Wikipedia talk:Requests for adminship/Optional RfA candidate poll/Archive 5

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

←

Archive 3

Courtesy archiving of closed poll seen by candidate

Regarding this edit: see the discussion above and this previous discussion on archiving a poll that was removed by the candidate, as a courtesy. It's not a big deal to leave it for the bot to archive, but personally the main reason I see for leaving a poll up is to let the candidate see the results. Since removing a closed poll implies the results have been seen, I see no issue in archiving the poll immediately. isaacl (talk) 03:41, 16 October 2017 (UTC)

It's also a courtesy to let the people who frequent the page see what has happened. If the poll is put up, closed, and archived in a few hours, and that's the standard methodology, then a series of these could be completely missed. It does help to know if we're getting a fair number of snow closes, and keeping them up for at least a short while could also warn off other new editors who haven't otherwise been paying attention. BlueMoonset (talk) 04:20, 16 October 2017 (UTC)

Personally I don't think having a poll up for a week is a big deal. Referring back to Wikipedia talk:Requests for adminship/Optional RfA candidate poll/Archive 4#Archive length, however, there was a concern raised about leaving negative criticism, particularly if non-constructive, up for some time. It does feel somewhat ungracious to restore a poll to the main page when it was deleted by the candidate. I feel confident that there are enough observers of this page to notice any trends and raise them on the talk page. (As for warning off editors, if the big edit notice isn't doing it, I doubt any closed polls will.) isaacl (talk) 04:59, 16 October 2017 (UTC)

I'm not sure that we are worried about this particular editor, as he is a banned sock. And warnings like this potentially might give other newbies the clue that this isn't the place for them. I'm ok with leaving it up to be bot-archived. — Insertcleverphrasehere ^{(or here)} 05:17, 16 October 2017 (UTC)

Would've been nice if I were pinged!, Anyway my reason for reverting was as I said because a bot will do it anyway so I personally don't see the point in someone doing it sooner, Had I known they were blocked prior to the revert then I probably wouldn't have reverted, Anyway we have bots that archive so human-archiving isn't necessary nor needed. –Davey2010^Talk 12:16, 16 October 2017 (UTC)

Discussion about the optional RfA candidate poll

Regarding these edits: I suggest that editors not discuss the concept of the poll within a specific poll, as it is unrelated to providing feedback to the candidate. It would be better to hold the discussion on this talk page. Thanks! isaacl (talk) 16:27, 1 November 2017 (UTC)

Comments with scores

For the past few threads here, I have declined to give a score and just given a bunch of opinions that explain what needs to be done to become a suitable RfA candidate, in my view. I think it's worked well, and other people have followed suit. Does anyone think it's a good idea to do away with the idea of marking people out of ten altogether? Ritchie333 ^(talk) ^(cont) 15:46, 30 November 2017 (UTC)

I'd support that. I've never given a score either, as I find it more useful to provide a detailed opinion. There's been a good amount of misunderstanding about what the number should be: does 10/10 mean "in my opinion you have a 10/10 chance of receiving >65% support (or 75%?)" or does it mean "in my opinion you will receive 100% support?" I've always thought it was the former, but that hasn't been too clear. When I took the poll nearly 18 months ago, it was the comments I received that helped me, more than the ratings. Vanamonde (talk) 15:58, 30 November 2017 (UTC)

As stated in the instructions, the score is the likelihood of having a request for administrative privileges be accepted. So it the probability of the bureaucrats determining that your request has succeeded. isaacl (talk) 17:20, 30 November 2017 (UTC)

I too would support - I've always based my numbers on "I think you have for instance a 5 out of 10 chance of succeeding ... but that's just a wild guess I don't really know what the numbers are supposed to mean as there's been so much confusement over them, Anyway I do agree with doing away with the numbers. –Davey2010^Talk 16:17, 30 November 2017 (UTC)

Yes can do; numbers can be ditched. While I've given scores, I think it's honestly kinder to leave sensible remarks than to give a rating that someone's 0/10 or 1/10 and leave them much the worse. We might have to rename ORCP to Optional Rfa Candidate Assessment in that case. Lourdes 16:46, 30 November 2017 (UTC)

In the more distant past, when this has come up, there have been numerous persons who expressed a desire to stick with a simple number to make the process more lightweight. (Anna for example holds this view.) However, last time this was discussed I asked for others with this viewpoint to speak up, as well as those who have gone through the process to talk about what format they would find useful. Since there were no responses beyond Anna's, perhaps the time is ripe to change the instructions accordingly. I noted before I was going to try to alter them to make providing a score optional, but was having difficulty writing a concise set of instructions to guide the non-numerical feedback. In the interest of a lightweight poll, the feedback shouldn't be too comprehensive. Additionally, concerns have been raised that the feedback is not sufficiently constructive. However the instructions can't be too lengthy, either, because people aren't even reading the current, 3 4-sentence long instructions. I'll give it another attempt.

I don't think any renaming of the poll is necessary; it's still a poll of a few persons regarding one's chances to succeed at becoming an administrator, with all the usual caveats of a small sample size survey. isaacl (talk) 17:20, 30 November 2017 (UTC)

How about something like this:

Responders, please provide short, constructive comments on the potential candidate's likelihood of passing an RfA, based on how you think the community will evaluate the candidate. (Note your personal evaluation of the candidate might be different.) Please be understanding of those who volunteer without fully understanding what is expected of an administrator, and phrase your comments in an encouraging manner as much as possible.

The sentence regarding the one-click rating would be removed. isaacl (talk) 17:27, 30 November 2017 (UTC)

Isaacl - I certainly prefer that statement - It would (or should) deal with some of the toxic comments (mine included), Short and to the point. –Davey2010^Talk 18:25, 30 November 2017 (UTC)

Well, it's pretty much the same as the current instructions, just with the part about the score removed. So unfortunately I don't think it will make too much of a difference. isaacl (talk) 18:31, 30 November 2017 (UTC)

If you do away with scores, how will you be able to evaluate the usefulness and accuracy of the ORCP, as recorded here. Cwmhiraeth (talk) 20:31, 30 November 2017 (UTC)

One can stop adding average scores, or can only provide a summary of the discussion ("Likely", "Unlikely", "NOTNOW", and "DEFINITELYNOW"). --Izno (talk) 20:56, 30 November 2017 (UTC)

From a "is this poll helping" perspective, it's enough for whoever updates the results page to make a rough estimate of the net consensus of the discussion, something like Izno suggests, and we can check if a related RfA follows the same trends as a given poll. Alternatively, there aren't so many polls and RfAs, anyway; someone can read through them and determine if the poll was helpful in identifying strengths and weaknesses in the candidate. (As previously discussed, several variables make any statistical analysis invalid. Even if the same people participated from poll to poll, and they were all self-consistent in their probability estimates, there's no good way to compare the probability to an RfA vote total. And since all candidates are different, it's also pretty hard to look at the aggregate RfA results to figure out if the probabilities are accurate across all of those who initiated a poll.) isaacl (talk) 21:14, 30 November 2017 (UTC)

I think we should keep the score system, leaving it as an option for respondents to choose to use it or not. It provides a rapid overview without having to read all the comments which, despite our appeals, are often far too long - let's face it, a candidate who has read the instructions at the top of the page should have read the linked advice pages, and their self-evaluation will provide them already with most of what they need to know about their chances at RfA. The poll is just a bonus. Being able to read instructions and take advice are prerequisites for admins, anyone who does not or cannot has already demonstrated their unlikelihood of being considered fro adminship. The fact that many of the respondents don't read the instructions is, well, frustrating. Kudpung กุดผึ้ง (talk) 20:16, 1 December 2017 (UTC)

The people who don't read through the comments are the same people who leave frivolous oppose votes at RfAs. Ritchie333 ^(talk) ^(cont) 10:06, 2 December 2017 (UTC)

Commenters will still be free to list a score (probability of passing) if they want. The instructions don't rule that option out, and even they did, of course people will just ignore them as they wish. isaacl (talk) 18:34, 2 December 2017 (UTC)

Lack of interest

It may be just my impression, but there seems to be a decrease in the number of candidates using ORCP. I am wondering if this is due to a lack of interest in using it, or (more likely) a continued lack of interest in candidates of the right calibre wanting to go through 7 days of RfA, which is still plagued by poor faith oppose votes and drama. Kudpung กุดผึ้ง (talk) 01:40, 24 December 2017 (UTC)

The latter is certainly an issue EvergreenFir (talk) 02:48, 24 December 2017 (UTC)
Without commenting anything about ORCP, the latter seems to be quite an issue.While reasonable dissent must be promoted, some blatantly disruptive schadenfreudist(s) are surely surplus to our requirements.Winged Blades^Godric 04:32, 24 December 2017 (UTC)
Pinging TonyBallioni who may have some interesting takes:)Winged Blades^Godric 04:32, 24 December 2017 (UTC)

The good candidates definitely dont use ORCP before running for actual RfA. I mean, they go straight for it. Then there are candidates who use ORCP, but most of them dont have any chance for a successful RfA. There are a lot of examples of such users. And then there are editors like Tony, megalibrarygirl, and most used example: goldenring. Tony, and MLG didnt go through ORCP, and goldenring would been discouraged here. I dont think we have a problem at RfA. Most of the recent RfA's have gone pretty smoothly i think. There were one or two odd oppose votes at Tony, and MLG's RfA, but apart from that the RfAs were pretty good.
I think it is either the good editors dont want to be an admin for some reason, or maybe they havent been asked by someone to be nominated. I think if someone encouraged a good editor for something, then they would go for it. This has been observed recently with Insertcleverphrasehere's activity with inviting editors to join NPR. I mean, this dude literally backlogged NPR's PERM division single-handedly.
During Ansh666's RfA, I initially voted oppose, but later Mz7, and Ritchie333 very politely asked me to reconsider, and i did it. I was going to suport him, but by the time I got free, his RfA was already closed as successful. Cullen's RfA was very good as well.
I dont think the problem is with the RfA but it seems like, all the voters start biting a candidate if he is not good enough. Something like, "ganging up". Almost all of the successful RfAs of 2017 were sort of smooth, with very little less drama; that too only from one or two editors. Even the currently ongoing RfA of Muboshgu is going pretty smoothly, except for one WP:GRENADE. But it is like only less than 1% of voters (on an average two voters from 200) that are causing drama because of their interaction with the candidate in the past. —usernamekiran(talk) 13:27, 24 December 2017 (UTC)

Well, out of the last three successful, joe roe did use ORCP. Galobtter (pingó mió) 13:34, 24 December 2017 (UTC)

Actually, yes. Even though Powerenwiki's RfAs was unsuccessful, and Riley's ORCP was unsuccessful; i consider them as good editors. And they used ORCP. But there is exception for almost everything.

—usernamekiran(talk) 13:56, 24 December 2017 (UTC)

I'll bite. — Insertcleverphrasehere ^{(or here)} 21:30, 26 December 2017 (UTC)

I feel (and I think others concur) that ORCP tends to have many of the cons (airing of grievances and perceived flaws + setting up future questions about "what have you done to improve since then?") of an actual RFA without the pros (possible promotion). This is not to say that I'm against candidates learning from folks' responses and becoming stronger candidates. But I do feel that the weirder kinds of oppose rationales (less than 31415 edits, less than 31.41 years of tenure, more than 3.14% incorrect CSD tags) get amplified through ORCP. Airbornemihir (talk) 22:06, 10 January 2018 (UTC)

side discussion

@Ritchie333:, regarding this edit: could you discuss this somewhere such as Hawkeye7's talk page, in the interest of keeping the poll focused on feedback for the candidate? isaacl (talk) 22:44, 14 February 2018 (UTC)

As a response to "SPI work meh" it was fine. >SerialNumber54129^{...speculates} 09:58, 15 February 2018 (UTC)

The conversation fortunately avoided the personalization that had been introduced. Nonetheless, as has been discussed previously on this talk page, prolonged discussion on the underlying philosophies of those evaluating administrator candidates is often overkill for the actual poll itself. No one's going to remove comments from this poll (unless they are flagrantly disruptive), so all we have is gentle reminders to try to keep the comments focused on the candidates. General discussion that isn't specific to one candidate can be held in numerous other places (including this talk page if desired), so it doesn't have to get repeated across multiple polls. isaacl (talk) 17:31, 15 February 2018 (UTC)

Shortcuts

Edit warring over shortcuts? Holy overkill, Batman. ←Baseball Bugs ^{What's up, Doc?} carrots→ 18:13, 23 February 2018 (UTC)

WP:SHORTCUTS3RREXEMPT, of course :D >SerialNumber54129^{...speculates} 18:24, 23 February 2018 (UTC)

So your revert absolves you clearly, just because you were the first one? :) I provided the guideline on the matter in the final edit, so whoever wants to deviate from there should show some talk page consensus (which has not so far been present, so far as I am aware). --Izno (talk) 19:06, 23 February 2018 (UTC)

I was surprised that this triviality was an issue. As for it being a lame edit war, it doesn't count as such until it's been taken to ANI and argued for a number of days or weeks. ←Baseball Bugs ^{What's up, Doc?} carrots→ 19:27, 23 February 2018 (UTC)

I see that Wikipedia:Administrators' noticeboard/Incidents has FOUR shortcuts listed. Maybe you shortcut control freaks should take your edit war to ANI and see how things go for you. ←Baseball Bugs ^{What's up, Doc?} carrots→ 03:07, 24 February 2018 (UTC)

I'll settle this. You all stand down. I'm going to remove:

Then we'll have no shortcuts for a year or so, then I will suddenly add:

Settled! And that's final!

Anna Frodesiak (talk) 06:12, 24 February 2018 (UTC)

I really didn't want to talk about this, but if we're going to, let's talk about the merits and not about each other. There are a couple of reasons why people create shortcuts: to save typing when linking to a page, and to have a mnemonic to (hopefully) more easily remember a page. Unfortunately, this often means that the shortcut gets used in the visible text, creating a jargon that makes the text less accessible to those unfamiliar with the page in question. (It doesn't have to be this way, but that's what has happened in practice on English Wikipedia.) As the number of shortcuts increases, editors have to devote more attention to keeping the shortcuts straight. It's a tragedy of the commons situation: adding one more shortcut might make your editing process easier, but at a cost you've imposed on everyone else. The advantage of jargon is having a unique term for a concept, not having multiple ones.

Now it's unlikely that much opposition will arise to stop someone from creating a new shortcut, because the cost of the resulting conversation outweighs the benefit, as long as the new shortcut is just getting used once or twice. But to limit the tax of a new shortcut, in my opinion it is reasonable to discourage the proliferation of multiple shortcut links to the same target, and to consolidate on as small a number of shortcuts as possible. To that end, I feel it is reasonable to limit the number of shortcuts displayed next to the target location. isaacl (talk) 06:37, 24 February 2018 (UTC)

Agree, if we start adding all possible extracts from the page title as shortcuts (WP:RFACP, WP:RFAP, WP:ORFA), or shortcuts that are mistypes of other shortcuts (WP:OCRP), there will be no end to it: Noyster (talk), 10:19, 24 February 2018 (UTC)

Scope

There was probably a discussion I've forgotten about, but why isn't the poll broad enough to allow a bit of editor review? Anna Frodesiak (talk) 18:14, 29 March 2018 (UTC)

Recall your original concept was to have a simple score, and very minimal written feedback. That format doesn't lend itself to providing general editor reviews. Wikipedia talk:Requests for adminship/Optional RfA candidate poll/Archive 1 has various threads discussing scope, and how the participants then wanted to keep the comments lightweight. Personally, I think it would be better not to expand the scope of this poll, and instead revive Wikipedia:Editor review. If that works out, then perhaps this poll can be folded into editor review. isaacl (talk) 01:43, 30 March 2018 (UTC)

Hello Isaacl!

We wanted it simple at first because we didn't know what to expect. Then a lot of paragraphs landed.

Why try reviving editor review and folding this into that when there is a simpler option?

I see this like a waning company offering a single product. Why not test market a new product? We could broadening the scope just a little with a tweak to the instructions.

We could let people take the poll who are considering "RfA." rather than considering RfA "...in the near future". We could allow "a short review" rather than it be "not intended to provide general reviews of editors". This would allow a bit if guidance for all to see and benefit from here rather than at the user talk.

The potential benefits:

Editor review comes alive in this poll.
We get a bit of a live guidance page.

What's the worst that could happen that would out outweight those (and other) benefits? Plus, if after a month it doesn't work out, we revert to where it is now.

I say, test that new product. Keep up with the times. Be pragmatic and dynamic. Companies that do not do that, find themselves missing opportunities or even "inactive and retained for historical reference"!

Is that a pitch or what!!

Anna Frodesiak (talk) 02:40, 30 March 2018 (UTC)

Well, that's the point: companies who succeed at selling bowling balls don't say, here, buy our bowling ball and use it to play golf. They actually release a new product. If people are interested in performing editor reviews, I think it's better to do it under a page called that, and not say you can get your editor reviews done as part of the RfA candidate poll. isaacl (talk) 03:14, 30 March 2018 (UTC)

As for the "near future" criterion: those reviewing have chosen to invest the time to estimate the candidate's chances based on their current characteristics, and not what they estimate their characteristics will be down the road. Can't we continue to have a process for those who prefer this choice, and a separate process for those who want to be more forward looking? (A score representing a probability would likely not be the best option for this scenario.) isaacl (talk) 03:21, 30 March 2018 (UTC)

Hmmmmm, good points. Let me think about that. Anna Frodesiak (talk) 03:43, 30 March 2018 (UTC)

Looks to me as if it's either a popularity issue with ORCP or the well known ever increasing disinterest in becoming an admin. The use of ORCP since it began appears to be (and these figures are very rough):

1st 12 months: 117
2nd 12 months: 26
3rd 12 months: 64
Q1 2018: 7

The editor Review project died a natural death and was deprecated. Reviving it, or anything similar, under a new mame would, IMO, not only be a waste of time but also defeating the object of ORCP. It's never been proven that ORCP has actually increased the number of potential candidates or the actual number of RfAs which still suffer from downright mean spirited voting. Nobody is keen to go through that kind of thing but the community appears to be reluctant to do anything about the serial RfA killjoys. Kudpung กุดผึ้ง (talk) 16:27, 30 March 2018 (UTC)

Fair enough. And I'm not seeing evidence that this poll page is useful either. Anna Frodesiak (talk) 19:05, 30 March 2018 (UTC)

As someone who recently used it, I found it useful. Cordless Larry (talk) 19:29, 30 March 2018 (UTC)

That's very good to hear, Larry. And congrats on your RfA! :) Anna Frodesiak (talk) 21:01, 30 March 2018 (UTC)

Thanks, Anna Frodesiak. I've passed. Phew. Thanks to you and Samwalton9 for first approaching me with the idea to run. It took me a while, but I'm glad I did. Cordless Larry (talk) 06:32, 2 April 2018 (UTC)

I think an editor review process can co-exist with this RfA candidate poll since this poll is more suitable for its specific purpose, and will likely always have a faster response time than a more comprehensive editor review. I doubt that over the past four years, a lot of new people willing to do editor reviews have emerged, but as long as expectations are managed, I don't think it's a waste of time for those who are interested to look for opportunities to perform reviews. — Preceding unsigned comment added by Isaacl (talk • contribs)

I think ORCP has run its course - it was a nice idea, but there are only going to be so many people interested in RfA, and they've probably already been through it all. Informal conversations are okay, and another candidate search like Anna did last year (maybe just on a userspace page this time) might be a good idea to gather potential candidates. Really, the best way to move forward is to get more people nominating candidates and encouraging people to go forward with it. Ritchie333 ^(talk) ^(cont) 12:53, 31 March 2018 (UTC)

I don't think there is a need to disband ORCP at all. I believe the lack of requests is more to do with a general lack of interest in running for adminship which has been a problem for years, and we know what the reasons are and have known since I launched WP:RFA2011. The community seems reluctant to address these issues. Instead, we had a bunch of reforms made by a new admin, who came and went in a short space of time, which have not altered the situation one iota. We need to start earnestly addressing those issues. Kudpung กุดผึ้ง (talk) 15:39, 2 April 2018 (UTC)

Well, I've managed to put forward 17 candidates so far, of which 14 passed. I'm trying to encourage one guy at the moment who would make a great admin, and have had to give some advice to another who I would possibly nominate except he's just had a run of declined A7 speedy tags, which kills an RfA stone dead these days. I'm writing some scripts to try and get another corpus of potential admins.

Having talked to candidates afterwards, most think RfA wasn't actually that big a problem and it wasn't as bad as they feared. A key issue seems to be that there are too many questions and too much cross-examination. I do have some idea of what questions might be, and am always around to give advice - although there tends to be a "what you should say to pass the RfA" answer and "what I would actually do" answer.

Looking through the last couple of RfAs, which spans back a year, the common theme of opposition seems to be "too many bad calls at CSD / AfD". I don't think Andrew Davidson is a problem anymore other than just causing mild disruption - on the most recent RfA he was openly mocked and taunted and the community now effectively has him on "mute". A further problem is once you have a sizeable number of support / oppose !votes (which generally happens after 24 hours), the RfA usually becomes a self-fulfilling prophecy. When people look at an RfA that's just open, they have an open mind. If they look at it and find 50 other supports, the "oppose" suddenly becomes "well I'm not sure but hey, 50 people can't all be wrong?" which affects the result. Ritchie333 ^(talk) ^(cont) 17:57, 2 April 2018 (UTC)

It's true that these days most serious RfA are almost certain to pass despite the persistent obviously deliberate disruption by Davidson. RfA failures by established editors are becoming more rare. RfA may therefore be somewhat less of a village pillory than it used to be but it still needs to live down that reputation. However, my recent involvement in The Signpost seems to demonstrate a growing, general apathy in most meta areas. Biblioworm's reforms a couple of years ago did nothing to address the situation (if indeed that was his intention). What we need are more candidates, not more voters. Kudpung กุดผึ้ง (talk) 02:56, 3 April 2018 (UTC)

On that note, props are due to SoWhy for conscientiously raising their concerns even in the face of (apparent) community consensus at RfA - and I'm not talking about just one instance, but a history of thoughtful participation. Airbornemihir (talk) 04:20, 3 April 2018 (UTC)

I think Biblioworm's reforms helped. Some of the changes were subtle improvements that nonetheless make RFA a better environment. The per-user question limit has helped eliminate some of the irrelevant questions that were sometimes asked. I think candidates sometimes felt compelled to answer all questions, no matter how silly or trollish, for fear of a "you didn't answer question #17" backlash. Now that RFAs regularly get more than 100 votes, the influence of individual editors is more limited. RFAs now typically fail for reasons other than politics. People also feel emboldened to move debates to the talk page, where the bickering hopefully won't cause massive amounts of drama. If RFA is no longer seen as a toxic free-for-all where your enemies are allowed to take potshots at you, that may help convince some candidates to run. NinjaRobotPirate (talk) 06:14, 3 April 2018 (UTC)

True, I grant you that about the questions, but we get a lot more drive-by votes even if they are mostly in support. The long arguments that get moved to the talk page however, are a relatively new phenomenon. Previously, the only thing that was generally posted to talk were the candidate's editing stats. Oh well, trends come and go. Kudpung กุดผึ้ง (talk) 06:46, 3 April 2018 (UTC)

I've dug through a bunch of user stats and identified Feminist and Whpq as editors who could make use of the tools and might be interested; however neither were (they didn't say why). I've also asked Crow and PinkAmpersand and got lukewarm responses. I'm beginning to come to the conclusion that adminship is simply dull and unexciting to most people, and the talk about "serial opposers" and "RfA reform" is actually a giant red herring. Ritchie333 ^(talk) ^(cont) 11:49, 4 April 2018 (UTC)

I can't speak for anyone else, but I don't think admin work is boring. I was a Wikidata sysop and quite enjoyed that. I just don't think that my current activity levels are high enough; that's the only reason that I haven't pursued adminship since you suggested it. And, to be clear, I really am honored that you think I'm qualified. Personally I'd thought I was several years away. So yes, I do plan on pursuing this, once I'm a bit more active. Sorry for not communicating well on that front. — PinkAmpers &^{(Je vous invite à me parler)} 19:25, 4 April 2018 (UTC)

@PinkAmpersand:...was a sysop? Past participle; prepare to explain that one, here or at RfA (which is the equivalent of whatsisname's In This ife Or The Next!!!) :) —SerialNumber54129^{paranoia /cheap shit room} 19:58, 4 April 2018 (UTC)

Inactivity. ~ Amory (u • t • c) 01:02, 5 April 2018 (UTC)

If people aren't going to RfA just because adminship is dull and boring, why are there any admins here at all? Adminship hasn't become more or less interesting since 2001. That said, I think that you (and Kudpung above) are on the right track here, and it's got me thinking. I still think that high standards are an issue, but I also think that it's fair to observe that a) meta-level participation has declined to some degree in at least some areas and b) eventually you reach a point where you run out of people interested in being an admin.

I don't think that we've quite hit the point of nobody being interested in being an admin. I think that a big issue is that the standards for RfA have gone well beyond the intrinsic reward that people get or expect to get from being an admin. I've said before that passing an RfA requires a year or more of prior work, including frequent editing of a large number of areas of the project in order to get the activity and experience required at RfA. I think that a lot of people who might have run for RfA in 2007 (when the standards were roughly a) be around for three-six months and b) be active wherever you want but have experience in at least one area of admin work) have no interest in getting involved in all of the areas expected of admins and putting in all of the time required to make a credible effort at RfA today. People have other things going on in their lives, and the intrinsic value to them of doing the menial admin tasks isn't worth the long-term time costs associated with running a credible attempt at RfA.

As to decreasing meta-level participation, I'll try to find some metrics we can use to measure how this might be happening, and specifically where it is happening. That's an interesting point that probably merits some real consideration. -- Ajraddatz (talk) 00:49, 6 April 2018 (UTC)

The community has gotten a lot bigger, and the site's popularity has attracted many people who don't adhere to Wikipedia's basic principles. Because it often falls to administrators to deal with unco-operative editors, it's a lot easier to burn out now. We need to reduce the effectiveness of being unco-operative, and increase the incentive to work towards agreement, thereby reducing the need for administrative actions restricting editors. But the community still hasn't learned the lessons of Clay Shirky's talk, "A Group is its own Worst Enemy". English Wikipedia's traditional decision-making process gives an outsized amount of influence to the less-collegial, and thus there is no incentive for them to enable changes. isaacl (talk) 03:19, 6 April 2018 (UTC)

Thewinrat's ORCP

This user's ORCP was recently brought to an unceremonious end; while I don't disagree with the application of SNOW to an ORCP for a user under 600 edits, could the closing user sign their name on the closing notice, at the very least? The way it went, it is likely to leave people with the very strong impression that unaccountable "punitive" measures are par for the course on the English Wikipedia. Airbornemihir (talk) 04:20, 3 April 2018 (UTC)

The user in question had recently been warned about filing requests for rollback three times in quick succession, each one declined, and was advised to forget about hat collecting. Ritchie333 ^(talk) ^(cont) 15:44, 3 April 2018 (UTC)

@Ritchie333: Fair enough, but I was under the impression that discussion closures are generally signed (although CLOSE and SIGN don't seem to require this.) Airbornemihir (talk) 11:08, 4 April 2018 (UTC)

They are, when the closing editor puts his brain into gear and remembers..... :-/ Ritchie333 ^(talk) ^(cont) 11:12, 4 April 2018 (UTC)

Additional instructions

Regarding these edits: Given that the infobox at the top of the page already instructs prospective candidates to read the "Advice for RfA candidates" page, I don't think these additional instructions are needed. Anyone not following the current instructions isn't going to read the new changes. As it is generally pretty obvious when someone hasn't read any of the advice and instructions or is not paying any heed to them, everyone should just feel free to ignore these cases and not respond at all (optionally, someone can immediately close the poll). isaacl (talk) 02:32, 5 July 2018 (UTC)

When I saw the changes, I thought that "Our users do not like having their time wasted" sounded a bit passive-aggressive. Also, if someone clearly unsuitable for adminship opens a poll, does it really waste anyone's time? Responding to polls isn't compulsory, so not really. Cordless Larry (talk) 06:02, 5 July 2018 (UTC)

No objections if anyone wants to revert. That's what BRD is all about. Isaacl's suggestions are not without merit. Kudpung กุดผึ้ง (talk) 13:00, 5 July 2018 (UTC)

Agree with Isaacl here. A strongly worded statement like that might discourage the type of editor we actually want to come forward, IMHO. — sparklism ^hey! 13:14, 5 July 2018 (UTC)

OK, I will revert the edit. This will still leave two pointers to the "Advice for RfA candidates" page. isaacl (talk) 14:35, 5 July 2018 (UTC)

Admin score?

Has anyone figured out what a "good score" (vs. a "poor score") on Admin score is? It looks like it's been developed since I last looked at it, and might be useful in ORCP appraisals, if the metrics can be worked out... --IJBall (contribs • talk) 03:40, 17 July 2018 (UTC)

I think the consensus opinion is that a statistical tool will never give reliable-enough results to be of much use. I do find it amusing that the tool seems to recommend that more pages that I create should be deleted, though. power~enwiki (π, ν) 03:48, 17 July 2018 (UTC)

Obligatory plug for my alternate version with a pretty graph: toollabs:apersonbot/aadminscore.

But, of course, I agree with power - I don't think many of the qualities that make a good admin can be scored numerically. It's sometimes fun to look at, though. --Enterprisey (talk!) 04:36, 17 July 2018 (UTC)

Interestingly, it counts 'patrols' but not 'reviews', so anyone who uses the page curation tools as an NPR won't get any score from it. Even if it did, I don;t think it is of any use at all. — Insertcleverphrasehere ^{(or here)} 04:50, 17 July 2018 (UTC)

Enterprisey - what is supposed to be a "good" score on your system? Not surprised to find out it doesn't like my CSD micro-history - I've temporarily sworn off after the system didn't mark my copyvio CSD as red even when 97% of it was such as someone saw it and rewrote the article (in itself good)...but that's a rant for a different talk page. Nosebagbear (talk) 09:02, 17 July 2018 (UTC)

Easy:

Good score: 8
Bad score: 6
Medium score: 7

Anna Frodesiak (talk) 05:48, 17 July 2018 (UTC)

Well, there's an actual info page linked from that tool, so that might be instructive. As has been said in previous discussions and above, these can be helpful for a broad-picture look, but in the end they all devolve to "number of things done in *AREA*." Assuming you care about those areas and the number of things done there, fine, but in the end, RfA doesn't really hinge on that and nobody particularly cares. Put another, maybe an okay negative selector absent other information, but not a good positive selector. ~ Amory (u • t • c) 14:27, 17 July 2018 (UTC)

@Amorymeltzer: It's odd, but the info page seems to have lots of criteria that the tool itself doesn't—page patrols, AIV AfD, RFPP? Unless I'm missing something; but I'm definitely looking at Enterprisey's AAS (Asynchronous Admin Score!!!) —SerialNumber54129^{paranoia /cheap sh*t room} 16:41, 20 July 2018 (UTC)

Sorry for not being clear, when I said "that tool" in reply to you, I meant the one you linked, the wmflabs one. It does use those measures and those multipliers. ~ Amory (u • t • c) 17:44, 20 July 2018 (UTC)

Tt's nice to score over a thousand, and I managed that on both systems. However if I was an actual RFA candidate I wouldn't get a positive for a CSD log with three tags, even if all three were redlinks. I got through my second RFA because my nominator went through my CSD tagging and thought highly of it - the CSD log is a much more recent innovation that doesn't include any tags from before my RFA. If I was nominating a candidate with that small a CSD log I would advise them not to mention CSD tagging as an area they were active in, unless of course they had only just created a CSd log and I was advising people to judge their CSD tagging by their deleted contributions. The other system gave me 100 points for 100% use of edit summary, which is odd and overly generous, and not just because my oldest edits include some without edit summaries, more importantly at RFA it only matters that you have now started using edit summaries, no-one has had a problem at RFA because they only started using edit summaries a few months earlier. One system gives me 100 of my 1200 points for having a userpage of a certain size, the other gives me 10 points out of over 1300 for having a userpage. I'm not convinced that it merits near 1% of the score, I can't recall a serious candidate without a userpage, but in principle I don't think it would bother me. Ϣere SpielChequers 16:09, 20 July 2018 (UTC)

@WereSpielChequers: Isn't that because they both automatically give 500 points for having the adinship user right? —SerialNumber54129^{paranoia /cheap sh*t room} 16:41, 20 July 2018 (UTC)

One gave 500 for that, the other gave me 100 for dishing out a bunch of blocks, I couldn't have got those points without the bit, but the other 1100 I think I could have got. Not sure that the current admin thing is helpful though I can see that we might look at it for former admins. Ϣere SpielChequers 16:52, 20 July 2018 (UTC)

Some quirks, omissions and exceptions are likely in any such automated tool. I mention, only as an example or information, that I lost nearly half my points on the Scotty Wong tool because of a mistaken 49-second block some years ago. For all I know, mistaken (and old) blocks may still deduct points under new tools. I don't see blocks itemized as a deduction in the calculation of a current score, but they are one of the factors on the explanation page. Perhaps they are used for deductions in the calculation since one or more items must not be explicit. I don't know how frequently this quirk occurs or whether it really matters but I have seen other users receiving mistaken blocks.

I have kept a manual CSD log on my user page. I have used Huggle for almost all reports and did not notice how to create a Huggle log or combined log and was interested in just keeping a record of results for myself. I mention this only because I have seen instances where users (at RfA, at least) were not credited, or credited much, for CSD activity when they had no log. It seems to me that a log, or a complete one, cannot be created retroactively or at least not without some difficult manual work. I don't know how frequently this might occur or whether it is the only type of item where something like this may occur. Or maybe it is a negative not to set up a log? And maybe the importance of speedy deletion performance will decline now in any event.

Nonetheless, despite any quirks, I think the tools are valuable and can give some ballpark ideas. They also can provide a quick, anonymous tool for self-evaluation as long as a user keeps in mind that it is not a full or precise evaluation. For what it's worth. Donner60 (talk) 04:51, 22 July 2018 (UTC)

The Admin score tool was created many years ago by my very good friend Scottywong, now retired, but I never understood how it worked and I never use it preferring to apply my own judgment. That said, Scott has created many other excellent and extremely useful user analysis tools. Kudpung กุดผึ้ง (talk) 06:27, 22 July 2018 (UTC)
I am endeared to Scottywong as well, and miss the times of his oft presence. Happily, SW is actually semi-retired, and though it is much less often than before, he does edit from time to time. In many sad ways, SW was underappreciated around these parts. I collaborated with SW in his development of the Admin Score tool and can tell you, IIRC, its better potential was lost to strawman fallacies and good ole battle fatigue; often consuming and derailing efforts born of the best intentions. It's a fairly accurate tool, nevertheless, and was, in the past at least, quite useful as well.--John Cline (talk) 15:18, 22 July 2018 (UTC)
Appreciate the kind words, gents. If I had more free time, I'd come back and brush the dust off the tools I made many years ago, many of which are likely broken or outdated now. Maybe some day. ‑Scottywong | talk _ 15:38, 1 August 2018 (UTC)

Oh, that score. Sorry. I didn't click the link. Anna Frodesiak (talk) 18:13, 22 July 2018 (UTC)

I should make clear that I am in no way criticizing Scotty Wong or his effort or the tool he devised. I have been around long enough to have seen and admired his work and his demeanor. It was a loss to the project that he retired or semi-retired. I agree with Kudpung กุดผึ้ง that the tool could not be used to make a final judgment on whether to support a candidate. As I began to !vote/comment in RfAs, I gave the score little if any weight in my final determination on whether to support an RfA, at least when many other facts came to light. However, I also agree with John Cline that the score very often pointed in the right direction. This makes the point that the quirks or omissions could result in higher as well as lower scores, and higher or lower evaluations. Again, the score, especially by itself, would not have been determined whether an RfA should be supported.

Given the wide variety of possible facts or actions to consider in devising such a tool and a few limitations on bot gathered data, it is not surprising that some apparent improvements have been made in the more recent tool and some quirks or exceptions, or important plus or minus factors, could not be considered then or now. The current tool is useful in evaluating candidates even if it cannot be used to make the final determination for an !vote. Of course, if it could determine the outcome, why even have RfA? I am not trying to set up strawmen or criticize the tools, just to mention some limitations that I have seen. Perhaps some of these are not limitations in the tool but in how at least a few !voters use them without fully understanding them or taking other factors into account.

I do conclude that the tool is a good one and the score, within a range and with other points in mind, can help with evaluation and even point toward a proper !vote, as I wrote before. I am not sure that my points about some additional considerations for evaluation and !voting were made clearly enough and that this comment might be useful.

The participants in this talk page know these things but not everyone who comments in RfA does. A few points that may not be directly related to the scores, such as why created articles might have been deleted, e.g. user reconsiders whether to proceed, at least at the time, and perhaps thinks a mistake of some sort can be corrected only by starting over and asks for deletion. I would not consider that type of deletion to be any sort of serious negative. This factor, which has been raised a few times, may not be closely considered by some at RfA when they comment on it.

Again, just my observations or opinions for what they are worth. (Angels on pinheads?) Unless I see a further comment that seems to suggest that a reply would be useful, I'll close with that. Donner60 (talk) 23:15, 22 July 2018 (UTC)

The main problem with RfA is that just anyone without any limitations whatsoever, is allowed to vote. A paradox created with the Dec 2015 reform was that of publishing RfAs on the watchlist. It doubled the participatipon, but simply doubled the intrinsic problems. With RfA being, like all meta areas, a magnet to newbies and inexperienced users, a significant number of voters just do not understand what adminship is all about, whichever side they vote on. They won't read WP:RFAV until you tell them to, and by then it's too late. Kudpung กุดผึ้ง (talk) 00:40, 7 August 2018 (UTC)

Very late to the party, but I wanted to say that while the XTools Admin Score contains some useful information, the score itself should be taken with a grain of salt. Enterprisey's tool may be superior, I am not sure. The XTools variant has not received much attention from developers, and remains largely unchanged from the logic created by Scottywong years ago — MusikAnimal ^talk 00:15, 17 October 2018 (UTC)

I prefer to use my judgement when !voting and to this day have never bothered to look at admin-score type tools, A tool in my eyes doesn't really define whether you're a good editor/soon to be admin or not and so like User:MusikAnimal says these should be taken with a grain of salt. –Davey2010^Talk 00:28, 17 October 2018 (UTC)