Wikipedia:Requests for adminship/2024 review/Phase II/Discussion-only period/Archive 2

Open discussion

Welcome back! Now that the trial period has concluded, I have a few questions to ask of everyone participating. Hopefully, these make for good yardsticks as we examine the effect of this trial period.

Were these RfAs more or less contentious than usual?
Were these RfAs better or worse for candidates than usual?
Were these RfAs more or less accessible for !voters than usual?
Is there anything else you noticed about these RfAs that positively or negatively impact your perception of the trial period?
Is there anything you would change about the discussion period?
If you were to RfA (again, possibly), would you rather your RfA had the discussion-only period or not?

Cheers, and happy discussing-only! I plan to delay the retain/remove RfC for at least a week, but I'm happy to wait as long as needed for open discussion to hit all the major points. theleekycauldron (talk • she/her) 09:06, 6 September 2024 (UTC)[reply]

We can't support/oppose yet? Is this RFA number six? ;-) –Novem Linguae (talk) 09:13, 6 September 2024 (UTC)[reply]
The discussion-only period probably could have been 24 hours (rather than 48)
— User:HouseBlaster, Wikipedia:Wikipedia Signpost/2024-08-14/Opinion
Aaron Liu (talk) 15:43, 6 September 2024 (UTC)[reply]
Personally, I prefer the 48-hour period, as it better accommodates some discussion across a global community. isaacl (talk) 17:40, 6 September 2024 (UTC)[reply]
But is the benefit of having more time when global peeps can't vote worth the additional stress? Aaron Liu (talk) 17:49, 6 September 2024 (UTC)[reply]
Personally, I believe the tradeoff is worthwhile, which is why I said I preferred the 48-hour period. isaacl (talk) 18:01, 6 September 2024 (UTC)[reply]
I'll raise several concerns that I have, none of which is a fatal flaw, but each of which seems to me to be worth serious thought.
1. It seems to me that the discussion-only period did provide an opportunity for more positive feedback to the candidates, so that aspect was helpful in making the process potentially less stressful. However, it's also important to look at how that played out in RfAs that were neither of the two poles: neither snow-close, nor near-unanimous support. When the RfA was successful, but where there was also significant opposition, I had the impression that the opposition got less serious consideration than it would have gotten under the "traditional" system. Editors tended not to engage so much with expressions of concern voiced during the first two days, as though comments made before the voting started seemed to count less. Editors who expressed concerns during those discussion-only days ended up having to repeat what they said in their "oppose" comments, which makes me question whether the initial discussion was a waste of time, not taken seriously enough. And then the voting period was shorter, which tended to truncate the traditional pattern in which opposes have tended to emerge only after the first few days, perhaps because editors who are inclined to oppose tend to wait for someone else to "go first". If the only goal is to get more successful RfAs, then arguably that's a good thing, but so long as the community consensus is that we don't want to give out the permissions carelessly, then it might become a problem.
2. It felt kind of frustrating, waiting before being able to actually express a support or oppose.
3. After the actual voting began, the RfAs seemed to end much sooner than before, because the overall time was the same (5 days of voting, instead of 7).
4. It might make sense, for the two reasons just above, to have just one day of discussion instead of two. Many editors have said this. On the other hand, does a single day of discussion-only really accomplish anything? Why have a single day at all (other than as an intellectually lazy way to arrive at a "compromise" in this discussion)?
5. We had a prolonged period of not-very-many successful RfAs during the trial period, accompanied by a lot of discussion about whether we are heading towards too few admins. There's no evidence that the trial discouraged any good candidates, but there's no evidence that it didn't, either. If there's an uptick in successful RfAs after the trial period, that would be worth paying attention to.
--Tryptofish (talk) 22:52, 6 September 2024 (UTC)[reply]

From what I see, concerns brought up in the discussion section were extensively discussed. I don't see how oppose !votes referencing/restating comments in the discussion sections means that the section was disregarded. Aaron Liu (talk) 01:34, 7 September 2024 (UTC)[reply]

Well, that just means that you didn't see what I did. I'm trying not to personalize this by naming individual RfAs, but there was one where I laid out my concerns in the pre-discussion, then found that I had to repeat myself in the main event. --Tryptofish (talk) 19:50, 7 September 2024 (UTC)[reply]

Regarding a change in the degree of back-and-forth discussion, I think that was an intended goal: let people raise their points, pro and con, without framing them as part of support or oppose positions, so there's less tendency to reflexively argue against dissenting views. Regarding feeling frustrated at not being able to set forth one's support or oppose viewpoint as soon as it is formed: it's a tradeoff to give everyone some buffer time to evaluate the candidate independently. I trust that many of those who feel that urgency to express their views will be understanding and patient with those who need more time for examination and consideration of the candidate's record. isaacl (talk) 04:15, 8 September 2024 (UTC)[reply]

Everything you describe seems to me to be desirable outcomes. But it all assumes that everyone pays close attention and thinks seriously about the merits. In reality, what I think I actually saw was a significant number of people (not everyone, of course) just paying less attention. --Tryptofish (talk) 20:55, 8 September 2024 (UTC)[reply]

The goals of reducing back-and-forth and giving people time to consider their viewpoints independently are still met, whether it is because participants are paying less attention during the discussion phase or other reasons. It's theoretically possible that if commenters were paying less attention, their analysis missed some points, but whether or not that affected the usefulness of the end outcome can't be evaluated for a while, if ever (since separating out that factor is hard). Personally, I don't think either format is much different in how easily someone can choose to drop in and express their viewpoint without paying attention to the comments of others. isaacl (talk) 21:52, 8 September 2024 (UTC)[reply]

I don't think that the existence of a discussion period makes much of a difference either way. But IMO 24 hours is unthinkably short. Even to just get a comment in would mean an editor having to notice and comment within one of their 8-16 hr. editing periods. And an actual discussion/exchange would mean squeezing 2 or more of those cycles into one 24 hour or less period. North8000 (talk) 01:45, 7 September 2024 (UTC)[reply]
I guess we could make it 36, but that might be too complicated. Aaron Liu (talk) 22:54, 8 September 2024 (UTC)[reply]
To me, the trial illustrated clearly that a brief discussion period does not do anything to fix problems RFA may have, and creates some new problems. Most people with substantive concerns about candidates rarely brought them up, instead reserving them for the oppose section. A lot of people - myself included - posted comments that are !votes in essence. Speaking for myself, it was because I wanted the candidate to feel the comfort of support during an undoubtedly stressful time. And a considerable number of people made comments and posted questions that verged on the inane, which, in my view, happened because Wikipedians feel the urge to participate in open proceedings, but were severely constrained in how they could do so. I am opposed to extending this trial, as I was opposed to starting it. Vanamonde93 (talk) 04:24, 9 September 2024 (UTC)[reply]
As a ORCP frequenter, IMHO this two-day trial period didn't affect the outcome of any of the trial period processes. At first we saw the surprising runs of three unready candidates (none of whom had gotten likes from ORCP commenters) each of whom crashed so hard we were rethinking this very trial, and then we saw runs from five fully qualified candidates, each with little baggage. In a way, I'm glad we had three initial trial period crashes. RfA is not a place where a candidate feels warm and fuzzy. It is not a place to seek validation. Is is a place where each and every one of our previous behaviors may be closely scrutinized, however trivial. I remember TLC suffering because of a non-standard archiving experiment. Even some of our finest admins did not pass RfA on the first try. IMHO, we shouldn't be doing anything to make it more pleasant to RfA. It's a job interview. BusterD (talk) 11:13, 9 September 2024 (UTC)[reply]
There is consensus that RfA has problems. Aaron Liu (talk) 11:32, 9 September 2024 (UTC)[reply]
I see nothing incompatible between these positions. A job interview for a lifetime post obviously SHOULD be challenging, daunt job seekers, and quickly screen out the grossly unready. RfA is not a platform for mere popularity. This proposal, based on our test cases, doesn't appear to help us in any measurable way. This proposal didn't meaningfully affect any of the test cases. This proposal wouldn't have affected my choice to run. This is my opinion. BusterD (talk) 11:58, 9 September 2024 (UTC)[reply]
I should add, IMHO this proposal was intended to make early opposes less damaging, not to correct all RfA injustices. At this point, I'd prefer to identify and remove recidivist "corrosive" troublemakers from the process. My solution might be unpopular but be more measurably effective. Again, my opinion. BusterD (talk) 12:10, 9 September 2024 (UTC)[reply]
How would that address such opposes? No individual RfA proposal is supposed to fix everything, and such standards would be unfairly harsh. Aaron Liu (talk) 17:31, 9 September 2024 (UTC)[reply]

There is consensus that the level of scrutiny is problematic, and all this does is have the scrutiny discussed. If you think it didn't change anything, why not keep it, especially with the potential protection against unproportionately-spoiling opposes? Aaron Liu (talk) 17:29, 9 September 2024 (UTC)[reply]
At this point surely until and unless there's an uptick (or a downtick) in RfAs, as Tryptofish suggests above, we have no evidence other than anecdotal to evaluate the trial. My anecdotal evidence is that the trial caused me, twice and for two different reasons, to step closer than i have ever been to making a Request: First, at the beginning it seemed as though no one was going to, so i considered it exactly so we'd have some evidence then, after two or three unsuccessful RfAs i felt that i might step forward as someone closer to the average candidate than the first few; fortunately, Elli and then several more did so. Much more important than mine, though, would be the experience of the candidates who did go through this process; while i recognise that some or all mayn't want to rehash, i'd very much like to hear from any of ToadetteEdit, Numberguy6, DreamRimmer, Elli, Pickersgill-Cunliffe, HouseBlaster, and Asilvering on their views of the experience. Happy days, ~ Lindsay^Hello 13:36, 9 September 2024 (UTC)[reply]
I personally liked the idea of the discussion-only period, and the proposal to end it early resulted in me running when I did. I've been meaning to write a debrief with my overall thoughts, but haven't been sure quite how to put everything together. My opinions are overall positive though. Elli (talk | contribs) 13:43, 9 September 2024 (UTC)[reply]

My thoughts on the discussion only period was divided. The pro is that it allows the editor to rethink twice before voting. The con though is it can put the candidate into much more stress, as what happened in my application where people were bringing up every ounce of my mistakes which forced me to stay away from Wikipedia for a whole day. In my opinion the discussion only period should continue. ToadetteEdit (talk) 14:48, 9 September 2024 (UTC)[reply]

Well, I'm not sure what to make of the delay between HouseBlaster's RFA and mine - was it scaring off other potential candidates? We go with gaps that long sometimes, so who knows. I did delay the start of my own run in the vague hope that someone else would take the last spot, but obviously it didn't stop me from running entirely. I'm also not sure that the "Wikipedians feel the urge to participate in open proceedings" thing Vanamonde mentions make much difference for my RFA in particular - ie, I'm not convinced that I wouldn't have had basically the same experience with questions if voting had opened at the outset. Perhaps I'm just well-trained by academic conferences, where people ask you mind-bendingly irrelevant questions all the time, some in earnest good faith and some not at all. The game is to come up with a much more interesting question in your head and answer that one instead, without unintentionally communicating to the questioner, who probably has no idea that their question is inane, that they said something stupid.

My RFA wasn't contentious, so I don't know that my experience can say much about whether the discussion period helps with keeping the temperature down. I do think, though, that framing concerns as a "topic for discussion", like happened in my RFA, did really help in that case. Other editors responded and disagreed, but because they could disagree by saying something like "eh, I don't think that's a concern" or "I agree with asilvering here" rather than effectively saying "I disagree with YOU PERSONALLY because of YOUR VOTE", it looked much less like hounding than lots of responses to oppose votes have in the past.

What I didn't like about it was that it added another "beginning of your RFA" point, and that's just unnecessarily exhausting no matter what emotions you feel about it. -- asilvering (talk) 17:36, 9 September 2024 (UTC)[reply]
I thank the three of you for those responses; to me they indicate that if the discussion only period was designed to remove (or reduce) stress in the candidate it has failed, but also that it has positive points in allowing discussion rather than hounding or bitter argument. (ping for thanks) Happy days, ~ Lindsay^Hello 07:26, 14 September 2024 (UTC)[reply]
I think we've learned that a discussion-only period isn't helpful.—S Marshall T/C 22:37, 10 September 2024 (UTC)[reply]
I feel like it might have made a minor difference in whether discussion was contentious. Which was the point, as I really do think contentious discussion, particularly in response to opposes that aren't well thought out and thoughtfully expressed, is one of the things that makes RfA so unpleasant. BusterD, I completely disagree that we shouldn't be trying to make RfA more pleasant. Most job interviews don't include up to two questions from your worst work enemy and the intern who started yesterday in front of a peanut gallery of hundreds who are bickering amongst themselves within your hearing 24/7 for a week. Valereee (talk) 11:04, 14 September 2024 (UTC)[reply]
The utility of the discussion-only period was best exemplified by ToadetteEdit's RfA avoiding caustic oppose votes. Once the discussion showed no possibility of success, ProcrastinatingReader's WP:NOTNOW close avoiding subjecting a valued editor to explicit opposition, instead leaving them with constructive suggestions for future eligibility. As for the length, I do not check WP:CENT on a daily basis and would therefore like to retain at least two days of discussion. Tryptofish notes that longer discussions eat into the voting period, but I do not see the downside since before this trial period, RfAs virtually never flipped in their consensus between the fifth and seventh day. BluePenguin18 🐧 ( 💬 ) 21:18, 14 September 2024 (UTC)[reply]
The primary downside that has been raised with a period shorter than 7 days for expressing support/oppose/neutral viewpoints is that it won't give those who only edit once a week (or perhaps only allocate one day a week to work on non-content-related tasks) the opportunity to weigh in. I agree that historically the data seems to indicate that there isn't much effect at the tail end of the RfA period, which suggests that either there isn't many people who want to participate and fit that profile, or that they look at the tallies and decide not to participate. Even so, I think it's still desirable to give occasional editors the opportunity to participate should they wish. isaacl (talk) 21:56, 14 September 2024 (UTC)[reply]

Although it may depend on how one defines "flipped" (completely changing direction, versus continuing in a direction already started), I think it's simply not true that RfA consensus has never changed "between the fifth and seventh day". For RfAs that have failed narrowly, by which I mean to exclude those that were SNOW failures, but which, instead, ended up near or in the Crats' discretionary zone, it has been fairly common for an RfA that was just barely passing around days 4–5 to move into narrowly failing by day 7. That's because it takes a while for opposition to develop. The first really substantive oppose arguments are often not posted until a few days in, and then other editors may need a few days to decide to agree with those opposes, or to switch from support to oppose. I recall Crat Chats where they noted shifts in sentiment towards the end of the discussion, sometimes in a negative direction, and other times in a positive one. With a shorter voting period, we truncate that process in its middle. And yet it seems to me that a consensus that comes into focus only after multiple editors have considered and responded to one another's arguments is a stronger one, than one in which they haven't, for lack of time. --Tryptofish (talk) 22:25, 14 September 2024 (UTC)[reply]
Discussion about a candidate's characteristics has already been ongoing by the time the viewpoint expression period begins, though. This should enable the support/oppose/neutral statements to start at roughly the same place as with the process without an exclusive discussion period. isaacl (talk) 22:47, 14 September 2024 (UTC)[reply]
Oh, FFS, near the beginning of this discussion, I already addressed that. I shouldn't have to repeat myself. --Tryptofish (talk) 23:00, 14 September 2024 (UTC)[reply]
I appreciate you have a different view than I on how much attention is ongoing during the exclusive discussion period. I apologize for voicing my disagreement again without making reference to it. isaacl (talk) 23:06, 14 September 2024 (UTC)[reply]

Went through all pre-trial 2021-2024 RfAs and only found two cases showing major shifts between the fifth and seventh days. In MB's RfA, we went from 80.89% support to only 68.38% as editors raised concerns about deviance from WP:BEFORE, ultimately ending with bureaucrats finding no consensus. In Tamzin's RfA, we went from 96.94% support to only 75.22% as editors strongly disavowed desysopping Trump supporters, ultimately ending with bureaucrats finding consensus to promote. I respect Tryptofish's view that even if we provided seven voting days, having a prior discussion may have some unavoidable dampening on organized opposition, but I am willing to accept that in exchange for civility. BluePenguin18 🐧 ( 💬 ) 02:07, 16 September 2024 (UTC)[reply]
Again, you have defined the terms in a particular way, and that gave you the result you were predisposed to finding. So let me put it this way: let's set aside "major shifts", and look instead at the overall course of discussions, with attention to the give-and-take between differing views that can give rise to a more solid decision in the end. And especially, let's set aside a narrow focus on days 5–7, and instead consider what happens in days 1–5, and whether any of that continues through days 5–7, because days 5–7 don't exist in a vacuum. Well, the trial for the pre-discussion is over, and now, we have a new RfA, under the "old" system. Here is a permalink to it, as of the beginning of the day today, a little under two days in: [1]. The !vote numbers were 76/0/0. And here is a permalink at the time that I post this, during day 3: [2]. Now, it is 89/5/0. That's a change, and it didn't happen until the third day. For about two days, this looked like unanimous support. Today, one person wrote an oppose with a rationale, and then other editors also opposed, agreeing with the first oppose. Where will consensus end after day 7? I don't know (and I haven't decided yet how I will !vote, myself). Maybe opposition will grow, or maybe the oppose rationales will be convincingly refuted. But right here, we have empirical evidence of opposers waiting to post, and some change in direction part way through. This pattern has long been typical of RfA, and the trend has been for opposition to emerge late. So what do we gain, and what might we lose, if we decrease the opportunities to refute opposes, and decrease the opportunities for opposition to grow?

You said a lot when you said, "having a prior discussion may have some unavoidable dampening on organized opposition, but I am willing to accept that in exchange for civility." I object to calling the opposition "organized", because it isn't. In the current RfA, I doubt that you can present evidence of canvassing or meatpuppetry among the editors who have opposed so far. And you frame it in terms of civility. Civility, really? Can you provide any diffs that demonstrate incivility in the oppose comments? This isn't about civility, and it really never has been. Successful RfA candidates end up having to deal with real civility problems, that go far beyond anything that happens at RfA. This is about discouraging opposes. And you pretty much admitted it. I've been here long enough to remember when RfA had very low standards, and we got admins who ended up being disasters. Wikipedians who don't remember history will be doomed to repeat it. --Tryptofish (talk) 18:47, 16 September 2024 (UTC)[reply]
You claimed it has been fairly common for an RfA that was just barely passing around days 4–5 to move into narrowly failing by day 7. That's because it takes a while for opposition to develop. This thread has been about days 5 to 7 from the start since the discussion period only eats 2 days. Days 1 to 5 are pretty much irrelevant to this argument. Aaron Liu (talk) 20:34, 16 September 2024 (UTC)[reply]

Also, I think the civility issue Blue meant is people badgering opposers that "appear out of nowhere" for "ruining the count", discussions that result in nothing but wasted energy, as opposed to having discussion beforehand that usually forewarns opposition, with a notable exception being when the opposer in question is a certain L-named individual. Aaron Liu (talk) 20:42, 16 September 2024 (UTC)[reply]
I'm trying to explain how discussion develops over the entire period, and the two days that get "eaten" have an effect on what does or does not get made use of from the days that came before. As for the problem of badgering opposers, that's a different issue, because what was trialed here has nothing to do with it. The discussion beforehand does nothing to "forewarn opposition", because editors who badger will still badger when someone opposes. And that individual you refer to is actually supporting in the current RfA, and also supported in the one before that. --Tryptofish (talk) 20:57, 16 September 2024 (UTC)[reply]
I'm trying to explain how discussion develops over the entire period, and the two days that get "eaten" have an effect on what does or does not get made use of from the days that came before.
How do days 1-4 affect this at all? If you're arguing this, you should be showing results substantially changing between days 5 and 7, which Blue correctly analyzed.
editors who badger will still badger when someone opposes.
From what I see in the trial's RfAs, people who laid their near-full argument at discussion below were not badgered, or at least were badgered significantly less than their prior opposes. An example of this is your own oppose in HouseBlaster's RfA. A ton of Support voters referenced your argument and heavily disagreed with it, but nobody came to badger you.
For L, I'm talking about how he once expressed support on day 1 before opposing on day 3 during the trial. That's the only instance of someone supporting under discussion and instigating an oppose argument during !voting I've seen and is under my context of predictability. Aaron Liu (talk) 01:56, 17 September 2024 (UTC)[reply]
I'm not going to try to explain to you what's going on in your first and third points, because I've already explained it. As for your second point, I had been trying to avoid pointing to specific RfAs, but since you mention HouseBlaster's, I'll go there. (He and I have already had a very good talk about it on my talk page, so I don't think it's unfair to him for me to discuss my oppose, now.) I expressed concerns during the discussion-only period. When I subsequently opposed, based on those reasons (and when other editors opposed based on what I had said during the discussion), there were responses from editors who obviously had never read the preliminary discussion. This forced me to repeat things I already said, and it caused other editors who opposed to have to explain that they based it on what I had said earlier. That's a perfect example of what I pointed out earlier in this discussion, about how editors often seemed to ignore the early discussion, and only start paying attention during the voting period, which is a bad thing. As for whether or not that stuff was badgering/incivility, you can see below what BluePenguin says was actually meant by that. --Tryptofish (talk) 21:48, 17 September 2024 (UTC)[reply]
@Tryptofish, do you think it would help if we had some kind of guideline that required all substantial comments to be in a discussion section, such that votes themselves (or maybe votes more than a basic "not jerk, has clue") were things like "see my comments in discussion section"? -- asilvering (talk) 21:58, 17 September 2024 (UTC)[reply]
Hmm, no, actually. I suppose, in retrospect, that I could have done more along the lines of saying, in my oppose rationale, "see what I said earlier in the general discussion (link)". But overall, I think we should always be doing more to encourage more (polite) discussion, and less of the basic !voting. In the end, I'm really coming down on the side of not wanting to continue this particular trial. The more we've been discussing it here, the more I've become convinced that it was a net negative. --Tryptofish (talk) 22:05, 17 September 2024 (UTC)[reply]
Hm, that's an interesting outlook to have in the context of the current RFA, where it appears to me that you're feeling badgered by the responses to your neutral !vote (which, at least as I read it, is more of a "current placeholder for my opinion" than a "true final assessment of the candidate"). It's sort of leading me to think that discussion is going to feel unpleasant to at least some people no matter where it is or how it's handled. -- asilvering (talk) 22:38, 17 September 2024 (UTC)[reply]
Intelligent discussion is better than unintelligent. --Tryptofish (talk) 22:45, 17 September 2024 (UTC)[reply]

Could you point me to edits where you had to repeat things you already said and the comments from editors who didn't read? I've looked through all of your edits to House's RfA since 02:37, 19 June 2024, which is when you added your oppose !vote, and found nothing.
Also, my point was that nobody badgered you. I also do understand now that's not what Blue meant, but IMO it was an effect.
And, if you're willing to, you could also point me to where you explained against my first and third points if you want. To clarify my first point, your premise is to set aside a narrow focus on days 5–7, and instead consider what happens in days 1–5. I'm saying that that premise is false because whatever happens in days 1−5 will happen during days 3−7 under the trial anyway. Yes, days 5−7 are usually effects of the first five days, but that doesn't mean what happens in five days of voting should matter in evaluating the discussion-only period since it always happens either way.
This tiny bit of magnification only mattered 2 times out of about 50 times. Yes, it decreases the opportunities to participate, but that tiny ~4% is too small to matter, especially in the face of Wikipedia:Admin reconfirmation. Aaron Liu (talk) 22:22, 17 September 2024 (UTC)[reply]

By "organized opposition," I simply meant sufficient opposition to block the RfA, not a malicious accusation of canvassing. I am not discouraging opposition, especially considering that I was the fourth oppose vote in the ongoing RfA, and my prior objection to premature drafticiation was referenced in Joe's initial oppose vote. Regarding incivility, the most recent support vote is Generalissima writing that she doesn't "mind the perennial opposes," attacking editors like myself as predisposed to vote against RfA candidates. As you allude to, it takes longer to write substantive opposition by investigating editor logs than to write "no big deal" and support. Where we differ is my belief that the discussion period assists both support and oppose voters by allowing them to critically analyze the candidates before they dig in based on how they voted on the first day. BluePenguin18 🐧 ( 💬 ) 00:05, 17 September 2024 (UTC)[reply]
OK, I think we understand one another, thanks. --Tryptofish (talk) 21:36, 17 September 2024 (UTC)[reply]

That the negative comments were not formally labelled as opposes did not make the RfA much less toxic, if at all - for my part I found it roughly as hurtful as any other quickfail. The problem isn't with the formalism, the problem is the way in which people make their points. A discussion-only phase isn't going to fix this. Java Hurricane 00:26, 22 September 2024 (UTC)[reply]
Bad people are still going to be bad, and this proposal has no intention of fixing that. (Rather, it's the job of the designated monitor.) However, what it does intend to fix is inertia. Aaron Liu (talk) 02:12, 22 September 2024 (UTC)[reply]
The (admittedly limited) evidence we have suggests that the two-day discussion hasn't done its job well enough, then. A case in point would be DR's RfA - the comments section looked like this at the start of the voting phase - mostly positive with only a couple of concerns. The RfA ended up being a dumpster fire that was withdrawn at near 50%. It's just one case, of course, but it's gone a similar way as the "typical" RfA that crashes and burns beginning a day after it starts. Just that for the first two days here, the "supports" were not officially labelled as such.

As I said, this is a people problem - oppose voters being too snarky/brusque, or early voters not doing enough homework on the candidate, among other things. I doubt discussion-only periods can fix this issue. Java Hurricane 03:41, 22 September 2024 (UTC)[reply]
I’m delighted the discussion period is over, and I hope it doesn’t return. Reading through a stream of comments from people all saying they are going to vote in support of a candidate for the two days before they then repeated the same comments was a complete waste of time. As was reading through the stream of ‘not now’ comments for the candidates who then withdrew. These were both sets of overt pseudo-votes which cluttered without providing any enlightenment - just let the voting open on day one and let people vote. - SchroCat (talk) 04:58, 15 September 2024 (UTC)[reply]
But you're still reading the comments instead of laying down impetus before any potential opposing grounds have been discovered. Aaron Liu (talk) 12:00, 15 September 2024 (UTC)[reply]
What an odd—and incorrect—comment. I read the pseudo-voting comments for the first two candidates, then realised what a complete waste of time it was before I stopped for the all the other five. My comment here was based on a retrospective analysis. - SchroCat (talk) 11:10, 16 September 2024 (UTC)[reply]
Overall I think the separate discussion period did no harm but the positive effects I hoped for have not materialised. Should we try this again (I think we should, especially for the admin elections), I would suggest to make the discussion period more of a structured evaluation of the candidate. Two days of "great candidate, can't wait to support" are not as useful as separate evaluations of AfD, AIV and NPP performance or an in-depth dive into the candidate's GAs. —Kusma (talk) 12:34, 15 September 2024 (UTC)[reply]
Sure, any commenter who would like to break down their evaluation into different aspects or just cover one is welcome! (This would start to approach my 2015 proposal to orient discussion around the strengths and weaknesses of the candidate.)

Regarding the upcoming admin election: an exclusive discussion period was part of the approved proposal. isaacl (talk) 17:15, 15 September 2024 (UTC)[reply]
I think we need to do more to encourage commenters to make comments that will be more useful to others. —Kusma (talk) 10:24, 16 September 2024 (UTC)[reply]
I think one of the achievable ways to do this is by example. For example, you can break down your comments into separate sections (with headings) covering different characteristics. isaacl (talk) 16:53, 16 September 2024 (UTC)[reply]
Yes, but that requires someone to do the actual work :) —Kusma (talk) 15:17, 17 September 2024 (UTC)[reply]
Sure; you're also suggesting that some actual work be done, right? ;) isaacl (talk) 16:51, 17 September 2024 (UTC)[reply]
I believe the trail was a positive experience. I can't speak for other editor's but after I do my investigations and !vote I rarely remember to come back in the days after and read what other editors have to say. During the trial I read every support and oppose and all the ensuing discussions as well as answers to any questions asked after my initial interaction period. I support this becoming a permanent feature. ^Thanks,L3X1 ◊distænt write◊ 15:00, 15 September 2024 (UTC)[reply]
It's hard to form concrete conclusions from such a small sample size, but overall I thought the trial period was a success and I am in favor of making this change permanent. There's no real harm in making editors wait a couple of days to register their !vote while allowing time for members of the community to present a fuller picture of the candidate. The old system often results in dozens of editors taking the nominator(s)' word for it before any real scrutiny takes place, and I consider that to be a bug, not a feature. LEPRICAVARK (talk) 16:11, 15 September 2024 (UTC)[reply]
I am in general agreement with Kusma. ~~ AirshipJungleman29 (talk) 10:53, 16 September 2024 (UTC)[reply]
I share the sentiments above of those who didn't find it helpful. -Fastily 22:51, 17 September 2024 (UTC)[reply]
It turns out that requiring opposers to provide arguments against a candidate is not helpful. It's already a problem with the "not a vote" idea behind requiring rationales for RfA oppose votes, but it has worsened when "not a vote" was actually enforced. I'm eagerly awaiting the experiment in the other direction – admin elections, finally allowing people to silently oppose for any reason, hopefully making RfA less toxic. ~ ToBeFree (talk) 08:11, 18 September 2024 (UTC)[reply]
Note that the enforcement is the result of Wikipedia:Requests for adminship/2024 review/Phase II/Designated RfA monitors, a separate proposal. Aaron Liu (talk) 11:51, 18 September 2024 (UTC)[reply]
I think the enforcement they're talking about is just the 2-day block on votes. -- asilvering (talk) 16:14, 18 September 2024 (UTC)[reply]
Well, I slightly doubt that since people aren't required to comment under discussion. Aaron Liu (talk) 16:44, 18 September 2024 (UTC)[reply]
Personally, I liked it. It gave users a chance to put what they had on the table before anyone voted, which is how most real elections work. I would like to see it clarified that obvious snowball/nonnow cases can be closed during this period as we saw some of that and they were left open until voting began, which was to nobody's benefit. Just Step Sideways ^{from this world ..... today} 22:21, 21 September 2024 (UTC)[reply]
Wikipedia:Requests for adminship/ToadetteEdit was closed before voting. Aaron Liu (talk) 22:54, 21 September 2024 (UTC)[reply]
Personally feel that this made little practical difference. As such, I don't really see the point of keeping this for future RfAs. Java Hurricane 00:20, 22 September 2024 (UTC)[reply]
As a historical proponent of RfA reforms I agree that this experiment seems desperate and unhelpful with no discernible benefits. A mandate for informal discussion preceding formal consensus building does not sit well with me, and the fact that the review discussion itself seems to take for granted that this is the way, and that an editor may deem that initial feedback is not to formally mean anything real, is not only nonsensical and arbitrary, but insulting to the fundamental system of governance of the project. ~Swarm~ ^{sting} 07:29, 27 September 2024 (UTC)[reply]
I don't think this was better, by splitting the phases it may prevent some contributors who are not active daily. — xaosflux ^Talk 09:44, 27 September 2024 (UTC)[reply]
It's a minor matter IMO. The big issue for me is that open voting is still too toxic for me to want to participate in it and so I'm still waiting for a secret ballot which will make this more of a safe space. Andrew🐉(talk) 09:49, 27 September 2024 (UTC)[reply]
As others have said, I don't think it necessarily had any positive effect on making RfAs less contentious. As with elections in the real world I think most people will have quite clear and unchangeable positions, which discussion will do little to affect. Furthermore, I think the idea that it can "prevent" people who are thought of as probably unfit from even doing an RfA subverts the system. Their success or failure should be up to the election, not a chosen few who discourage them from going forward. I'd assume that if someone puts themselves forward, they're prepared to handle a potential loss. ᴢxᴄᴠʙɴᴍ (ᴛ) 09:50, 27 September 2024 (UTC)[reply]
I definitely think we should keep it... I'm not seeing any major downsides from what's happened to far; sure, in some instances such as easy passes it didn't add any particular value but in one or two cases it can clearly be shown to be beneficial. Wikipedia:Requests for adminship/ToadetteEdit, for example, was withdrawn peacefully before the possibility of an oppose bloodbath could begin, and hopefully this makes for a more positive experience for Toadette in the long run, giving them the knowledge required for a successful re-run in the future. I see the two-day wait period as an improved or alternative version of WP:ORCP myself (and personally I like ORCP since it gave me the confidence to have a run for the real thing). And that's to be applauded. — Amakuru (talk) 09:59, 27 September 2024 (UTC)[reply]
@Xaosflux, Andrew Davidson, Zxcvbnm, and Amakuru: Sorry for a bit of a mix-up! I've archived the open discussion section, since we're moving to structured discussion. Feel free to repeat or move your comments in the relevant section. theleekycauldron (talk • she/her) 10:04, 27 September 2024 (UTC)[reply]