie333 (talk) (cont) 22:33, 14 November 2019 (UTC)
Seems unnecessary, and would add undue complexity to the process. A candidate needs to be available mostly at the beginning of the process, to fill out the required info and answer the initial questions. After that, it's not essential for them to be constantly available. ‑Scottywong| [soliloquize] || 22:35, 14 November 2019 (UTC)
When you start your RfA is one of the few things you can control about others around here! So start it when you have at least 2 days to reply - feel free to note that you will be off for a days 3 and 4 for example, then come back and catch up. — xaosflux Talk 22:46, 14 November 2019 (UTC)
The problem is that the philosophy "see my co contributions and decide for yourself, full stop" just does not work. Most people are too lazy to go and check the contributions; the project is still big enough that most voters do not know the candidate or work in different areas of Wikipedia, and it is enough to have a couple of enemies ready to cherry-pick a couple of episodes and highlight them, and the nomination is dead. Even if these episodes are old, and there are plenty of good more recent contribution. I had this illusion in the past, and I run into a troubles trying to follow it. If somebody shows up with any completely absurd accusation based on cherry-picking something I said and putting it out of context - I thought users would go look up diffs and analyze the whole episode, and then it becomes obvious that the accusation is bullshit - nope, nobody would do it, not even the arbs. They would just stick to the accusation and would decide whether I should be blocked now or given a warning and blocked later. That's life.--Ymblanter (talk) 08:54, 15 November 2019 (UTC)
I don't believe it is necessary for candidates to be continually available for 7 days. If they state up front what days they will be active that will forestall complaints that questions aren't being answered (though I think we ought to tamp down expectations on that front, too). If they really wanted to, I imagine it should be feasible for candidates to state at the start that they want the request to stay open for an extra length of time until date X, to accommodate their schedule. isaacl (talk) 22:48, 14 November 2019 (UTC)
If an RFA goes really well, a candidate can pretty much disappear for the final day, maybe even the one before. I have known a candidate put an apology in that something had come up and they wouldn't be around for 24 hours in the middle. However that's about as much latitude as the community currently gives admin candidates. My suggestion is much broader, I'd like us to open up to the sort of person who might only be active on wet weekends. The only way such a candidate could run an RFA would be to have it in a series of chunks, or to rearrange their schedule to be around for 7 days continuously. I don't see any downside to such an RFA, other of course than that an editor who only edits on Thursdays and Fridays would not be able to comment on an RFA that ran in two Saturday to Tuesday chunks. ϢereSpielChequers 00:24, 15 November 2019 (UTC)
I don't see a need to halt the request process in the middle, though. The candidate can ask for it to run for nine days from Saturday to the second following Sunday to cover two weekends. isaacl (talk) 00:47, 15 November 2019 (UTC)
As an admin, the sort of person who might only be active on wet weekends is likely to be a damp squib. Unless they live in England. Kudpung กุดผึ้ง (talk) 05:25, 15 November 2019 (UTC)
Quite a few of us do live in England.... and I'm sure I'm not the only one who is more likely to edit when it is dark or wet in the garden. But more seriously, do we want to restrict adminship to those who are here at least as many hours as a part time job, or are we open to someone who puts in the equivalent of an evening every week or so? Given that the RFA expectation of recent activity is 100 plus edits a month, I'm assuming the latter. If so I think it would be helpful if candidates could pause their RFA and not hold it in one continuous 7 days. ϢereSpielChequers 16:56, 15 November 2019 (UTC)
But then you're asking editors who want to vote on an RfA to understand when they can or can't vote, or ask a question, or add to the discussion. Imagine you're an editor who notices an RfA and wants to vote on it, and you try to do so, but ohhh, it's paused now, I guess I'll need to come back in a few days and try again? That would be a very strange experience. How about we just build a culture where it's ok if it takes an RfA candidate 24 hours to respond to a question? Does that culture not already exist? I don't think anyone has a stopwatch on the candidate, and if they don't respond to a question within an hour, they get an automatic oppose for being too slow. Additionally, while no editors are on Wikipedia 24/7, it's probably not too much to ask of an admin candidate to be available a bit more than normal during the week of their RfA (and I think most of them naturally are, because it's an exciting time). ‑Scottywong| [talk] || 17:28, 15 November 2019 (UTC)
Yes, doing this with a series of pauses would be different and you'd need a clear way of doing the pausing. Currently the expectation is within 24 hours, after the first few hours there is no problem in the candidate only being around a few hours a day. For our current candidates that is rarely a problem, but I'd like to open adminship up to people who aren't able to be here every day. ϢereSpielChequers 18:08, 15 November 2019 (UTC)
We should not do anything that could prevent !voters from participating, one reason it is a week instead of say 3 days is precisely because some people can only access the site on some days. — xaosflux Talk 20:19, 15 November 2019 (UTC)
  • Is there any reason, anecdotally or otherwise, to believe that the need to answer questions over a seven day period is actually holding back any candidates from running? Boing! said Zebedee (talk) 19:53, 15 November 2019 (UTC)
    That would be difficult to figure out, but I am sure that multiple candidates, I believe including myself, posted notices during their RFA that they would be unavailable during a day or two. This means that they took the possibility of opposition due to lack of answers in a day, real or imaginary, seriously.--Ymblanter (talk) 20:07, 15 November 2019 (UTC)
    Interesting, thanks. If a notice like that worked for you (and for others), I don't see why it shouldn't work for anyone else. Boing! said Zebedee (talk) 20:22, 15 November 2019 (UTC) I'll add that the reason I'm asking this is that I've seen countless attempts over the years here to solve what isn't the problem. Boing! said Zebedee (talk) 20:24, 15 November 2019 (UTC)
    It certainly should work in a normal situation, I guess the original question of WSC was how we facilitate RFA for people who can only edit say two times per week.--Ymblanter (talk) 20:42, 15 November 2019 (UTC)
    Sure, I understand. I'm just wondering whether there really are people who would be that restricted even during an RfA. Boing! said Zebedee (talk) 21:04, 15 November 2019 (UTC)
    - As a matter of fact, not being available continuously for a whole seven days has certainly put me off standing at an RFA over the last 12 months (as Ritchie333 and Amorymeltzer could probably confirm). But thereagain, I welcome any excuse to procrastinate(!), so I'm not too bothered about waiting until I'm free enough to do it justice. After all, there is no rush, is there? That said, why do we not ask candidates to indicate in their application the approximate UTC times they envisage being online/available in the RFA period ahead? One person's daytime is another person's beauty sleep. It could also be argued that, if an RFA were to run over 9 days, not 7, it would indeed allow its timing to straddle two weekends. But inserting a pause in proceedings, except in exceptional circumstances, seems rather unworkable to me. Nick Moyes (talk) 22:45, 15 November 2019 (UTC)
    Yes, that's what I said: candidates can indicate when they are active, and ask for the RfA to run for nine days covering two weekends. isaacl (talk) 23:20, 15 November 2019 (UTC)
    I like the idea of extending RfA to 9 days more than I like the idea of an intermediate pause, and then asking the candidate to say when they'll be available for questions - it should be easy enough to make an availability template to display in the RfA. It would be simpler than tailoring intermissions to suit individuals, and I'd hope it would fix the availability problem for most people. Boing! said Zebedee (talk) 12:52, 16 November 2019 (UTC) (Updated, Boing! said Zebedee (talk) 10:45, 17 November 2019 (UTC))
    FWIW, I don't agree with the premise of this thread that the requestor needs be available continuously for a whole week. A few days up front to answer the slew of questions, but if nothing significant arises, these days it's a good five days of a slow trickle of supports and some opposes. There's little a candidate needs to do except check in an answer a question here or there; popping on for a quick bit before going to bed or after waking up should be be sufficient. More controversial candidates should likely be more available, but even then, unless they do something new, there's not much to change.. ~ Amory (utc) 15:31, 17 November 2019 (UTC)
  • Do we know that candidates would want a 9 day RfA? I would have preferred to go 5 days rather than 7 let alone 9 - less time for the slow drip anxiety that RfA can engender. It seems at least one potential candidate for RfA would also prefer shorter to longer (Nick Moyes). If going to 9 days would help real candidates than by all means let's do it. But I just have to wonder if we wouldn't be better off doing some sort of RfA survey where we would ask "Would this make you more likely, less likely, or not make a difference for you to RfA?" of potential reforms. In this idea I would see us polling X number of the next RfAs and an equivalent number of people who we identify (somehow) as potential candidates who didn't run and see what we come up with. Best, Barkeep49 (talk) 18:17, 16 November 2019 (UTC)
    The idea is to allow those who participate once a week to have a chance to comment, so the request should remain open at least seven days. Most requests are pretty cut-and-dried and don't have a lot of new comments by the end of the period, so I don't think extending the request a couple of days will make much difference. (And for a weekend-only editor, the current alternatives are for the editor to either schedule the request so it ends without making any last replies, or to schedule it so the editor isn't available at some other time.) I don't think we need a poll on this specific issue: it's just an option that they can choose if it helps. If not, they can choose any other available options that they prefer. isaacl (talk) 21:56, 16 November 2019 (UTC)
  • Mine went easily enough -- the majority of work on my part was in the first two days -- so I would still prefer 7 days minimum. If a user feels he cannot attend to an RFA except on the weekends, I don't see an issue with a request to the crats to hold on closing, since it's always been 7 minimum, not 7 absolute. --Izno (talk) 20:57, 16 November 2019 (UTC)
  • The greater issue is the community (in its form of a bunch of individuals, not a hive mind) being fine with it. I think if the candidate posted at the start that they'd be non-answering for days x, y & z but it would be open another 2 days, most would be fine with it. There is the potential for some incidental difficulty (most RfAs that ultimately end in the 55-85% range go "big pile of supports, a few opposes, mixed pile of opposes, relatively equal trickles of supports and opposes), though the most recent was a distinct anomaly on that front. As such, a longer RfA might get a couple of % lower than an otherwise equivalent RfA, but that would be relatively minimal (few people !vote on day 6) Nosebagbear (talk) 13:55, 21 November 2019 (UTC)
  • I don't think there should be a pause. Voters also have the right to only edit on Monday mornings and Wednesday afternoons, and still have their voice heard in an RfA. Just make it clear to anyone who expects candidates to be always available within 48 hours that this is not a reasonable thing to expect from other volunteers. —Kusma (t·c) 14:27, 21 November 2019 (UTC)
  • I think the best timing for an RfA would be to set aside at least 2 or 3 days where you can look at an RfA attentively, perhaps a weekend on which you are free, and schedule those three days as the first three days of your RfA. After three days, if you've done well, then there will generally be enough support votes that you can delay answering questions until you get the time. Mz7 (talk) 18:47, 25 November 2019 (UTC)

Separate discussion from polling?[edit]

Copied from Special:PermanentLink/928097395#General comments to continue in an even more general manner
  • Do we have to have people arguing with every oppose? It creates such a poor atmosphere that Oppose !voters gets attacked in every RfC and Support !voters who post trite supports ("No big deal", "Why not?", etc) never get questioned over it (and I've seen people who do question those !votes get attacked for doing so). It's not about "having a discussion" in the RfC - these are people who just get attacked for having a different opinion. - SchroCat (talk) 10:58, 26 November 2019 (UTC)
    Hi SchroCat: in a clerking capacity, I reviewed the four responses to the four oppose comments currently placed: I’m having trouble viewing them as “attacking” the participants - RfAs are still a discussion at the basic level and these counter-arguments are delivered in a calm manner (save for the word “ridiculous”, which does add a bit of heat) and mostly speak to the arguments being made, not the participants. I understand that it can be somewhat uncomfortable (and effect a chill) to have to defend one’s position (or for one’s position to be questioned) in this context, but this is the consensus process as it exists at RfA today. Speaking generally (and this should maybe move to WT:RFA), but I wonder if this “uncomfortableness” would be lessened if we attempted to separate the polling from the discussion, so specific arguments or reasons in support or opposition to the candidate would be cataloged and discussed individually in sections, and then the polling could be done separately with only brief remarks referring the “Oppose due to oppose reasons 2,4,5; reasons 3,6 are irrelevant to me; moral support due to support reasons 1,7” and no threaded discussion would occur in that section. What do you think? –xenotalk 11:17, 26 November 2019 (UTC)
    (edit conflict) I certainly think that comments and !votes should be separated. We have people putting in a support vote, then attacking the rationale of the opposers - it's a bad look and will keep people away from contributing. I've seen people being berated for questioning the more trite support !votes, and yet a thought-out oppose (whether others agree with it or not) will automatically come under fire. Look above, a couple of people think the edit levels are not good enough to support the candidate: we don't need three people to jump on them and tell them they are wrong - they are no more right or wrong than the supporters who do not think it a problem. - SchroCat (talk) 11:29, 26 November 2019 (UTC)
    I have previously suggested separating discussion for counter-arguments, to avoid repetitive conversation threads, and so certainly support this approach. This will make the conversation more effective and depersonalize discussion, thereby making it less confrontational. It would also be more conducive to building up a consolidated list of pros and cons, which can provide a concise recap for anyone looking to join the conversation or review what has gone on since they last participated. isaacl (talk) 17:45, 26 November 2019 (UTC)
    As one of the four opposers, I definitely don't feel attacked by having a point challenged. It's a reasonable difference of opinion. – Teratix 11:21, 26 November 2019 (UTC)
    (edit conflict) While you yourself may not feel it, there is already an attacking air for anyone who feels they do not want to support this candidate, but does not want ot be attacked for doing so. - SchroCat (talk) 11:29, 26 November 2019 (UTC)
  • Support !votes are (and should be) challenged if they are felt to be wrong. Here's one example and here's a challenge I made. Ritchie333 (talk) (cont) 12:30, 26 November 2019 (UTC)
    Initial opposes at an RfA regularly recieve pushback that early supports, some of which could be challenged here, don't. It's one of the ways I've identified that help make RfA right now favorable to the candidate. Best, Barkeep49 (talk) 14:00, 26 November 2019 (UTC)
    There are currently six supports that show insufficient or trite rationales. I know that if I press all six of those to given some decent rationale, I'll be pulled up for harassment or badgering (I've seen it happen to others when they have tried to get people to explain). One of the reasons RfA can turn into a bear pit from time to time isn't people attacking the rationales of the support votes, but those attacking the rationales of the opposes. No-one wants to see time wasting opposes or abuse of candidates, but if any opposers are going to have their opinions trashed at each and every RfC, then it will inevitably turn sour from time to time. -SchroCat (talk) 14:09, 26 November 2019 (UTC)
    While the default position is 'promote', I've previously commented in bureaucrat discussions that bare supports (and the like) will typically carry less weight in close call situations. –xenotalk 14:43, 26 November 2019 (UTC)
    Why count them at all? If people are unable to leave even a basic rationale, then it's not a discussion or a consensus-weighing process, but a straight vote, which is not what this is supposed to be about. - SchroCat (talk) 15:31, 26 November 2019 (UTC)
  • I typically don’t heckle opposes, but I think when the initial criticism is something as ridiculous as saying someone who has over two years of experience and has never dipped below active editor status in that time does not have enough experience or that their activity is inconsistent, that does need to be pointed out lest we have a meme effect. To xeno’s point on the word ridiculous: it’s a fair description of the view that over two years is not enough time. TonyBallioni (talk) 15:22, 26 November 2019 (UTC)
    Tony, that's your opinion (and I think I agree that two years is sufficient), but just because someone else has a different opinion, it doesn't need people to pile in and argue the opposite. Telling them they have the "wrong" opinion seems wong. Surely they are allowed to have their opinion without people sniping at them? Or should we allo mass sniping and criticism of all votes based on all opinions? The 'crats are sufficiently proficient in weighing up whether it is a viable Oppose or not - it doesn't need everyone else souring the atmosphere. - SchroCat (talk) 15:31, 26 November 2019 (UTC)
    Personally, I’d make RfA a straight vote so we didn’t have to worry about any of this. I think that would be substantially more fair to everyone and would lower the temperature in the room significantly. The community has for reasons beyond me decided that it thinks RfA should be a discussion, which means that opposes tend to draw discussion if people disagree with them. I think people should be prepared to defend their view if challenged, but those challenging them also should know when to stop.
    Since my standard support is probably one of the 6 you mention above, in fairness I’ll expand: I don’t have much experience with Dreamy Jazz, but I’ve seen them around enough to realize they have a basic understanding of how the project works, know they’re typically pretty competent, and have never seen them behaving rudely to others or in a way that I think they will abuse the tools. If someone meets those standards, I’m fine supporting (though I typically do it in a much shorter fashion.) TonyBallioni (talk) 16:10, 26 November 2019 (UTC)
    I wasn't necessarily thinking of yours (although it is more borderline than some others), but numbers 5, 7 and 49 are meaningless in the "discussion" sense and are just straight votes (rather than !votes). Maybe going for a straight vote is the way forward. Maybe 3 days of discussion followed by three days of straight votes would be an idea, or possibly separating the !voes from a discussion on the same page. Either way, the badgering of each opposer while the supporters leave no rationale is not conducive to a free and open discussion. - SchroCat (talk) 19:31, 26 November 2019 (UTC)
  • I disagree with separating the !votes and discussion on them, as they get significantly more eyesight and means that opposition to the oppose reasoning is actually seen by anyone (not just those careful at reading everything before !voting) see it. If needs be an in-page link (similar to the ones to the talk page we do for big threaded discussions) could be put in, but that's a 2nd preference. I am happy to concede that the lack of challenging of unsupported support votes can be viewed as unfair, and I'd be perfectly happy if they were also viewed as open to discussion. Nosebagbear (talk) 20:03, 26 November 2019 (UTC)
    • Perhaps not allowing any "discussion"(/badgering/comments/harrassing/whatever) at all after the initial comment? Any ridiculous statements, insults or trolling can be moderated, so why have the interaction at all? - SchroCat (talk) 20:38, 26 November 2019 (UTC)
    • I think at present the discussion in the oppose section gets more attention anyway. Consolidating discussion would avoid covering the same arguments in multiple threads. The way these discussions are repeated across many people's stated opinions magnifies them in the readers' minds (including the candidate's). isaacl (talk) 23:22, 26 November 2019 (UTC)
  • Separating the !votes and the discussion would lead to even more drive-by participation as !voters would feel even less need to explain their position. After all, nobody is going to double-check that each !voter also posted in a separate discussion. Lepricavark (talk) 20:30, 26 November 2019 (UTC)
    Given there are some supports that give no rationale at present and a heap of others that barely register as a fill sentence, it looks like we're already in that position. - SchroCat (talk) 20:38, 26 November 2019 (UTC)
    And I've said I'd be happy to have support voters need to justify themselves, if that was the "cost" to allow threaded consideration of oppose !votes. More active clerking would be fine. Nosebagbear (talk) 22:14, 26 November 2019 (UTC)
    Commenters would still enter their opinion with a concise summary, just as is typically done now, which is easily distinguishable from a bare support or oppose. Further discussion about the points made would be separated. This way the conversation about a given argument can be held once, in one location, instead of being spread out in many people's comments. isaacl (talk) 23:15, 26 November 2019 (UTC)
    There are some mockups RfAs at Special:PrefixIndex/User:MBisanz/Qs/RfACandidate. Maybe someone could reformat one into the format being discussed. –xenotalk 23:32, 26 November 2019 (UTC)
  • Separating discussion and voting is fine only if the votes themselves do not contain a rationale. Very often the comment you make while supporting/opposing/neutraling has the effect of campaigning for others to join you in your opinion (even if that is not the intention). Nonsense rationales ("too many deleted edits, user clearly does not understand notability") should not be allowed to stand. Either disallow rationales or allow them to be challenged. (I deliberately use straight votes when I don't want to influence others, and so I am one of the main offenders in the "voting without a rationale" category). —Kusma (t·c) 20:44, 26 November 2019 (UTC)
  • Well, I oppose all of the possible changes proposed here (and feel free to reply to me!). As I see it, over time, the most frequent concern expressed by editors about RfA is that potential good candidates are reluctant to run because of the harshness of the process. The last thing we should be doing is to make it harder to support and easier to oppose. (And the second-last is to make it less of a discussion and more of a vote.) When I support, I almost always make an effort to provide a rationale for supporting, and I do wish more supporting editors would do so. I do find a lot of the "canned" support rationales lazy and annoying. But I don't think it's something we should legislate. I support the current practice of the 'crats to give less weight to comments without rationales in close cases, and that really is the bottom line. When I'm supporting and I think there are serious flaws with some of the oppose comments, I find that a lower-drama way to respond is to refrain from commenting in the oppose section, and instead explain why those opposes are flawed in my own support comment. (And I trust the 'crats to see it there.) One thing I could support would be to have some kind of notice added to the top of RfA pages, that points out explicitly that !votes without clear rationales are likely to be given less weight. --Tryptofish (talk) 23:52, 26 November 2019 (UTC)
    I don't see having a separate section for extended discussion doing either of those things: writing an initial support opinion would be the same as now, and writing an oppose opinion won't get any easier. It would make RfA more of a discussion by actually holding a group conversation on specific topics, rather than individual threads with individual commenters. isaacl (talk) 00:10, 27 November 2019 (UTC)
  • I am opposed for two reasons. To begin with, I agree with Trypto that making opposes easier doesn't make RfA less toxic for candidates and that should be our focus. I'm not seeing any argument from people that consensus has changed in the last six months when we had an RfC close with a consensus of consenus processes rather than more vote like. And to me this idea makes it much more vote like. Best, Barkeep49 (talk) 04:24, 27 November 2019 (UTC)
    For example, Wikipedia:Pending changes/Request for Comment 2012 separated initial statements by each commenter from subsequent discussion. I think it worked quite well in letting everyone state their positions upfront, without having to wade through interleaved discussion threads, while still fostering lots of discussion. isaacl (talk) 04:51, 27 November 2019 (UTC)
  • A lot of the time, the opposition comes down to a single isolated issue. For WP:Requests for adminship/HickoryOughtShirt?4, it was a couple of ECP requests. For WP:Requests for adminship/Valereee it was running the same as WP:FRAMGATE. For WP:Requests for adminship/RexxS, it was saying "don't fuck about" once. For WP:Requests for adminship/EvergreenFir it was reporting Eric five years ago for something. These attract pile-on votes that make it sound like an ongoing problem. It's therefore no real surprise, particularly when an RfA is running at 85%+ support, that people will be naturally attracted towards weak counterarguments and refute them. The reason you don't see the reverse - ie: initial opposes and little support, is because it's now technically much harder to get into that situation in the first place. But we used to have WP:NOTNOW RfAs that closed on (0/10/2) where nobody cared about the opposition. Ritchie333 (talk) (cont) 16:34, 27 November 2019 (UTC)
  • Thanks for the suggestion, SchroCat, and for all the thought here. But I don’t agree with the proposed changes to our current system. First, let’s make it clear what kind of RfA’s we are talking about. It’s usually the cases where there are many support !votes and only a few oppose ones. In those cases, the oppose !voters, especially the first few, tend to get argued with. Sometimes it can come across as pile-on or badgering by the candidate’s supporters, and that’s unfortunate. Sometimes it turns into an extended argument, which is then properly clerked to the talk page. But I don’t agree with any of the proposed solutions, especially separating the responses from the !vote. Many opposes NEED to be responded to; they may contain a misstatement of fact, or the !voter’s unique interpretation of an action that most people do not find to be problematic. If opposes are left in place unchallenged, something like that could be taken at face value and turned into a quite unjustified meme. I think we should keep our current system, especially now that some bureaucrats (thank you, xeno!) are monitoring the discussions and clerking when necessary. I oppose a straight !vote without comment; many people put research and thought into their position and argument. And I disagree with the suggestion that support and oppose reasons could be somehow catalogued, numbered, and referred to by number. The reasons given are way too varied and complex to be catalogued. If someone wants to reference a particular argument, they simply say “support/oppose per so-and-so”. -- MelanieN (talk) 17:22, 27 November 2019 (UTC)
    • What we have seen here (and it's not uncommon in almost any RfA you look at) is that someone has an opinion on the number of edits they think an editor should have, or that they're asked for a block to help take a break. It's an opinion, a judgement: there is nothing wrong or right in what they have said, because it's their opinion only. Why, in that case, have people been lining up to argue with them, telling they are wrong to hold such opinions. That's wrong as it creates a rather unpleasant atmosphere to others who may want to oppose but don't want to get dragged into a long argument. It's also wrong when support voters who give either absolutely zero rationale or rather trite reasons ("why not" or "NOBIGDEAL") do not face any questioning at all - or worse, when someone does ask them for a rationale, they are told to stop being disruptive or harassing supporters. It's a skewed and unlevel playing field that allows more passive-aggressive behaviour than anywhere outside ANI. - SchroCat (talk) 17:30, 27 November 2019 (UTC)
      SchroCat, I don't agree with your analysis of the situation. But even if I did, I'm not sure that would merit changing our system. You seem to be arguing that we need an environment where opposers feel more comfortable. I think it's to our benefit as a project that people don't feel that they can just toss off thoughtless opposes - we should all be forced to be careful before tossing off a criticism of another editor. What you are suggesting doesn't seem to me like it will change what you see as passive-agressive behavior (and again I don't agree with this characterization) it's just that rather than being aimed at people who want to oppose it'll be aimed at candidates. That doesn't strike me, on balance, as a positive change. Best, Barkeep49 (talk) 18:20, 27 November 2019 (UTC)
      I certainly don't want people to think I'm tossing anything off anything! ([FBDB] I think the term may mean something different where you are to me!)
      Yes, I certainly want an environment where opposers feel more comfortable, just as they should be, as long as they are making a good faith oppose based on a thought out rationale. Opposers should feel just as comfortable in posting their vote as supporters do. We already have a different threshold to how support and oppose votes are treated (an oppose without a rationale is ignored; a support without a rationale is only ignored if it is a close call) and we have a situation where supporters are given a free ride at little or no rationale where they are not questioned, yet most people opposing will be badgered despite having already given a rationale. Why would you want anyone to be uncomfortable voting at RfA? What we have - and what you are defending - is treating people differently just because they may see the threshold of adminship differently to you. It's a shoddy system we have at the moment, but I guess the usual inertia to change will mean we will retain a second rate process. - SchroCat (talk) 18:32, 27 November 2019 (UTC)
      Yes it does mean something different. Sorry for the colloquial expression there. "Why would you want anyone to be uncomfortable voting at RfA?" Because RfA isn't a vote and so if someone thinks it is they should be uncomfortable. It is, according to our most recent discussion, a consensus building process. Unlike most forms of consensus, RfA requires a substantial consensus so there is already an extra importance attached to opposers. A minority can prevent someone from becoming a sysop. I think this a good thing - protections of minority viewpoints from tyranny of the majority is important. However, it does, for me, mean (because I also operate from the premise that we could use more sysops, and editors in general) that leveling this structural imbalance slightly by placing a higher "cost" to opposing than supporting is alright. Especially because in this instance the issue under consideration is a person at the other end of the keyboard who deserves to be treated with respect. A system that makes it easier to say nice things rather than potentially hurtful things (even if done respectfully and with evidence) again is a system whose bias I am ok with. I understand that you feel differently and hopefully have been respectful myself in this discussion. I would suggest, however, that if you're interested in improving RfA that changing your mindset of "it's a vote" to "it's a consensus building exercise" might help you become more likely to overcome the status quo bias that permeates any efforts at change on Wikipedia. Best, Barkeep49 (talk) 18:53, 27 November 2019 (UTC)
      (edit conflict) I know people keep claiming it's a discussion, but that's obviously farcial, given the number of support votes (not !votes) at each RfA. (And it's not my mindset that I think of it as a "vote": I have just got bored with typing "!vote" so many times in this thread; RfA has been a clusterfuck for so long, and while there have been a couple of improvements, it still a second rate process with which there is too much inertia to change). I disagree with much of what you have written about desiring a structural inbalance against opposers: !voters of all types should be treated the same, and the supporters should not attack the opinions of others. People's opinions differ, but no-one berates the supporters for some rationales that can be seen as "ridiculous" or "insane" (as some of the opposers have been characterised in the current RfA). - SchroCat (talk) 19:07, 27 November 2019 (UTC)
      I agree people should be treated with respect. This includes people voting oppose. I agree it's not great for their reasoning to be called ridiculous or insane. And the reason I keep saying it's a discussion is because we had an RfC six months ago that closed with that consensus. Disagreeing with that consensus is fine of course and I respect your efforts to change that consensus. Best, Barkeep49 (talk) 23:27, 27 November 2019 (UTC)
      I think you need to read the full thread: I am not necessarily saying that we should stop this being a discussion. That is one suggestion that someone else has made that deserves wider examination. Other possibilities also include splitting the discussion from the !voting, or any other method that stops opposers being harassed as a matter of course, while the supporters can leave no rationale and still be counted as a full vote. - SchroCat (talk) 23:43, 27 November 2019 (UTC)
      I think we know where each other stand on the bigger topic and we've had a spirited hopefully productive discussion. I am curious about one thing though. If someone left in response to an oppose I think you need to read the full page would that be harassing in your mind? Despite my having read this full thread I don't think of it as such but am curious your thinking. Best, Barkeep49 (talk) 01:22, 28 November 2019 (UTC)
      We already have a different threshold to how support and oppose votes are treated Yes, we do: Oppose !votes carry three times as much weight as Support votes, thanks to our 75% threshold. Is it any wonder that Oppose !votes are taken seriously, and sometimes argued with? One Oppose !vote cancels out three !Supports. -- MelanieN (talk) 18:58, 27 November 2019 (UTC)
      It's 65% and one this year was lower than that. - SchroCat (talk) 19:07, 27 November 2019 (UTC)
      Fine. One oppose cancels out TWO support !votes in terms of leaving an avenue open via the bureaucrat discretion zone. One oppose cancels out three supports in terms of passing directly. My point remains: Oppose !votes carry more weight than supports do. You said that opposes have a different threshold; I absolutely agree, but for a different reason. -- MelanieN (talk) 19:36, 27 November 2019 (UTC)
      I've been stalked, harrassed and trolled by more than one admin. I've been badly blocked more than once. The thought of a higher threshold for tyro admins comes as something of a bonus to ensure that we do not hand out powerful tools to those who are not fit to use them. I am not happy with an unequal system that gives licence for passive-aggressive bullying of people for well thought out opinions. It also means we continue to retain a second rate infrastructure for running the place. - SchroCat (talk) 20:01, 27 November 2019 (UTC)
      Letting commenters make an initial, concise statement laying out their reasoning, with further discussion occurring in a separate section, doesn't change the discussion into a vote. It just allows discussion to be consolidated. I'm confident that the bureaucrats will read the entire discussion and still be able to suitably evaluate each person's opinion. isaacl (talk) 01:42, 28 November 2019 (UTC)
"Numbers of roughs ..."
  • Secret ballot There's currently a general election in the UK in which the issues are being discussed and then the voters will get their say. This is done by secret ballot and there are rules to prevent partisans from interfering with this (see Maintaining order for some guidelines). When this was not done, bribery and intimidation of voters was quite common. RfA currently has this character whereas the voting procedure for arbcom elections seems much more respectable. I cast my arbcom votes recently, going through the ballot to tick the support and oppose boxes and this was quite fuss-free. I did a bit of browsing of candidate statements and other background, where I wasn't sure, and so was reasonably well-informed. Perhaps there was some noisy discussion somewhere but I'm not interested in having people tell me how to vote as I have a mind of my own and that's the point of the process – to establish a broad consensus and representation. This is best done with secret ballot as all civilised places do. Andrew D. (talk) 13:08, 28 November 2019 (UTC)
Would it be practical to run SecurePoll on RfAs, given the relative lack of them compared to the past? Ritchie333 (talk) (cont) 13:26, 28 November 2019 (UTC)
xaosflux would probably be able to answer that. –xenotalk 13:53, 28 November 2019 (UTC)
A practical process would be to do them in batches so that the voting form and mechanisms would resemble the arbcom one, with a selection of candidates. If this were done monthly, or quarterly, then there would be a regular cycle and process which might encourage people to come forward. My impression of the current process is that the business of preparing nomination statements is quite intimidating because each one is isolated. If candidates were grouped in bundles then you'd get economies of scale and safety in numbers. The assessment process might also be fairer as candidates would be compared with each other rather than being judged against an idealised, abstract standard. Andrew D. (talk) 14:18, 28 November 2019 (UTC)
@Ritchie333, Xeno, and Andrew Davidson: - doing it on votewiki (like the way we do arbcom) would not be practical unless we changed RfA's to be like ArbCom elections, mainly occurring very rarely (maybe twice a year?) The votewiki server currently requires WMF staff to deal with, and has issues running more than one election (globally) simultaneously (I think this is mostly due to information disclosure and encryption systems). I suppose we could run 'local' securepolls but there would be a few limitations issues:
(a) encryption issues may mean that the "secret" votes are only "private" (i.e. sysadmins may be able to access the data),
(b) dealing with the "private information" disclosures: election administrators (which could be limited) can view USER::IP ADDRESS::USER AGENT correlations in secrepoll for all voters, unlike with the checkuser tool - viewing this information is not logged - so you would have to assume that election administrators would certainly have this disclosed to them, such disclosure may alter the people likely to vote (I suspect sockpuppetts would be less likely to vote, but privacy minded editors may also be disuaded).
(c) We would need a system for dealing with election administation, a group of users would need to be the "election administrators" and would need to be trained and available to run the elections.
(d) extension spport for SecurePoll would need to be looked at - if something went wrong in the middle of an election with the software, there may be noone that could assist with repairs.
Of these I would see "C" as the biggest community issue to deal with, follwed by D, B, then A. — xaosflux Talk 14:43, 28 November 2019 (UTC)
"Economies of scale and safety in numbers"
"If candidates were grouped in bundles then you'd get economies of scale and safety in numbers." That's exactly why I ran Kosack and Valereee's RfAs together, and why I snuck EvergreenFir's in while two others were going. Ritchie333 (talk) (cont) 16:07, 28 November 2019 (UTC)
And in fact it did feel less exposed to be running an RfA at the same time as someone else. I think three or more at a time would be ideal from the candidates' standpoint. --valereee (talk) 16:46, 28 November 2019 (UTC)
The idea of running a bunch of RfAs together at specific points in the year rather than "on demand" is an interesting one that we should spend more time discussing, perhaps in a different discussion. Even if we keep the current format and don't use SecurePoll, I think it can certainly make some candidates more willing to run if they know a bunch of other people will run at the same time. Mz7 (talk) 23:54, 28 November 2019 (UTC)
Running editors in flights is already being experimented with (though you might not have noticed); AA88, Bradv, and Chetsford earlier this year were the flight in question. (I just missed it though I was offered; busy in real life that week.) --Izno (talk) 17:07, 29 November 2019 (UTC)
Hrm. While I see the problem with people stating an opinion and suddenly a ton of verbiage falls on their heads it is kind of important to know why someone is taking a certain stance. Not all opinions are equally well grounded and sometimes you need to know which ones to consider convincing and which ones not. And unlike a real world election where either one candidate - or party - or the other wins, the number of administrators isn't fixed and thus not an either-or proposition, you don't need to compare one candidate to another. I don't agree with using real life elections as a comparator, either; a real life election can have life-or-death consequences while the outcome of any given RfA only affects a website, and not all that strongly. Jo-Jo Eumerus (talk) 18:11, 28 November 2019 (UTC)
  • Mm, RfA is supposed to be a discussion in the system we have set up, and when you oppose a proposal in a discussion, I do believe you should expect to defend your position reasonably from counterarguments (after all, you are presenting them a counterargument). I also don't see it as unfair that oppose votes are responded to more frequently than support votes; someone who makes a proposal is more likely to respond to comments disagreeing with them than comments agreeing with them simply because that is how constructive discussions tend to work (as opposed to "echochambers" and whatnot). In some sense you can view an oppose vote as a response to all of the support votes.
    I would agree that the "badgering" line is crossed perhaps when a single editor starts responding to many opposes in a redundant way (i.e. like a broken record) or when multiple editors pile on against a single oppose redundantly, repeating what each other say instead of adding something meaningful. It is also inappropriate to make comments about the person, and not the comment itselfthat would be a true "attack" on the opposers, in my understanding of the term. However, in general I would disagree with discouraging responses to oppose votes placed directly underneath the oppose vote in question. Mz7 (talk) 23:37, 28 November 2019 (UTC)
    I agree with your comments on the nature of a discussion, but I don't feel that spawning off discussion threads below each initial comment is a necessary condition for having this discussion. To me it's a tragedy of the commons situation: it may be easier for the commenter and the respondent to branch off a conversation directly below the initial comment. But it's much harder for everyone else to follow multiple repetitive threads, which is a barrier for reaching a group consensus. isaacl (talk) 00:18, 29 November 2019 (UTC)
  • Any secret poll method would still bring to the front that the oppose reasons are mainly needed so editors find out whether there is a reason they too need to !vote oppose. Counter-arguments to those opposes, while perhaps intended to convince them not to oppose, are again mainly critical because they persuade those viewing the original opposes. Too many RfA participants would not read both the opposes and then always check the comments for follow-ups. Threaded discussion guarantees that the most people will see the most consideration of concerns and rebuttals to those concerns. Nosebagbear (talk) 23:17, 29 November 2019 (UTC)
  • But that means we are still left with Oppposer being harassed, which creates a negative atmosphere that deters others with good view AND does nothing to deter the more pointless Support !votes that explain nothing and that should be ignored. - SchroCat (talk) 23:28, 29 November 2019 (UTC)
  • Over 1500 editors have voted in the current arbcom elections. They managed to do so without needing to know exactly how anyone else voted. Nosebagbear was among the voters participating in that secret poll. It not only works, it works far better than RfA in generating a high turnout – an order of magnitude better. Andrew D. (talk) 23:38, 29 November 2019 (UTC)
    Is it the secret ballot or is it the fact that we spam every talk page (in addition to the watchlist notice we do for RfA)? Best, Barkeep49 (talk) 06:14, 30 November 2019 (UTC)
    That, and I am not sure if it actually results in better performance. My impression from Arbcom re-election rates is that RfA is a better vetting system than an Arbcom election, which might imply that anonymity reduces the quality of vetting. Jo-Jo Eumerus (talk) 11:21, 30 November 2019 (UTC)