Wikipedia talk:Requests for adminship/Optional RfA candidate poll/Poll candidate search/Archive 1

Setting it up properly
Okay, so the Asynchronous Admin Score tool looks for edit count, blocks, account age, existence of userpage, user rights, pages created, and activity. I'm not sure how to set up the flow page to fit with that. Are the sections in the right order? Are some missing? Is this whole method dumb? Thoughts? Anna Frodesiak (talk) 00:39, 10 April 2017 (UTC)

How to list names
or....

?

Anna Frodesiak (talk) 02:47, 10 April 2017 (UTC)

Question about non-admins
, can non-administrators add their names? -- Marvellous Spider-Man  01:46, 10 April 2017 (UTC)
 * I think everybody should be allowed to do anything they think is reasonable. This is a wiki, after all. :) Anna Frodesiak (talk) 01:52, 10 April 2017 (UTC)
 * I mean, I added my name. RileyBugz Yell at me  &#124; Edits  03:02, 10 April 2017 (UTC)
 * Ah, okay. I misunderstood. Yes, of course! That is fine. Anna Frodesiak (talk) 03:03, 10 April 2017 (UTC)

Another way to do this
How it looks in edit mode:

Thoughts? Anna Frodesiak (talk) 02:57, 10 April 2017 (UTC)


 * First way, unless they move to the AfDs or beyond section, in which case it should be switched to the second way. This is so that we can get more info for candidates that are more likely to actually try, and so that we don't clutter up the page. I don't like the third way, the other ways make categorization easier and it makes it easier to navigate. RileyBugz Yell at me  &#124; Edits  02:58, 10 April 2017 (UTC)


 * Fair enough. Cheers. Anna Frodesiak (talk) 03:51, 10 April 2017 (UTC)

Initial discussion
...is here. Anna Frodesiak (talk) 05:00, 10 April 2017 (UTC)

CSD taggings?
Shouldn't there be a section for evaluating CSD taggings? I'm aware that I have been called a bit strict at times when it comes to candidates' speedy deletion work but then again, oftentimes the community agreed with me and we don't really need to ask people to step forward who have problems in that area just to be shot down later. Regards  So Why  09:21, 10 April 2017 (UTC) PS: On that note, I'm trying to write a tool that automatically scans a user's CSD log (created by Twinkle) or their contributions for speedy taggings and displays whether the page was deleted and if so, if it was deleted because of the speedy tagging or for another reason. If someone more familiar with writing API-using tools feels like they could do it and is interesting in doing it, I'd be grateful since I have only rudimentary PHP skills. Regards  So Why  09:48, 10 April 2017 (UTC)
 * Yeah, I think that there should be a section for that. Maybe we should have to vet them for CSD tagging before AfD vetting? I think that the process for CSD vetting can involve bots, but I think that humans will have to be involved too. This is because it seems, to me at least, that some tags are more "permanent", as in even if the article is changed wholesale, those tags are still applicable. Thus, I think that we should value these tags more when vetting candidates. The best example of this tag is A7. So, we should hope to see a good percentage of articles tagged with a permanent tag to be deleted under the same permanent tag. Thoughts? RileyBugz Yell at me  &#124; Edits  11:34, 10 April 2017 (UTC)


 * Agree. Especially with A7 oftentimes articles end up being deleted through PROD or AFD, meaning despite the redlink the tagging was incorrect. A bot or tool could filter such taggings and then display those where tagging and deletion reason diverge. Regards  So Why  12:12, 10 April 2017 (UTC)
 * And, conversely, I suppose the opposite applies too- think of say, G12s which were correctly deleted but then restored sans cv- they would incorrectly show as incorrectly tagged. In fact, every successful challenge at WP:DRV cloud be a eventual example. Basically, there's a nuance that I'm not sure a bot would recognize? &mdash;  O Fortuna   velut luna  12:23, 10 April 2017 (UTC)
 * While I dont disagree that CSD tagging is a useful metric, we should be wary that we don't start requiring candidates to be using Twinkle or otherwise logging their tags. I'm sure plenty of good admin candidates, who do lots of CSD tagging, don't use Twinkle or keep a list. There have also been plenty of successful admins who didn't do much if any CSD work. Sam Walton (talk) 12:42, 10 April 2017 (UTC)
 * Agree with the first part but a bot could scan the (deleted) contributions for edit summaries that indicate such taggings. Not perfect but pretty useful nonetheless. As for the last part, there is probably no admin who has not interacted with speedy deletion at all and this metric should be to vet candidates who do much speedy work. As such, I don't think "does speedy work" is required to pass but "if they do speedy work, it's good". Regards  So Why  12:49, 10 April 2017 (UTC)
 * Sounds reasonable :) Sam Walton (talk) 12:59, 10 April 2017 (UTC)
 * I have never had a CSD log; however I have had User:Ritchie333/saves, which is kind of an anti-CSD log. Ritchie333 (talk) (cont)  15:41, 10 April 2017 (UTC)
 * ...and the last thing that I saw as I lay there on the floor was painted by an atheist nutter on a British Rail poster reading 'have an away day'.... :)

Asynchronous Admin Score and previously blocked users
Ping Bri.

The Asynchronous Admin Score excludes previously blocked users, right? But some were joke blocked for 1 second and some self requested during exam time. Anna Frodesiak (talk) 03:38, 10 April 2017 (UTC)


 * It seems best to try and manually vet those people. RileyBugz Yell at me  &#124; Edits  03:39, 10 April 2017 (UTC)


 * You may be right. Anna Frodesiak (talk) 03:40, 10 April 2017 (UTC)

Thoughts? Anna Frodesiak (talk) 03:40, 10 April 2017 (UTC)
 * It doesn't exclude them, it just deducts points (the script clearly believes in rehabilitation!) &mdash;  O Fortuna   velut luna  09:04, 10 April 2017 (UTC)
 * Fortuna Imperatrix Mundi is correct, the penalty for having a block keeps most editors with that from being "approved" by the Asynchronous Admin Score tool. But I noticed a couple in my scan: Exec8 for instance, also Zigzig20s who has a very high score. - Bri (talk) 14:42, 10 April 2017 (UTC)
 * Would it be possible for the tool to exclude blocks that were less than a few minutes (e.g. 30 minutes) in duration? This might cut down on most erroneous blocks. Mz7 (talk) 14:49, 10 April 2017 (UTC)
 * as in blocks that were removed in minutes? That'd be nice... CHRISSY MAD  ❯❯❯  ¯\_(ツ)_/¯  18:01, 10 April 2017 (UTC)

The list of what we are vetting for
We don't have a simple list. This could be helpful as a checklist, for sequencing, for the record, to add to, etc., so here goes:


 * Civility
 * Account age
 * Edit count
 * Recent activity
 * Blocks
 * AfD
 * Backpage participation
 * Content creation
 * Already took the poll
 * Name exists in the holding pen

Anna Frodesiak (talk) 22:01, 10 April 2017 (UTC)

Civility vetting
I would like to add an extra stage to the vetting process, for civility. It is a standard check I have done when looking for candidates. A candidate can pass all the other areas, but if they write "Delete - no sources, couldn't find a thing, fucking spammers like this shouldn't be allowed to breathe let alone edit Wikipedia", they won't be suitable to be an admin.

The general heuristic is I look through the candidate's talk page, and archives going back about six months, and then search for their name on ANI. If I find helpful replies and a generally positive atmosphere, that's a big plus. If I see lots of unanswered questions, particularly from newbies, that's a minus as I wouldn't trust them to honour WP:ADMINACCT. If I see clear incivility, that's a complete fail. Ritchie333 <sup style="color:#7F007F">(talk) <sup style="color:#7F007F">(cont)  15:50, 10 April 2017 (UTC)
 * Thanks, that is an essential check. RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  16:34, 10 April 2017 (UTC)


 * I totally agree, Ritchie333. Should we ensure that all who get added to the raw list are already vetted for civility? If so, then these scripts or tools that will generate raw list names can check that. However, this raises the question about who can add names? Can anyone just come along and add several to go through the process. Anna Frodesiak (talk) 21:58, 10 April 2017 (UTC)


 * Okay, I see civility has been added as a section. I feel that a minimum number of sections is good. Fewer steps. Combine civility with something else? Early vet to ensure they don't make the raw list? We can do this with a statement at the top of that section like "If you are adding items here, please ensure they are civil by searching...." In any case, incivility should be an early deal-breaker and be higher on the vet list, no? Anna Frodesiak (talk) 22:06, 10 April 2017 (UTC)
 * I agree with combining civility vetting with the raw list, as uncivil editors should not become admins. RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  22:09, 10 April 2017 (UTC)

Okay, I'll boldy go ahead and remove that section and add a sentence to the raw section. Please revert if you disagree. Anna Frodesiak (talk) 22:40, 10 April 2017 (UTC)
 * Actually, it might be better to create a dispute vetted section, and then include stuff like no violations of the three-revert rule, and then combine civility into that. RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  22:48, 10 April 2017 (UTC)
 * I'm really reluctant to add sections. More work could cause all this to die. In fact, we hope lots of these sections will be eliminated as queries and such get developed. Anna Frodesiak (talk) 22:51, 10 April 2017 (UTC)

Low risk
All that being said, this is about encouraging them to take the poll, not RfA. The poll will catch incivility. If we wanted to be super-thorough, we'd be posting suggesting they actually run. Anna Frodesiak (talk) 22:49, 10 April 2017 (UTC)

Readiness
Is it alright to send in its current form? Anna Frodesiak (talk) 07:48, 10 April 2017 (UTC)
 * "If you have already taken the poll recently, please disregard this notice." Is that part necessary? Wouldn't it be easy enough to check whether that editor has already run? I'd be happy to compile a list of everybody whose run over the last 12 months, say. An offer to answer any initial queries might be a good thing to add to the standard notice.  Schwede 66  10:04, 10 April 2017 (UTC)


 * Good thinking, Schwede. I've removed that text. It was just so much work vetting, I didn't take that last step. Anna Frodesiak (talk) 11:14, 10 April 2017 (UTC)
 * I'm not sure the addition of "so far" in the first sentence is desirable; it seems to temper the appreciation begin offered with a caveat. isaacl (talk) 05:28, 11 April 2017 (UTC)

Corpus of new page reviewers
I've put everyone in Category:Wikipedia new page reviewers in the raw list to sift through. Essentially, everyone there has asked for the "patroller" right via WP:PERM and got it. This is a low bar of 90 days activity and 500 non-deleted edits; however everyone on the list is doing maintenance stuff that is good experience for adminship. If we "divide and conquer" this lot, we should hopefully sort the wheat from the chaff. <b style="color:#7F007F">Ritchie333</b> <sup style="color:#7F007F">(talk) <sup style="color:#7F007F">(cont)  14:27, 12 April 2017 (UTC)
 * - surely self-requested blocks don't count? How do they indicate a poor behavioural pattern? It was exams :) &mdash;  O Fortuna   velut luna  14:51, 12 April 2017 (UTC)
 * Well I won't get upset if you revert them back in! <b style="color:#7F007F">Ritchie333</b> <sup style="color:#7F007F">(talk) <sup style="color:#7F007F">(cont)  14:59, 12 April 2017 (UTC)
 * Self-requested blocks in recent history would be received poorly by the community, in my opinion. If it was in the last 6 months, I would question whether they can handle the stress of adminship. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 15:06, 12 April 2017 (UTC)
 * Converseley, it could be argued that they would be demonstrating the maturity required in an admin, fully appreciating, and in the spirit of, WP:5THWHEEL :) Anyhow, I don't think it would be a deal-breaker for almost any candidate, going by the frequency with which it is raised. As to reverting - well.  &mdash;  O Fortuna   velut luna  15:15, 12 April 2017 (UTC)
 * Oddly enough, I'm not in the new page reviewers category. Also, I would suggest readding those who have self-requested blocks that were imposed to enforce a wiki break. RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  15:22, 12 April 2017 (UTC)

New section at holding pen
We should probably add a section on the holden pen page for those that don't pass the vetting, but we think would pass an RfA anyways. RileyBugz <sup style="color:#D7000B;">Yell at me &#124; <sub style="color:#D7000B;">Edits  22:05, 10 April 2017 (UTC)
 * An interesting idea, but unrelated to this post, I think. Make it into a new section? Anna Frodesiak (talk) 22:15, 10 April 2017 (UTC)
 * I made this into own section called "New section at holding pen" in good faith. Anna Frodesiak (talk) 23:08, 14 April 2017 (UTC)

Orphan
This is an orphan. Thoughts? Anna Frodesiak (talk) 20:56, 15 April 2017 (UTC)

Centralised discussion page
There is currently a discussion going on here about the possibility of having one talk page for all discussions related to this project. If you'd like, you can voice your opinion there.  Anarchyte  ( work  &#124;  talk )   04:21, 16 April 2017 (UTC)

Centralised discussion page?
At the moment every subpage has its own talk page. Wouldn't it be better for them all to redirect to here, where everything related can be archived together?  Anarchyte  ( work  &#124;  talk )   08:49, 13 April 2017 (UTC)


 * Hmmmm, maybe a good plan, Anarchyte. I can see pros and cons. Anna Frodesiak (talk) 23:10, 14 April 2017 (UTC)
 * I'd appreciate this; there's a lot of discussion on "Poll candidates" talk page that I'd like to follow, but I don't want to follow the "Poll candidates" page itself. isaacl (talk) 22:21, 15 April 2017 (UTC)
 * There are currently two redlinked talk pages, one and two, that probably won't get that much discussion on them anyway. Same goes for this one. I don't see why we'd need four talk pages for one project.  Anarchyte  ( work  &#124;  talk )   04:24, 16 April 2017 (UTC)
 * Agreed; accordingly, I'd appreciate having all discussion consolidated on this talk page. isaacl (talk) 05:46, 16 April 2017 (UTC)


 * Agree with the proposal. Too many pages varying degrees of activity just makes it easier to miss something. Keep it all in one place. &mdash;  O Fortuna   semper crescis, aut decrescis  05:31, 16 April 2017 (UTC)
 * Alright, I'm gonna redirect and move the content from the three talk pages I mentioned earlier, but I'll leave this one for now because it's the most popular one and it'd be better to get more opinions before moving all of the content on it.  Anarchyte  ( work  &#124;  talk )   05:58, 16 April 2017 (UTC)
 * Fair enough, my friend. Thanks for being bold. It seems like a good plan. Anna Frodesiak (talk) 22:43, 16 April 2017 (UTC)

Any opposition to moving the contents and redirecting the remaining talk page? Pinging all users listed as participants:.  Anarchyte  ( work  &#124;  talk )   10:10, 17 April 2017 (UTC)
 * I love the idea. Now I can unwatch all of those pages. RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  11:28, 17 April 2017 (UTC)
 * Ditto here. -- Iazyges   Consermonor   Opus meum  11:57, 17 April 2017 (UTC)
 * Seems like a good idea to me. ThePlatypusofDoom (talk) 12:09, 17 April 2017 (UTC)
 * Alright, there's been no objection to this idea, so I'm gonna merge and redirect everything now.  Anarchyte  ( work  &#124;  talk )   12:12, 17 April 2017 (UTC)
 * Not much opinion here.  J 947 (c)  20:14, 17 April 2017 (UTC)

Maintenance work as a criteria
I think that we might be using our time more wisely if we actually focus on people who do "maintenance" work, as that might be the most important criteria. Just look at the recent polls as ORCP. Thus, I think that we should possibly attempt to limit our search to those that do maintenance work. This does not mean adding another vetting process, but instead encouraging people to place those that they have seen doing maintenance work. Thoughts? RileyBugz <sup style="color:#D7000B;">Yell at me &#124; <sub style="color:#D7000B;">Edits  16:52, 17 April 2017 (UTC)


 * That's a pretty good idea, RileyBugz. Anna Frodesiak (talk) 21:13, 17 April 2017 (UTC)

Threshold for final vetting
I am erring on the side of inclusion rather than holding pen. My reasons are:


 * This is only for a poll, not an actual RfA.
 * Any low poll score comes with suggestions for improvement. This could set them on the road to another poll or RfA.

Anna Frodesiak (talk) 03:08, 14 April 2017 (UTC)


 * Nah, otherwise it will bog up the page. Also, a holding pen shows what the candidates need to improve on. RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  19:05, 14 April 2017 (UTC)


 * Hi RileyBugz. Bog up the poll page? There are only new 6 entries per month there. Anna Frodesiak (talk) 23:07, 14 April 2017 (UTC)


 * Ok, not a very good argument, but it is so that we also don't keep going over the same names. RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  23:38, 14 April 2017 (UTC)
 * I agree; the regulars at ORCP actually advise against coming back too soon, and indeed mark candidates who do accordingly lower. So we should not be advising candidates to just keep submitting requests- it will work against them. &mdash;  O Fortuna   semper crescis, aut decrescis  05:29, 16 April 2017 (UTC)
 * As I understand it, Anna isn't saying not to track who has been previously invited to start a poll; she is saying she prefers to invite an editor who is a borderline pass to start a poll for the first time. isaacl (talk) 05:55, 16 April 2017 (UTC)
 * Well; she might have been saying that I suppose. Cheers, &mdash;  O Fortuna   semper crescis, aut decrescis  13:12, 16 April 2017 (UTC)


 * Hi isaacl and Fortuna Imperatrix Mundi. Indeed, I meant "...prefer to invite an editor who is a borderline pass to start a poll for the first time..."


 * Anna Frodesiak (talk) 21:21, 17 April 2017 (UTC)

What does "vetting" actually entail?
(paging Anna Frodesiak although I assume you have this page watchlisted.) The idea behind "use an automated metric to weed out the no-hopers and spot the potential high achievers" is sound in theory (and assuming you're using this tool I can hardly complain about a metric which AFAIK claims to provide empirical proof that I'm the single best editor in the history of Wikipedia*). However, I do have some concerns that the way this is currently being run might actually drive people away. At the moment, people seem to be failing "vetting" for completely spurious reasons, while obvious no-hopers are passing vetting and being sent invites. For obvious reasons I'm not going to name those I consider no-hopers being inappropriately accepted, but to give an example of a candidate who'd be a near-certainty at RFA failing vetting, consider Rodw. Here's someone with 6000 WP space edits (or to put it more clearly, roughly twice as much back-office participation as you) who's founded an active Wikiproject from scratch and is an active volunteer for WMUK who's literally taught the course on how academics can work with Wikipedia, but he's been summarily dismissed by your vetting process as Did not pass constructive backpage participation vetting. If I were someone in that position and saw myself being arbitrarily dismissed as unimportant in that way, my reaction would be something along the lines of "fuck the lot of you then, I'm off". Given the number of names you now have on your shit-list, even if only one in twenty of them actually take offence and walk out it's still going to cause a significant hit to Wikipedia since by definition these are highly active editors. I don't want to piss on anyone's chips, and I do support the intent here, but I do get the feeling that this is trying to automate something which is inherently subjective, something even the billion-dollar algorithms of Google and Facebook struggle pitifully to manage, and in its current form as the potential to annoy a lot of people very quickly once the ball really starts rolling. At the very least, you need to have a clear and consistent explanation for what you actually mean by "blocks vetting", "AfD participation vetting", "civility vetting", "constructive backpage participation vetting" and "content vetting". &#8209; Iridescent 22:02, 15 April 2017 (UTC) * At least, I'm unaware of anyone who scores higher than me.


 * Hi Iridescent. The Asynchronous Admin Score tool is/was used to create raw list entries. The other vetting after that is done manually. It is new, subjective, and not an exact science. Your constructive input into how to improve things is welcome.


 * About Rodw. I added his name in this edit after searching like this. It shows little or no recent posts at AN/I, AIV RFPP, etc. Have I made an error there? Would he pass an RfA without recent work in those areas?


 * As for posting where the user has, as you put it "no hope", well, the user can get feedback from the poll and get on the road to preparing for an RfA. I posted about that here. And frankly, if there is a "no-hoper" in the list of post recipients, I must have missed that. Recent successful RfAs have included users with practically zero AfD votes, and that used to be a complete deal breaker.


 * Anyhow, you make a good point about upsetting people. What would you suggest? Should all this be scrapped and the pages deleted? Is there a way that this can go forward without upsetting people? Anna Frodesiak (talk) 22:29, 15 April 2017 (UTC)
 * My attitude at RFA has always been primarily about the social aspects. I don't care if anyone's active at AFD etc unless they say they want to work in those areas (IIRC my total comments at AIV at the time I passed RFA was zero), or doesn't give the official answer to that obnoxious "problematic username" question that's started popping up again in RFAs unless they've actually given an indication that they have the slightest interest in usernames. What I look for can basically be boiled down to "does anything in this person's past indicate that they can demonstrate empathy for people who disagree with them and are behaving erratically owing to stress?"; people who slavishly adhere to the letter of the law because they don't understand its nuances, or who throw blocks around without trying to calm the situation, are the ones who cause damage. (Bad deletions or pagemoves can easily be reverted, resignations can't.) These social factors are very hard to measure, and Snottywong's tool doesn't even try. (As an exercise, I've just put the name of one of Wikipedia's most notoriously objectionable editors through it, and that editor scores a very respectable 800+.) If you want concrete examples of the community accepting "isn't active in admin areas but would nonetheless make a good admin since their actions elsewhere show they can be trusted and won't act recklessly, and thus even if they only make a few admin actions it will still be a positive", then Ealdgyth's RFA is a good recent example.
 * I wouldn't go as far as deleting it, but I'd festoon it with very prominent "If a name is listed here it doesn't necessarily mean someone is considered good, and if they're not listed here it doesn't necessarily mean they're considered bad" disclaimers. If I were setting it up from scratch, I'd do away with the whole "holding pen" setup—which, whatever the intent, makes the whole thing look like you're separating the wheat from the chaff; if you must keep lists of names, I'd suggest a single list of everyone that's been suggested and inviting all of them, rather than trying to second-guess voters as to who they'll consider "constructive".
 * If you're going to rely on Snottywong's tool, it would probably make more sense to run it over every bluelinked name at WP:WBE and invite every single person scoring more than a set target, even if you look at some of the names and think "good god, no". The no-hopers will with any luck have enough self-awareness to make an "I never wanted to be an admin" excuse (as appears to already be happening), and if not, as you say they'll just be shot down at ORCP. (The WMF will scream blue murder when they notice the server load generated, but by then it will be a fait accompli since it's an exercise that only needs to be run once.) Even if you have no published list of names at all, once people start seeing the invites popping up on other editors' talkpages at least some of them will take offence that they haven't received one as well. This is not just an angels-on-pinheads thought experiment. It may have been before your time, but take a look at WP:ACPD and the RFC that killed it to see just how much bad blood anything that smacks of "sending out invitations to a selected group" can generate on Wikipedia. &#8209; Iridescent 23:03, 15 April 2017 (UTC)
 * Iridescent, I will respond in point form because you make so many points:
 * If you're going to rely on Snottywong's tool, it would probably make more sense to run it over every bluelinked name at WP:WBE and invite every single person scoring more than a set target, even if you look at some of the names and think "good god, no". The no-hopers will with any luck have enough self-awareness to make an "I never wanted to be an admin" excuse (as appears to already be happening), and if not, as you say they'll just be shot down at ORCP. (The WMF will scream blue murder when they notice the server load generated, but by then it will be a fait accompli since it's an exercise that only needs to be run once.) Even if you have no published list of names at all, once people start seeing the invites popping up on other editors' talkpages at least some of them will take offence that they haven't received one as well. This is not just an angels-on-pinheads thought experiment. It may have been before your time, but take a look at WP:ACPD and the RFC that killed it to see just how much bad blood anything that smacks of "sending out invitations to a selected group" can generate on Wikipedia. &#8209; Iridescent 23:03, 15 April 2017 (UTC)
 * Iridescent, I will respond in point form because you make so many points:
 * Iridescent, I will respond in point form because you make so many points:


 * Bloody good points, and well put!
 * Losing the holding pen was something I was about to suggest, but then what stops the names from ending up in the raw list again?
 * Maybe we should get rid of the sections in the holding pen and make the lead say '"...for closer examination..." or something like that.
 * I too have little faith in the Asynchronous Admin Score tool. I'd prefer raw list items come from elsewhere. But considering that there is more vetting afterward, does it matter?
 * Your attitude at RfA is the same as mine. However, this is all about the community's attitude and why they support and oppose.
 * Maybe in the invitation and lead of these project pages should read "...If you are reading this and interested in becoming an admin, you are all invited and encouraged to take the poll..."
 * I am happy to see someone use "festoon". It is a good word.
 * ACPD makes me think that this Poll candidate search‎ project shouldn't exist.
 * The fact that this is only about an invitation to take a poll makes me think that maybe this should exist.
 * I have no idea where to go from here.


 * Anna Frodesiak (talk) 23:24, 15 April 2017 (UTC)
 * Off the top of my head in the middle of the night, so treat this with some caution:
 * Run the tool over WP:WBE (or even WP:WBRE if it's in one of its rare phases of actually working); maybe also WP:WBFAN and other "active editor list" type pages;
 * Take the 500 people with the highest scores;
 * Use massmessaging to invite them all, without exception (the "without exception"—except for those who are already admins—is important, even if it means inviting people who haven't edited for six months, took the poll two weeks ago, or have just come off a checkuser block).
 * Although this will reach a lot of wildly unsuitable people, and not reach some suitable ones, it means that it will appear on multiple talkpages being watched by anyone who's active and engaged with the community, and consequently every viable admin candidate will see it. Because you've explicitly just invited the top 500 based on an arbitrary criteria (this is where the "without exception" is important), you're not making a conscious decision not to invite anyone, and thus there isn't the potential for people to be offended as there's no element of blackballing. &#8209; Iridescent 23:34, 15 April 2017 (UTC)
 * (Adding) If you want another lesson in ancient wikihistory, the Association of Established Editors and subsequent RFC are also worth reading. If you really have a strong stomach for people bitching about invite-only groups, head on over to WP:ESPERANZA and start following links. &#8209; Iridescent 23:37, 15 April 2017 (UTC)
 * I don't think we should be afraid of subjectivity. As long as it's made complete clear that the search is indeed subjective, editors should be understanding about not being invited to start a poll. After all, nothing stops them from starting a poll on their own initiative. I'm not a big fan of inviting everyone who meets some arbitrary metric; I think having personal judgment involved is a good thing. isaacl (talk) 00:16, 16 April 2017 (UTC)
 * You raise good points, and this is the concern I've had in the past about a process like this being public - if, for whatever reason, you put people on your "I don't want to nominate you" list, it creates all sorts of potential for offending them, even if the reason is minor. Sam Walton (talk) 22:33, 15 April 2017 (UTC)
 * I think the involved parties should take care to highlight wherever possible that the so-called vetting isn't the final word; it typically just represents one person's opinion. From the other direction, I hope anyone on these lists will realize this as well, and understand no process will be anywhere close to perfect. From a third direction, I hope anyone reading these lists won't use them to denigrate others, appreciating that this is really just a very rough screening mechanism; lots of good candidates may slip through, and some poor ones will get identified as potential administrator candidates. In the spirit of wikis, I trust that we will take as much benefit from incremental steps as we can, even if they aren't perfect. isaacl (talk) 22:38, 15 April 2017 (UTC)

I suggest immediately making a lead at the list and holding pen saying that decisions are subjective, and that users ought not be offended because intentions are good, that sort of thing. Anna Frodesiak (talk) 22:42, 15 April 2017 (UTC)
 * That WP:ACPD is interesting, and I can see the (unfortunate) relevance here. The phrase "" does rather jump out at one :)  &mdash;  O Fortuna   semper crescis, aut decrescis  05:55, 16 April 2017 (UTC)

I don't like the idea of listing all the failed candidates on a page, even if they do get readded to the list later. , the comment Losing the holding pen was something I was about to suggest, but then what stops the names from ending up in the raw list again probably applies here. A user being added to the list could be perceived in a few main ways: It's, as many of you have said, subjective and to disallow the same person to be vetted twice sort of defeats the purpose of the page, in my opinion. The page is used to find possible candidates for RfA and to say that if you failed the vetting once you can't go again is foolish. Though with this said, I don't think that it should be deleted, instead, modify it. One way to do this would be to make it so it's cleared very month or so because it's a) going to become very long, and b) it's going to limit the possible candidates even if they've improved since the last vetting, which shouldn't be what that list is for. It should also be noted that it only takes one user to remove the possible candidate from the whole vetting process, even if they pass every single other criteria (some are more important than others, such as the civility vetting). makes a good point here, why should they need to pass a certain part, such as AfD, if they don't plan to work there if they become an admin? I can see both for and againsts for that, but that's been discussed to death all over the place. I like the overall idea of this mini-project, but it could be executed better and possibly one of the first steps is to allow the same user to be vetted more than once (though there should probably be a max amount, such as once a month or so, bringing us back to my point of cleaning the holding pen every month).  Anarchyte  ( work  &#124;  talk )   13:45, 16 April 2017 (UTC)
 * 1) Multiple people believe that this user has a chance at RfA
 * 2) If they get rejected once, they'll get rejected again.
 * 3) If they're readded, maybe they've improved since their last vetting
 * I think it may be useful to have distinct search rounds, so it's clear the holding list primary role is just to track who isn't being invited in the current round. Also note that this search is a collaboration: any one can invite any other editor to participate in a poll. So if you think someone on the holding list should be asked, go for it! If you think you have a better screening process, go ahead and sort through a raw list of candidates your own way. There's room for many approaches. isaacl (talk) 14:22, 16 April 2017 (UTC)

Some more responses in point form:
 * Thanks to everyone for working to sort this out. Your shaping  this is very wiki.
 * This project could bring at least a few admins each year that otherwise wouldn't be. That is worth something.
 * However, stuck in my mind is Iridescent's "Don't upset people." We must do no harm.
 * We cannot do some harm because it is worth it. There must be the teensiest amount of irked users and zero users leaving.
 * The invitation is highly visible so it must make non-invites feel okay. Maybe we can turn it in our favour by having it contain a special message for non-invites that invites the non-invites. I'd prefer a coloured box within a box. I'll make something. Revert if you don't like it.
 * A holding pen to convenience us in exchange for upsetting people should not stand. Would off-wiki lists be appropriate?
 * Invite-only groups are horrifying, as pointed out by Iridescent. This is invite but not really a group, and not "invite-only". Also, we have other groups like 100,000 edits. We also have invite posts to groups that anyone can join. So, we must just avoid any exclusivity feel in this project. Festoon pages with "...take the poll, really, we want you to take the poll..."

Festoon this page with your thoughts. Anna Frodesiak (talk) 23:18, 16 April 2017 (UTC)
 * I think the key is, as you said, this is not an invite-only list. If editors A, B, and C decide they want to invite editors 1, 2, and 3, this doesn't stop editor D from deciding that editor 4 is worthy of inviting. And personally I think editor 5 should be able to understand that not passing screening by editors A to D doesn't mean their work is not appreciated, and that screening is highly subjective, so if they want to start a poll, they should feel free to do so.
 * Obviously, if you don't feel comfortable with the process as it stands, you should certainly change it to one that you prefer. But I think we should be careful not to underestimate the ability of editors to recognize that the search project is a highly subjective process that doesn't hamper them from pursuing greater involvement in English Wikipedia, should they so choose. I agree with continuing to highlight (as I have added in a couple of locations) that everyone is free to decide for themselves if they would like to start a poll; no invitation is needed. isaacl (talk) 23:36, 16 April 2017 (UTC)
 * Here's a different take on it. I'm placing it here as an FYI rather than any kind of warning. &mdash;  O Fortuna   semper crescis, aut decrescis  09:32, 17 April 2017 (UTC)
 * Perhaps Really simple guide to requests for adminship or something similar should be a link in the invitation (right now it is only in the new proposed message targeted at people reading the invitation on someone else's page). If there is an unexpectedly high response rate to the invitations, there is a risk of swamping the poll page (I hope that the regular poll commenters will be understanding of a little increase in open polls, even if some seem premature). To combat this, prospective candidates should be encouraged to do some self evaluation of their chances of successfully requesting administrative privileges and how they would use them for maximum benefit, as they should be in the best position to evaluate themselves. isaacl (talk) 18:21, 17 April 2017 (UTC)
 * Hi isaacl. Sure, link Really simple guide to requests for adminship. That post is hungry for copy editing. Have at it. As for the poll being "swamped", well, that would be great, but I doubt this poll candidate search effort will do that. Admins, and even people taking the poll, are hard won. I've been posting at wikiproject pages for over a year, and it doesn't do much. Best, Anna Frodesiak (talk) 21:17, 17 April 2017 (UTC)
 * I don't really expect there will be a problem with a huge flood of polls, but I confess I didn't think the candidate poll would be as active as it has been, either :). So here's hoping there will be a happy landing between the two extremes. isaacl (talk) 22:55, 17 April 2017 (UTC)

I picked up this process as an extension of the stuff I do manually, just to see if I would ferret out any possible admin candidates I hadn't thought of. Unfortunately, the whole thing seems to have confirmed my theory that there are few people I would personally like to nominate right now, as I haven't found those who've turned up on the poll to be particularly strong candidates. Indeed, I've had a chat with a few regulars off-wiki about some more suitable RfA candidates who didn't make it through the vetting process here.

Essentially, I think this was a worthwhile experiment, but it hasn't really got us any further towards getting any good admin candidates, so I suggest we knock the whole thing on the head and chalk it up to experience. <b style="color:#7F007F">Ritchie333</b> <sup style="color:#7F007F">(talk) <sup style="color:#7F007F">(cont)  15:12, 19 April 2017 (UTC)
 * I think it is a good experience with trying to spread the workload of screening potential candidates. Perhaps the table-based format Anna proposed would be better than the multiple queue approach, which I think overemphasizes the elimination of candidates versus getting an overall picture of them that can be evaluated. I hope that the involved parties will continue now that some momentum has been built. With a little patience, I think a reasonably effective process can be honed that distributes the effort. isaacl (talk) 15:25, 19 April 2017 (UTC)
 * Hi isaacl. I'd be happy to switch to the table format and give it a try. And yes, perhaps a bit of patience is required. Cheers. Anna Frodesiak (talk) 20:13, 19 April 2017 (UTC)


 * Hi Ritchie333. I'm not sure we can call this a failure just yet. My reasoning:


 * It is only ten days old. Thirty two posts were sent out. One in eight has taken the poll. Not bad.


 * I do not see this as an extension of what you and others do manually. Kudpung, you, and many others, find exceptional users and tell them they are ready for RfA. I see this as a parallel process. We sent messages to any potential who is at least satisfactory. The poll lets them know of their immediate chances, and gives feedback about what to do to have higher chances.


 * The aim is that, if the poll score is not comfortably high enough for them to run immediately, they work the right areas and then run in, say, six months. So, as this project is only ten days old, it won't have produced new admins yet.


 * I am quite happy to plug away at vetting if others can continue to provide raw list items. I understand some scripts are being made now to do that.


 * The future? If we can get a decent handful into the poll each month, it is reasonable to think it should result in a nice handful of new admins each year that otherwise wouldn't be. Considering the quickly declining numbers, this project seems reasonable. Anna Frodesiak (talk) 20:13, 19 April 2017 (UTC)

The fate of the holding pen
Can we please get a decision on the fate of the holding pen? (There are opinions on it above here and there. Search "holding pen".)
 * A: Keep it.
 * B: Delete it.
 * C: Keep it, but remove the section headings.
 * D: Something else

Anna Frodesiak (talk) 18:39, 18 April 2017 (UTC)

I am boldly blanking it for now. Anna Frodesiak (talk) 06:18, 19 April 2017 (UTC)
 * I would just archive it how it is with the headers and add to the archives when names get stale. Regards, — Moe   Epsilon  00:05, 20 April 2017 (UTC)
 * Hi Moe Epsilon. But how is an archive page different from the holding pen, which is essentially an archive? Anna Frodesiak (talk) 20:23, 20 April 2017 (UTC)

So, we really need this resolved. Keeping a record of those removed from the main list is needed to prevent same work being done multiple times. How do we do this without offending people? This really needs to get sorted out. Anna Frodesiak (talk) 20:23, 20 April 2017 (UTC)
 * We keep doing what we're doing. Potential admin candidates should take advice wisely and not react strongly.  J 947 (c)  20:42, 20 April 2017 (UTC)
 * Hi J. Are you suggesting I restore the holding pen? Anna Frodesiak (talk) 20:45, 20 April 2017 (UTC)
 * Yes.  J 947 (c)  20:55, 20 April 2017 (UTC)


 * Hi Anna. Maybe you could just retain the holding pen without the subheaders then and make it a simple table with names and the link to them being removed from the list of candidates. This method should be largely unoffensive. You could link the usernames or not, either way.


 * {| class="wikitable sortable" style="font-size: 90%;"

!Username !Link
 * 333-blue
 * 
 * Manxrule
 * 
 * 
 * }
 * 
 * }
 * 
 * }
 * }


 * Regards, — Moe   Epsilon  20:45, 20 April 2017 (UTC)
 * Hi Moe Epsilon. So, you're suggesting the diff as a way to record the stage of vetting they didn't pass? I must say that is quite a bit of extra work. What would you say to either a plain list or a list with those old sections. Anna Frodesiak (talk) 20:56, 20 April 2017 (UTC)
 * If you need the list, then keep the list. You can call it something non-committal like "candidates deferred for future rounds". If you move to the table format you had proposed earlier, then all the entries could just be left in the table, with an "invite sent" column that gets checked when an invite is sent. When the current round of invites is done, the table can be archived, and a new one started. The new table could have columns for different criteria, to catch different types of potential candidates. Those doing the screening could resurrect any names they wished to the next round. isaacl (talk) 20:49, 20 April 2017 (UTC)
 * isaacl, this makes the best sense. Anna Frodesiak (talk) 20:56, 20 April 2017 (UTC)

I suggest isaacl's table idea. Objections? Anna Frodesiak (talk) 20:56, 20 April 2017 (UTC)

Idea
How about we redirect the holding pen to the candidates and mark crosses on things that they fail? e.g.

 J 947 (c)  21:54, 20 April 2017 (UTC)


 * Hi J. See below. I didn't want to do all that work. :) Anna Frodesiak (talk) 22:04, 20 April 2017 (UTC)

Boldly sorted out
The problem was how to handle names that did not vet. Here is what I did per advice above and my own judgement. I hope it is okay:

The holding pen content is restored and should be complete. It is needed so that we do not vet the same names again and again. It contains:
 * A section April 20, 2017 and earlier, not in table format. Too much work to do that.
 * An empty table for April 21, 2017 and later. Names from the main list can go there in table format.
 * A raw list of all names on the page to conveniently cross-check against new, incoming items for the main raw list. There is also a tool provided there to do that check.

I have switched the main list to table format. There needs to be a way to indicate names that do not vet without adding a giant, offensive. Maybe a subtle full stop in the last column? Anna Frodesiak (talk) 21:58, 20 April 2017 (UTC)
 * May be instead of having a (my idea), we could just put nothing in columns with which the candidates fail in.  J  947 (c)  22:04, 20 April 2017 (UTC)


 * Hi J. But if we put nothing after we check it, then others could come along and check the same thing. If a stage of the vetting ends in a fail, we need to indicate that. Anna Frodesiak (talk) 22:06, 20 April 2017 (UTC)


 * Dandy. Looks like we've settled on a full stop. Anna Frodesiak (talk) 02:08, 21 April 2017 (UTC)

Removal of candidates
So, I suggest that we move people from the list to a general area before vetting them. In that area, participants would be able to remove names from the list as they see fit. Thoughts? RileyBugz <sup style="color:#D7000B;">Yell at me &#124; <sub style="color:#D7000B;">Edits  00:41, 10 April 2017 (UTC)
 * The first section is the "before vetting area" right? We can remove unsuitable names from any place and any stage, right? Anna Frodesiak (talk) 00:46, 10 April 2017 (UTC)
 * But will the bot re-add them? RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  01:18, 10 April 2017 (UTC)  there is no bot... is there
 * Ok, so we are allowed to remove names if we believe that they would not pass, anywhere? RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  01:20, 10 April 2017 (UTC)
 * Sure. Fine with me. Anna Frodesiak (talk) 01:23, 10 April 2017 (UTC)
 * Should I remove candidates if they don't pass on to the next area? RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  02:59, 10 April 2017 (UTC)

Jumping in here, but please check for User wikipedia/Anti-Administrator or similar. I had a look at a couple of random names on the page and found one. – Train2104 (t • c) 04:48, 10 April 2017 (UTC)
 * One thing I noticed – Cullen328 already polled on ORCP sometime back. I don't think he should be bugged with a Talk page notice – he scored excellently at ORCP last time, but decided not to run. Indeed, with any name you generate here, it'll be worth it to check that user's Talk page archives and make sure they weren't already asked to do RfA and turned down the offer. Just some thoughts... --IJBall (contribs • talk) 02:29, 11 April 2017 (UTC)


 * I'm fine with adding or removing candidates at will but I boldly added an editnotice that requests that people use edit summaries when they remove a candidate so others don't have to guess why. For example, I added a candidate and removed them without explanation, so I have no idea what this candidate's problem was. Regards  So  Why  13:27, 21 April 2017 (UTC)


 * It was AfD.  J 947 (c)  18:51, 21 April 2017 (UTC)

Good vetting, let's get it sorted out
I am out of touch with what draws supports and opposes at an RfA. (Read if you wish. The bottom line is that I'm not that good at it.)

If a user doesn't have any AfD, but is great in all other areas, could they still pass an RfA?

If a user doesn't make an AIV or RfPP posts, could they still pass an RfA?

Please help get guidelines sorted out for vetters.

Anna Frodesiak (talk) 23:50, 22 April 2017 (UTC)
 * Both of those are possible, although, in my view, AIV and RfPP have more of an influence than AfD. CSD tagging, although, is the thing that seems to be looked at the most. Overall, we are just trying to find people who would surely be good for adminship. We aren't, although, trying to find everybody that would be good for adminship. Overall, our current system is good, although is does need a bit of refinement. RileyBugz <sup style="color:#D7000B;">Yell at me  &#124; <sub style="color:#D7000B;">Edits  00:01, 23 April 2017 (UTC)
 * Agree. As I mentioned above, CSD tagging should definitely be a factor to check against and one I would help with. Incorrect speedy deletions are much more likely to hurt the project in terms of driving away new editors and removing content that should be included without much overview. For example, 7 out of my (admittedly low number of) 33 DYK credits were for articles that I rescued from the speedy queue (roughly 21%!), i. e. articles that another admin might have deleted. As such, I have (prominently) opposed a number of RfAs on bad speedy tagging and probably helped sink one or two of them. I thus see no use in vetting candidates here if we don't include issues a significant number of editors consider important enough to oppose. Regards  So Why  11:23, 24 April 2017 (UTC)

Hi RileyBugz and SoWhy. Fair enough. So, what's the best way to check for CSD tagging? And SoWhy, I did not understand your last sentence. Cheers. Anna Frodesiak (talk) 22:00, 24 April 2017 (UTC)


 * A high ratio of declined taggings or egregious errors in applying the criteria are imho sufficient to derail an RfA, e.g. A7 taggings of ineligible or clearly even notable subjects, which show that the candidate has not understood the speedy deletion policy. I have encountered a number of editors while patrolling CAT:CSD (who I don't want to name now) whose approach to CSD will probably (and hopefully) sink any future RfA. Basically, for vetting here we should flag those candidates who clearly misunderstand the point of CSD, otherwise this vetting is not really useful. Which is what I meant with my last sentence. Regards  So Why  07:02, 25 April 2017 (UTC)


 * Hi SoWhy. Understood. That makes sense. Now, is ther an efficient way to do searches? Best, Anna Frodesiak (talk) 18:52, 26 April 2017 (UTC)


 * Depends on the user in question. Most use tools to nominate pages for speedy deletion and you can search their contributions for the standard edit summary (e.g. "requesting speedy deletion" for Twinkle). For the others, I have no idea, one would probably need a tool to check the contributions for certain additions... That shouldn't prevent us from adding a CSD column to the list at this time. Regards  So Why  20:29, 26 April 2017 (UTC)
 * SoWhy, thank you. So, in checking contribs, seeing many CSD tagging being reverted should mean they don't pass the vetting. The absence of CSD tagging is still okay, I gather? Case-by-case, right?


 * Now, is there any way to avoid adding an extra column? Keeping a minimum of cols is really something I advocate. Could that just be part of backpage contribs? How about a vetting guideline section at the page? In fact, I'll add that now to get it started. Best, Anna Frodesiak (talk) 20:36, 26 April 2017 (UTC)


 * If it's clearly part of a certain col, I'm not against it not having a separate col although I think it would be more fitting to combine AfD and CSD into one column since both are deletion-related. Btw, you seem to have forgotten a bit of your sentence there. Regards  So Why  12:13, 27 April 2017 (UTC)
 * Hi SoWhy. Yes, CSD/AfD together is good. I like that. Oh, and that bit of sentence was just stray text, now removed. Cheers. Anna Frodesiak (talk) 23:00, 29 April 2017 (UTC)

Need more raw list items
All items on the raw list have been done. We need more. Please populate that section.

(Summary: Thirty two users received the post. One in eight subsequently took the poll.)

Thank you kindly.

Anna Frodesiak (talk) 21:44, 19 April 2017 (UTC)


 * I can't think of anyone else who I'd put forward. Well, that's not quite true, there are people like who probably know exactly what the admin toolset entails and have done for years, but I fear they'd find it patronising to be lumped on this list. <b style="color:#7F007F">Ritchie333</b> <sup style="color:#7F007F">(talk)  <sup style="color:#7F007F">(cont)  12:31, 20 April 2017 (UTC)
 * The people over at AfC could be vetted, though they've got a reasonably low bar of entry, IIRC.  Anarchyte  ( work  &#124;  talk )   12:34, 20 April 2017 (UTC)


 * With the current system, i. e. everyone just being able to remove candidates who failed any step, it's quite hard to check who has not already been named. I think we should change this first before adding new names in large numbers, lest we have to do the same work multiple times. Regards  So Why  20:05, 20 April 2017 (UTC)


 * Hi SoWhy. Good point, and that has been discussed and not resolved. Moe Epsilon, in the section above, thinks just archiving is the best plan. Would you be a dear and weigh in there? Anna Frodesiak (talk) 20:19, 20 April 2017 (UTC)


 * Anna, maybe you can find and search for more candidates at List of administrator hopefuls. TheGeneralUser (talk) 19:53, 29 April 2017 (UTC)
 * Hi TheGeneralUser. I acutally checked for the heck of it a few weeks ago. It was hard to find any that still edit. Plus, that template is often mocked and I have read about those who display that template "...exactly who we do not want..." What we need is the script people to generate big, fat lists based on some boffinesque algorithm. Anna Frodesiak (talk) 00:10, 30 April 2017 (UTC)

2 proposals
I've got two new proposals for the project: Thoughts?  J 947 (c)  02:33, 27 May 2017 (UTC)
 * 1) If the candidate being vetted fails AfD but passes all other checks the notice will still be given.
 * 2) New column: Edit count and tenure. This will prevent me being passed. :) A tenure of a year and 10,000+ edits is what I'll be standing by. I'm also standing by a 70% AfD rate.
 * I like the first, but I don't like the second. The tenure would be good, although, one year specifically. RileyBugz <sup style="color:#D7000B;">会話 <sub style="color:#D7000B;">投稿記録  22:00, 28 May 2017 (UTC)

We have a raw list
Continued from here:

Okay, I'm going to ping the users involved in past discussions at Poll candidate search talk archives.

Anna Frodesiak (talk) 21:32, 28 May 2017 (UTC)

Pinging the rest:

Friends, what do you think of this last query (provided here by Cryptic)? Anna Frodesiak (talk) 21:35, 28 May 2017 (UTC)
 * In its current state, this seems like a bad idea. So, I propose some extra needed qualifications:


 * 1) At 50 edits in the last two months. This is to check if they are active or not.
 * 2) Check to see that they have (this will check to see if they edit in adminy areas):
 * 3) 25 edits to AIV OR
 * 4) 25 edits to UAA OR
 * 5) 25 edits to RFPP OR
 * 6) 25 edits to SPI and its subpages (since one actually does sockpuppet investigations on the subpages)
 * 7) Make sure that they have not been blocked (unless the block has been undone within one hour) for any time during the previous year.
 * 8) Check to see that there is not a link to the users page on the holding pen page.
 * If these things were added, I would think it better. The problem is the fact that there are a lot of people on that list, and I doubt that most of them would pass an RfA. RileyBugz <sup style="color:#D7000B;">会話 <sub style="color:#D7000B;">投稿記録  21:59, 28 May 2017 (UTC)


 * For my part, I still don't think that these kinds of stats are any kind of reasonable indicator of whether someone would make a good administrator. At best, you'll find Wikipedia addicts fortunate to have found the wiki a long time ago (even if 9500 of their 10004 edits were in the past month), without any kind of indication that they understand policy, can distinguish a good source from a regurgitated press release, or successfully assume good faith of someone cut-and-pasting a company ad into a new article.A better place to look would be pick your favorite administrative process page and look for non-admins there who say sane things.  AFD is probably your best bet; my own preferred haunt, DRV, is esoteric enough that most of the non-admin regulars have already largely retired from articlespace, so would go over like lead balloons at RFA. —Cryptic 22:07, 28 May 2017 (UTC)
 * Understood. Is there a way to query for good AfD participation and high % of green? And what about thinking outside the box, like the existing query list then checking user talk archives for key phrases like "excellent work"? Anna Frodesiak (talk) 22:15, 28 May 2017 (UTC)
 * Well, to be honest, not many people care too strongly about AfD. Really, I think that it would be best to remove it, as this is searching for people who would likely pass an RfA and make a pretty good admin, not just those who would make a good admin. RileyBugz <sup style="color:#D7000B;">会話 <sub style="color:#D7000B;">投稿記録  22:16, 28 May 2017 (UTC)
 * There are several good tools, in need of adoption perhaps, on Scottywong's user page that are still quite useful; like, for example: Admin score which seems to no longer generate a valid score, per its instructions, yet it does generate a report that shows edit count, block log statistics, and user permissions with quickness and ease.--John Cline (talk) 22:41, 28 May 2017 (UTC)
 * , I made apersonbot/aadminscore, a resurrection of Scottywong's tool that uses approximately the same formulas. (Maybe could put in a note somewhere about it, since they maintain the tool you linked to?) Enterprisey (talk!) 23:16, 28 May 2017 (UTC)
 * ✅ https://tools.wmflabs.org/jackbot/snottywong/adminscore.html now redirects to https://tools.wmflabs.org/apersonbot/aadminscore/. JackPotte (talk) 23:34, 28 May 2017 (UTC)
 * There's something funny going on with that tool. I've been testing it a few times with my own username, in succession, and I get different results. Usually I score around 800, other times around 1000. How is that possible?  Y intan  07:15, 30 May 2017 (UTC)
 * Update: It has been giving me the same 800 result for the last few runs. Any idea where the 1000+ hiccups came from? Bug?  Y intan  07:28, 30 May 2017 (UTC)
 * Sometimes the tool queries databases that take a really long time to return results. If one of these queries comes back after the user has navigated to a different set of results (or reloaded, or something), the tool will still think that it has to add the results to the existing score. Definitely a bug, and I'll try to get around to fixing it soon. Enterprisey (talk!) 19:42, 30 May 2017 (UTC)
 * Just enabled my CSD log to do some more testing and found that the script calculates points based on the number of blue and red links in the log. Not sure if that's a good idea for 2 reasons:
 * 1. Articles can be properly tagged, they just haven't been deleted yet. Speedy isn't always that speedy.
 * 2. Articles that were properly tagged and deleted at the time can return when, for example, the subject gets notable.
 * In both cases the script will subtract points and calculate a lower-than-fair CSD score. In short, are you sure reading the CSD log is a good idea? Maybe just see if there is a recent CSD log, add some points for that (like for the User page), and leave it at that? Kind regards,  Y intan  09:34, 3 June 2017 (UTC)
 * , you make great points. For your first point, not considering the past two weeks will probably fix that - tracked here. For the second, I figured that I might as well penalize people who have logs filled with blue links and reward people with logs filled with red links, because on average that works. As a result, there are people whose tagging habits (say, lots of G12s, which probably can get recreated with non-copyvio text) will get unfairly judged by the tool. I can definitely reduce the weight of CSD logs - maybe make it out of 50 instead of out of 100, so it counts half as much. Thoughts? Enterprisey (talk!) 02:59, 4 June 2017 (UTC)
 * Hi . Going back two weeks (or even one week) would be a good way to fix my first point indeed. The blue/red thing is more complicated, I think. CSD logs are fast moving, and sometimes fast changing, beasts and I think any formula based on blue/red links will at some point be unfair to candidates. If you want to calculate a candidate's clue about deletion, I would use the AfD stats instead. There a justified 'delete' still shows valid in the stats, even if the deleted article returns at a later date. For CSD logs I would suggest to just give the candidate a few points for keeping a log. Mind you, I'm only saying this because I can't think of a fail-safe way to calculate CSD stats, I'm not objecting on principle.  Y intan  08:18, 4 June 2017 (UTC)


 * Who says not many people care too strongly about AfD? I've seen it be a deal-breaker at RfA, but less so these days. But maybe you're right and we should drop it. Anna Frodesiak (talk) 22:42, 28 May 2017 (UTC)
 * It's definitely possible to add extra code to the AfD stats tool that returns machine-readable results, so we can start querying it from other tools. I could even make user-selectable criteria for percentages, such as counting no-consensus results as keeps or something. If I do that, it would also probably be a good idea to do the same for xfd-stats, which is AfD stats for TfD, CfD, MfD, etc.` Enterprisey (talk!) 23:19, 28 May 2017 (UTC)


 * I recall using the holding pen before, but the design of the pages seems to have changed. If I am vetting a candidate who does not pass one of the vetting stages (e.g. doesn't have enough content creation), do we now remove the name or move it to the holding pen? Mz7 (talk) 17:58, 29 May 2017 (UTC)
 * Hi Mz7 . We move all entries to the holding pen one way or the other. Anna Frodesiak (talk) 04:59, 2 June 2017 (UTC)

It looks like it's coming along. Great! As soon as you have something, and it doesn't have to be perfect, just something we can pick through, please paste it at Requests for adminship/Optional RfA candidate poll/Poll candidate search/Poll candidates and we can get started. Sorry I cannot be more useful in this list generation area, but I'll try to make up for it in the vetting area. Cheers, Anna Frodesiak (talk) 05:02, 2 June 2017 (UTC)

This is it
I'm so, so, so sorry for the mass pinging. Please, please don't kill me or trout me or recall me or do horrible things to my userpage.

This is it. If this doesn't end up in a raw list, I'll give up and ping you no more.

Just post whatever the best raw list you can come up with as soon as you can to Requests for adminship/Optional RfA candidate poll/Poll candidate search/Poll candidates.

Again, I'm so sorry to bother you all with this. I'm just trying to move things forward.

Best wishes, Anna Frodesiak (talk) 05:13, 2 June 2017 (UTC)


 * I didn't receive the ping. Try all editors with rollback.  J <i style="color:#137412;">947</i>(c) (m) 06:06, 2 June 2017 (UTC)
 * With rollback? Anna Frodesiak (talk) 06:10, 2 June 2017 (UTC)
 * I suspect you hit the maximum number of editors in a ping (I don't know the exact number, but 26 sounds like it's over the limit) and thus none went through. Enterprisey (talk!) 18:24, 2 June 2017 (UTC)
 * It's either 24 or 25.  J <i style="color:#137412;">947</i>(c) (m) 18:26, 2 June 2017 (UTC)
 * Hmmmm, I did it in four separate saves, though. Anna Frodesiak (talk) 23:30, 2 June 2017 (UTC)
 * Only signed posts will trigger notifications (so if you copy edit a post, a notification will not be generated). Perhaps the editors listed in your last post that you signed were notified? isaacl (talk) 23:37, 2 June 2017 (UTC)


 * Ah, I see. Let's try this:

Anna Frodesiak (talk) 23:44, 2 June 2017 (UTC) Anna Frodesiak (talk) 23:45, 2 June 2017 (UTC) Anna Frodesiak (talk) 23:45, 2 June 2017 (UTC) Anna Frodesiak (talk) 23:45, 2 June 2017 (UTC)

Hi pinged people: Please read above "I'm so, so, so sorry for the mass pinging..."

Anna Frodesiak (talk) 23:46, 2 June 2017 (UTC)
 * Anna, I got your ping and assume that you're really hard up for fresh candidates so I'm trying something new. I'm searching the last N edits to the Conflict of Interest Noticeboard, which I'm familiar with, for reasonable candidates and will plop them on the search page. It's going to be pretty raw. It's just an idea for a way to come up with active, involved people. If it works maybe we can try it with other noticeboards. - Bri (talk) 00:00, 3 June 2017 (UTC)
 * That's fantastic, Bri. Thank you! Sure, let's experiment and see how it goes. Best, Anna Frodesiak (talk) 00:02, 3 June 2017 (UTC)
 * In a quick pass I see several editors who are not yet sysops with scores from the tool above 500. Is that a reasonable lower bound, or should I just dump everything with a positive score? - Bri (talk) 00:07, 3 June 2017 (UTC)
 * Don't think a negative score is actually possible for anyone remotely active. Even my two (legit) socks only score -38 and -69 and they have < 100 edits. So a reasonable threshold should be used, lest you add basically everyone with a few edits. As for what the correct threshold should be,, who recently passed RfA currently scores 498 (if you deduct the +500 sysop bonus), so anyone scoring at least 300 should be taken into consideration. Speaking of which, , how about yourself? Regards  So  Why  18:38, 3 June 2017 (UTC)
 * Don't you think that's a bit low? I expect a 300 treshold will give us "a cast of thousands".  Y intan  19:44, 3 June 2017 (UTC)
 * (Sorry, got distracted from the serious answer, which was, better late than never) - Yes, GR's is exceptionally low- and exceptional! I wouldn't take GR to be a control for anything, really (in a nice way)- his whole RfA was unusual, if not abnormal. Yes, there might be others like him- but there will be literally thousands to work through before finding them. &mdash;  O Fortuna   semper crescis, aut decrescis  17:41, 14 June 2017 (UTC)
 * Great, we can re-enact Ben Hur ;) &mdash;  O Fortuna   semper crescis, aut decrescis  19:53, 3 June 2017 (UTC)
 * Always nice to see people know their classics ;-)  Y intan  19:56, 3 June 2017 (UTC)
 * Well, I said considered, not added to the list. The lower the number, the more thought should go into whether to add them of course. Regards  So Why  20:04, 3 June 2017 (UTC)


 * AIV, UAA and SPI should work as well --Kostas20142 (talk) 14:40, 3 June 2017 (UTC)
 * Regulars at RfP too. I'll take a look. Lectonar (talk) 14:12, 8 June 2017 (UTC)

I'm watching, waiting for the dust to settle. Then I will get going on the vetting. Thanks all for keeping this up. Anna Frodesiak (talk) 16:53, 8 June 2017 (UTC)


 * did a lot of this years ago. I lost interest on most things admin related because it always ends up in circular arguments.  Dennis Brown - 2&cent; 13:46, 9 June 2017 (UTC)