Wikipedia talk:Requests for adminship/Archive 78

From Wikipedia, the free encyclopedia

Contents

[edit] Anons asking questions

Since only registered users are allowed to vote, should questions from not-logged-in users be removed. —Malber (talk • contribs) 14:02, 8 January 2007 (UTC)

That's what I'd do, unless it's a legitimately good question (which if it is, some other user can re-ask it anyway). --Wizardman 14:03, 8 January 2007 (UTC)
When I saw that happen, I wondered the same thing myself. WP:RFA seems to give leeway, saying that anons can comment in the comments and questions sections. – Chacor 14:16, 8 January 2007 (UTC)

Anons can place comments and ask questions. Kim Bruning 16:20, 8 January 2007 (UTC)

Kim's right, of course anon users ask questions and make comments. Do not immediately assume that just because somebody chooses not to create an account, or chooses not to log in, that they do not count and should be ignored. We do not allow them to vote because of obvious reasons (possibility of sockpuppetry, gaming etc), but we take their contribution on its own strength. If I saw an RFA candidate refuse to answer, or removed, an IP's question, I wouldn't hesitate to oppose them, as they have failed to understand what Wikipedia is. Proto::► 16:25, 8 January 2007 (UTC)
I agree with Proto who agrees with Kim. -- nae'blis 20:56, 8 January 2007 (UTC)

Does this perhaps open the door for abuse by blocked or banned users? —Malber (talk • contribs) 20:55, 8 January 2007 (UTC)

In what way? No one is proposing to allow them carte blanche, that I can tell.. -- nae'blis 20:56, 8 January 2007 (UTC)
Well, as has been pointed out to me earlier, questions can be used to press a particular Wikipolitical point. There's one question in a current RfA from a registered user about the controversial issue of admin decisions being discussed and made with off-wiki IRC discussion. The question in User: Pascal.Tesson's RfA from the anon is rather odd. For the record, I'm trying to craft my own questions to be more policy related and less pointy. —Malber (talk • contribs) 21:31, 8 January 2007 (UTC)
How does that question violate WP:POINT? Anyway, WP:BAN states that banned users can be automatically reverted if they're evading their ban, so no, this doesn't open up any doors for anyone. (In fact, this doesn't change anything at all. Anons were always allowed to participate productively.) --Rory096 00:13, 9 January 2007 (UTC)

This is nonsense. Anons have ALWAYS been free to participate in discussion. --Rory096 00:09, 9 January 2007 (UTC)

Precisely, and if the question/comment is valid and not disruptive, then it should remain. Thanks! Flcelloguy (A note?) 01:24, 9 January 2007 (UTC)
Please Rory, assume good faith. Just because I'm asking the question doesn't mean I'm suggesting that anons be immediately prevented from posing questions. I'm just wondering if there is a process to filter out banned users from abusing the process to prove a point. It's not like we're going to do a CheckUser on every anon that poses a question and it may not be as easy to identify banned users participating in an RfA as it would be a banned user editing from an anon IP on their pet article. Also, participating in an RfA is a bit different from editing an article. So even if the policy were to change to exclude anons from participating, it might not violate one of the Meta principles. —Malber (talk • contribs) 02:57, 9 January 2007 (UTC)
I am assuming good faith. I'm sure you didn't mean for it to be nonsense. ;) Anyway, you clearly did ask if we should remove questions from unregistered users. While it's true that banned users could use IPs, that's not a reason to ban all anons (many of whom are helpful and productive). Banned users can be automatically reverted under a separate, existing policy. To the contrary, it would definitely be violating at least the spirit of the Foundation issues, and even if it didn't, it violates the spirit of the wiki. We should not be excluding a wide group of people just to get rid of a couple bad ones that can already be reverted automatically anyway. --Rory096 01:59, 10 January 2007 (UTC)

As long as it's a good faith question, what possible reason is there for not allowing anons to ask questions? The only reason we don't let anons !vote is because there is no way to check no-one !votes multiple times. Someone asking multiple questions is not a problem, so who cares if we know who asked each one? --Tango 01:38, 9 January 2007 (UTC)

Agreed, so long as the question is appropriate and productive, I see no problems with an anon asking. A registered user might want to log out and ask a question, but the only reason I can see someone doing that is if there was a disruptive question in which case it would be ignored. James086Talk | Contribs 01:42, 9 January 2007 (UTC)
Even if the anon IP is that of a banned user, if the question is a good, pertinent one, what do we gain by removing it? Nothing. Judge each edit on its merits. Similarly, if it's clear trolling, or incendiary, or whatever, it would be removed regardless of where it came from Proto::► 11:12, 9 January 2007 (UTC)

[edit] Vote first, ask questions later

Since we're talking about asking questions, I've noticed some editors !voting first and asking optional questions later. In fact, the example that lead me to post here was a situation where it seemed that being the first to !vote and "beat the nom" and then posing an optional question was more important than asking a question and waiting for the answer before !voting. If people are going to ask a nominee a question, it really should be for the purpose of getting further information that will assist the questioner in making a decision. Otherwise, it smacks of making a point. If the question isn't important enough for a person to consider having answered before the questioner !votes, it probably shouldn't be asked at all. Agent 86 02:32, 9 January 2007 (UTC)

It's a consensus process. People stating their position for or against a candidate can expect to have that position challenged. Equally, if someone asks a question, then their position may change based on the answer to that question. The position they state initially may in fact be just that, an initial position.
If you yourself are not accustomed to this, just remember to keep watching an RFA you are participating in, because people may well ask you questions, or offer suggestions or compromises where it might turn out to be a good idea to change your own position. Even if all you do is reply, at times your continued presence alone might convince others of the merits of your position. Your participation is not over until the bureaucrat officially closes.
Have a nice consensus ;-) And have fun! Kim Bruning 02:49, 9 January 2007 (UTC)
I think you missed my point, and your reply puzzles me. It seems to be addressed to someone who is new to the process. I've been watching RfAs for almost a year now.
I'm not talking about people changing their mind after someone poses a question, or thinking about a question posed by someone else after they've !voted, or having second thoughts down the road. I'm talking about the practice of !voting (often in a mad rush to be first or "beat the nom") and then posting questions under "optional questions" that are slapped on most RfAs. If the person commenting has a reason for the support !vote, that person should say so, or at least offer some constructive reason. I chose not to call out the editors who I noticed doing this, because there's no need to single out anyone in particular.
The issue is that this is to be a discussion, not some sort of race to see who can !vote first. We ought to expect !voters to make a well-considered or thought-out decision. It does not seem to be so when some will race to !vote, and within 10 minutes post a seen-too-often "optional" question (to the point where they seem to be de facto standard questions). At the very least, it trivializes the process and at worst, it's insulting to other editors that the person posing the question can't be bothered to consider the answer in his or her !vote but expects other !voters to do so.
It's really not a "nice consensus" when there's little of the process that is supposed to lead to a consensus. Agent 86 08:55, 9 January 2007 (UTC)

Voting oppose or neutral before asking a question makes sense. You are stating how you feel at the moment, and giving the candidate a chance to change your mind. Voting support and then asking a question doesn't make sense - the only thing you can do in response to their answer is stop supporting, which means you're asking the question specifically to trip them up despite thinking they're ok to become an admin. It's blatant WP:POINT. --Tango 13:11, 9 January 2007 (UTC)

I can imagine a situation where a user supports, then several users oppose (all based on the same incident), then the original supporter asks a question designed to give the candidate a chance to persuade the opposers to support. Didn't something of the sort happen on TawkerbotTorA's RfA (although that was obviously a special case)? --ais523 13:16, 9 January 2007 (UTC)
I agree with ais523. For example, if I support a candidate but then a couple of people oppose (or are neutral) because of "insufficient edit summary use," I will ask the optional question "will you reset your Preferences to automatically prompt for an edit summary when you forget?" Usually the candidate responds "I didn't even know you could do that, I'm doing it right now" and a concern of the opposers is addressed. I agree one wouldn't support and then turn around and ask an open-ended policy type question. Newyorkbrad 13:22, 9 January 2007 (UTC)
Good point, but that would normally involve a delay between the support vote and the question (you have to wait for the oppose votes), and the question was about almost immeadiately asking the question. It was also about questions that are asked of everybody, which obviously aren't in response to specific oppose votes. --Tango 13:27, 9 January 2007 (UTC)
I guess I was responding more to the original, open-ended question at the top of the threat rather than the narrower one; thanks for the clarification. Newyorkbrad 13:42, 9 January 2007 (UTC)

[edit] We're getting a lot of WP:100 recently.

Has there been a non-controversial RfA in a month that hasn't gotten 100 supports? —Preceding unsigned comment added by Amarkov (talk • contribs)

Don't know. If no-one's done any stats, go and survey the RfAs the past few months. Carcharoth 10:56, 9 January 2007 (UTC)
Mine. Too uncontroversial - And a few running at about the same time Agathoclea 11:03, 9 January 2007 (UTC)
That may have been because they ran over the holiday period. Proto::► 11:14, 9 January 2007 (UTC)

Reproduced below is the chart I have on User:NoSeptember/RfA voting records showing the number of RfAs that have received at least 100 votes (including oppose and neutral votes). I hadn't expected the peak and decline in the number of heavily voted on RfAs to occur. Did you?

Quarter Total number
of RfAs		 RfAs with 100
or more votes		 Pass Fail
1st quarter 2005
64
0 (0%)
0
0
2nd quarter 2005
108
2 (2%)
2 (100%)
0 (0%)
3rd quarter 2005
168
3 (2%)
2 (67%)
1 (33%)
4th quarter 2005
262
6 (2%)
4 (67%)
2 (33%)
1st quarter 2006
241
20 (8%)
11 (55%)
9 (45%)
2nd quarter 2006
235
33 (14%)
22 (67%)
11 (33%)
3rd quarter 2006
213
24 (11%)
17 (71%)
7 (29%)
4th quarter 2006
196
12 (6%)
7(58%)
5 (42%)
1st quarter 2007
??
2 (so far)
1
1

(Total RfAs column and minor alterations by Carcharoth 14:52, 9 January 2007 (UTC)) If you look at the RfAs by date range, you can find all of the RfAs represented in the chart above. NoSeptember 12:37, 9 January 2007 (UTC)

Several explanations suggest themselves: (1) A clique of people swept through RfA voting themselves into power (I know, wrong language there, but it's to make a point that a genuine explanation would be nice to rebut such accusations); (2) Numbers participating at RfA fluctuates over time; (3) Numbers participating in Wikipedia fluctuates over time; (4) Quality of candidates fluctuates over time.
Of course, the WP:100 benchmark is decidedly arbitrary, as it focuses on support votes. But on closer examination, I see No September is correctly using total number of votes, not just the number of support votes. What would be more representative is the total number of votes in a given RfA. Also, what might be interesting is to see how many of the voters over a period of time really are regulars. How many people just turn up and vote in a few RfAs (could be several reasons for this - I've only ever voted in a few because I knew the people involved and felt my input would help reassure those who might not have heard of them), and how many vote time after time after time, week after week after week (and why they are doing this - because participation here is looked upon favourably, or because they carefully research candidates). Carcharoth 12:52, 9 January 2007 (UTC)
At the risk of sounding unappreciative (since I do appreciate the work involved in gathering statistics and presenting information), there is a column missing from the table that might explain a lot - the total number of RfAs each quarter. John Broughton | Talk 14:19, 9 January 2007 (UTC)
I'm guessing that column may appear momentarily... :-) Carcharoth 14:30, 9 January 2007 (UTC)
  • That info could be gotten from the first and third columns of this chart. NoSeptember 14:34, 9 January 2007 (UTC)
  • I updated the table and tweaked it accordingly. Now someone needs to do percentages to make sense of it, I think. Carcharoth 14:52, 9 January 2007 (UTC)
    I've added the percentages (100+ of total, and pass/fail of 100+). --ais523 15:03, 9 January 2007 (UTC)
On the Pass and Fail, wouldn't their percentage of the total RfAs be more meaningful than their percentage of the 100+ vote RfAs? (Although the numbers are too small to be meaningful anyway.) NoSeptember 15:29, 9 January 2007 (UTC)
No, because the value is 'Number of RfAs with at least 100 votes that passed'; I think taking that as a proportion of the total RfA count would be less meaningful than taking it as a proportion of the 100-vote-plus count. --ais523 15:35, 9 January 2007 (UTC)

[edit] Trends within an RfA

Can those who follow RfAs answer a question I have about trends within an RfA in general? Is the normal trend for the first few days of a possibly borderline RfA (one where there is a majority of initial support, but enough opposes to see the number of opposes steadily increasing over time) to see a clear majority for support appear in the initial rush of votes (over the first 2 or 3 days), and then for the remaining 4 or 5 days to see the votes split equally between 'support' and 'oppose' as the arguments for either side succeed in splitting the remaining votes? If so, that would lead to something like the vote stabilizing at around 85% after the initial rush of votes, and then, as equal numbers of support and oppose are added from that point on, the trend is irrevocably downwards. If allowed to run for several weeks, the percentage might drop below 80%, and sometimes in the remaining 4-5 days, the support dips to around 80%. Does that fairly describe a common way for a borderline RfA to develop? Is this a good thing or not? Oh, and to forestall the "its not a vote" comments, I know that RfA is meant to be a discussion, but people do use the percentage as one factor in evaluating an RfA, even though it should not be the only one. Carcharoth 11:06, 9 January 2007 (UTC)

What often happens is many people vote 'support' for every RFA without taking into account the merits of the applicant, usually because they want to make lots of friends as they have an RFA themselves coming up, then when they see some very good reasons to oppose being raised by the editors who actually bothered to look into the candidates edit history, they "change to neutral per Foo" or even to oppose. So there's often a decrease in the support:oppose ratio for that reason. One well reasoned and worthwhile oppose will simultaneously bring in both more opposes from those who have not yet voted, and people changing their support vote to neutral or oppose. As the well reasoned and researched votes don't happen straight way, much of what you see in the first few hours is the effect of that first wave of unthinking 'I support so you will support' support. Proto::► 12:34, 9 January 2007 (UTC)
I wasn't going to mention it, but this 'I support so you will support' mentality certainly doesn't apply in the RfA I am thinking of! :-) I am genuinely interested in the general case as well, so thanks for the reply. The point about well-reasoned votes coming in later is a good point. Carcharoth 12:40, 9 January 2007 (UTC)
I think it's possible to be a bit more charitable about earlier supporters. They may have made very well-reasoned decisions to support a candidate, without having information that becomes available later in the process (on day 4, for example). Also, a bit more research might be helpful to understand vote dynamics. For example, when a count changes from (say) 60-10 to 70-20, was that because 20 new comments split 10-10, or was it because (say) 4 editors switched opinions from support to oppose, and the 20 new comments were actually split 14-6? (And finally, to complicate matters, there probably are a number of "neutrals" who decide during the last few days to switch to "support" or "oppose".)
So I suppose one could suggest a change in process: say, 2 days for opponents to bring up what they see as problems, and other editors to question and comment on those, then a day for supporters to post arguments in favor of the candidate, then a standard window (of, say, 48 hours) for support and oppose opinions to be formally tallied. That way, there would be fewer "uninformed" early voters. (Logrolling, however, probably wouldn't be affected much.) John Broughton | Talk 14:13, 9 January 2007 (UTC)
Oh, I'm not saying all early supportes don't really pay attention to who the candidate actually is. Just some of them. Proto::► 14:16, 9 January 2007 (UTC)
Logrolling? LOL! Nice term. As that article says, over here, we are more used to the terms quid pro quo and horse trading. Carcharoth 14:28, 9 January 2007 (UTC)

[edit] Ban asking everyone the same questions

Let's try and resolve this once and forall: Can we ban these copy and paste questions that people put on every RfA and introduce a system for getting questions added to the "standard questions"? If something is going to be a standard question, it should be done properly, it should not be decided unilaterally, which is exactly what these copy and paste questions are. --Tango 13:31, 9 January 2007 (UTC)

There is every chance that the cut and paste questions might be ambiguous, while the standard questions are not. I see no problem with this format. — Nearly Headless Nick 13:35, 9 January 2007 (UTC)
If the repeated questions get added to the standard questions, then there will be no reason to repeat them. If they aren't added, then if people are still interested, people should still be able to ask them. Anything that will limit peoples ability to get the information they'd like to make a decision will be voraciously opposed by me. --badlydrawnjeff talk 13:36, 9 January 2007 (UTC)
Fully support. There should be a review process to introduce and retire questions. Having someone just decide to copy/paste the same into every RfA circumvents any sort of oversight. --StuffOfInterest 13:37, 9 January 2007 (UTC)
I couldn't agree more with StuffOfInterest. Some of those optional questions are better than the compulsory ones (I'm not talkingabout the age one, btw). JorcogaYell! 13:42, 9 January 2007 (UTC)
So what would you do with someone who asks every candidate questions on their user talk page instead of on the RfA, and then votes "Oppose, didn't answer my questions" if they don't answer? NoSeptember 13:50, 9 January 2007 (UTC)
As with any other bad reason, the crat will not give their !vote as much weight. --Tango 14:05, 9 January 2007 (UTC)
I'd hope that's not considered a "bad reason." --badlydrawnjeff talk 14:06, 9 January 2007 (UTC)
If this proposal has concensus, then it would be a bad reason. --Tango 14:13, 9 January 2007 (UTC)
People should have the freedom to ask any questions they see fit to the candidates, even if they have been used before. Lets not introduce bureaucracy over this. Having an "official" set of "legitimate" questions that "should" be answered would be even worse. I don't like the form-letter style questions, but they're preferable to the alternatives. — David Remahl 13:55, 9 January 2007 (UTC)
Absolutely. More talk, less bureaucraticfuckism (neologism credit: freakofnurture). — Nearly Headless Nick 14:00, 9 January 2007 (UTC)

I'd like to clarify what questions would be banned under my proposal. It's not so much asking everyone the same questions that's a problem, it's asking questions indiscriminantly. If a question is relevant to a particular candidate, it's fine to ask them, even if you've asked the same question to lots of people. For example, asking every candidate that is active in a wikiproject: "I see you are a member of WikiProject X. Do you intend to use your admin tools primarily to help that project, or will you be working on Wikipedia as a whole?" would be a perfectly acceptable question, even though a large proportion of candidates could be asked it. A question should only be considered relevant to everyone if there is concensus to support it, and if there is such a consensus, it can be added to the standard questions. --Tango 14:05, 9 January 2007 (UTC)

This proposal reads worse now than it did in its original form. A horrid idea. --badlydrawnjeff talk 14:06, 9 January 2007 (UTC)
Second the "horrid idea". If a user is being disruptive, we have methods of dealing with that. This is rules creep, and unwarranted. KillerChihuahua?!? 14:08, 9 January 2007 (UTC)
WP:CREEP? — Nearly Headless Nick 14:10, 9 January 2007 (UTC)
Yes, we have ways of dealing with disruptive users. This discussion is trying to determine if asking indiscriminant questions is disruptive or not. --Tango 14:12, 9 January 2007 (UTC)
Discussion is never disruptive. --badlydrawnjeff talk 15:13, 9 January 2007 (UTC)
I disagree. Repeatedly taking something to AfD when it's been closed with "keep" every time and not giving any new reasons is disruptive, but it's just discussion. --Tango 15:15, 9 January 2007 (UTC)
No, that's abuse of process. Apples and oranges. --badlydrawnjeff talk 01:47, 10 January 2007 (UTC)
RfA is a process that can be abused, just as much as AfD can. Asking questions that most people don't want asked (if they wanted them asked, they would become standard questions) could be considered abuse of process. (That's pretty much what this discussion is trying to decide.) --Tango 02:11, 10 January 2007 (UTC)
Could you explain how it's worse? --Tango 14:12, 9 January 2007 (UTC)
There's nothing stopping a candidate ignoring a question if they don't want to answer it. Tango, there's also nothing stopping you or anyone else removing questions you feel are unwarranted or unnecessary (and do you mean indiscriminate? :p ) Asking the same question of every candidate isn't disruptive. It is, however, a bit annoying, particularly as many of the questions are inane. If it bothers you that much, remove them. Proto::► 14:15, 9 January 2007 (UTC)
I think removing other people's questions without a concensus to do so would be generally considered vandalism. (And yes, I mean indiscriminate... I'm not sure where that extra 'n' came from, thanks!) --Tango 14:22, 9 January 2007 (UTC)

Please review WP:VAND. — Nearly Headless Nick 14:27, 9 January 2007 (UTC)

Is that targetted at me? I'm well aware of what WP:VAND says. That doesn't change the fact that if I removed a question, I'm pretty sure someone would revert it. --Tango 14:45, 9 January 2007 (UTC)
Yes, targeted at you, without reservations. There is a difference between WP:POINT and WP:VAND, and we always go by community consensus. — Nearly Headless Nick 14:47, 9 January 2007 (UTC)
Call it what you like, but if I remove a question without concensus, I will be reverted. --Tango 15:12, 9 January 2007 (UTC)
If there was a particularly poor question, or a dunderheaded one that was a variation of a dunderheaded one that had already been asked (or so on), then you removing it and putting in <small> 'This question was removed by Tango, see talk page' </small> would be fine - anybody who labelled that as vandalism would be both a churl and a fop. Proto::► 15:28, 9 January 2007 (UTC)
Wait... dunderhead, churl, fop... let me write these down - I'll need them for my next melt down :). NoSeptember 15:36, 9 January 2007 (UTC)

Amazing, I agree with NHN twice in one week. Taken from the RfA page: "Who may participate: Any user in good standing is welcome to participate in discussions on candidates." As I see it, there are basically three activities an admin does that your regular editor cannot do: block disruptive editors, delete pages, and protect pages from editing. It goes without saying that these are powerful functions, but they are the same whether or not an admin focuses in one particular area of the 'pedia. Certian editors may hold a particular view about one of the functions, especially if they regularly particpate in one area such as XfD and may have the same concern about all editors. So you may see this editor ask the same question of every nominee. I don't think there is anything necessarily wrong with this and to restrict this would be unnecessarily stifiling to discussion. I do think that questions should be crafted to be the least disruptive and to try to not lean toward any particular Wikipolitical point of view. Controversial questions can be asked, but every attempt should be made to prevent them from being leading questions. —Malber (talk • contribs) 16:01, 9 January 2007 (UTC)

  • I think that Tango has a very good point; the standard questions are kept to a minimum for a reason, and any one editor should not unilaterally decide that there should be more, especially as not answering questions tends to lead to oppose-votes. Note that there have been leading questions lately. In particular, the IAR/SNOW question is very obviously a leading question; it's just as factionalizing as when (about a year ago) people would oppose others for being inclusionist/deletionist. >Radiant< 17:05, 9 January 2007 (UTC)
    I completely agree with this. Fault found with Tango's proposed solution does not mean there is not a problem. -- Renesis (talk) 20:47, 9 January 2007 (UTC)
  • I don't object to anyone asking legitimate questions. I dislike the recent practice of some editors slapping the same questions on each and every RfA. However, rather than banning them, I'd simply move the place where optional questions are to be put. That way it removes the patina of their being de facto standard questions. See the discussion at the RfA template talk page. Agent 86 19:00, 9 January 2007 (UTC)
  • I oppose restricting who can ask what questions. I am not too sussed about where the other questions go but think it's better form that they fit in right below the first 3. Put a divider in, sure. But not move them to talk as someone suggests above. I have to say in general, I have some bias against moving things to the "talk page"... I've contended in the past that this isn't a vote, it's a discussion, so why would we even need to have a talk page? :) Nevertheless I do think really long things maybe can move there. And optional info like edit counts :). ++Lar: t/c 19:46, 9 January 2007 (UTC)
  • I dislike people asking each and every nominee the same optional questions, but to be honest, it's no longer the case that people get loads of oppose votes for not answering them, so I wonder whether this is a big deal or not. I don't think it is. --Deskana (For Great Justice!) 01:56, 10 January 2007 (UTC)
Now, that would be an excellent solution: Everyone just ignoring the questions. Maybe it's worth waiting and seeing if the problem does solve itself in such a way. --Tango 02:11, 10 January 2007 (UTC)
Interesting thought. --Deskana (For Great Justice!) 02:21, 10 January 2007 (UTC)
  • I suppose we could add the question "are you aware that you may ignore all these questions?" :) >Radiant< 08:47, 11 January 2007 (UTC)
    If you did, it might just be ignored... --ais523 09:39, 11 January 2007 (UTC)
  • Leading questions can be removed by users and discussion can continue on this page. — Nearly Headless Nick 09:44, 11 January 2007 (UTC)
    • I suspect that if one would remove a leading question, the person who added it would just put it back and add an accusation of vandalism. >Radiant< 10:43, 11 January 2007 (UTC)

Now here is a really way-out idea, which I will probably get shot down in flames for. OK, I can take it. Now, comments, which as we know should not be called votes, should be made on the basis of a users edit history and edit count and answere to the questions put; but quite often it seems to me, !voters comment not on these parameters but on oppose/support comments already made. So, my suggestion: how about arranging things so that !votes are not visible to other editors? That way, everyone decides without being influenced by what anyone else has said. It's novel. It's reactionary. It may not be possible. But is is reasonable? I would value comments. if I get bitten, I have a thick skin.--Anthony.bradbury 21:56, 13 January 2007 (UTC)

See the secret ballot discussion below. In full, bad idea, because then there is no discussion. Even though it is technically possible. -Amarkov blahedits 21:58, 13 January 2007 (UTC)

[edit] Subheadings in RfAs?

As RfAs get longer and longer (that is, more people participate), it gets more difficult to add a comment. I noticed that one of the present RfAs has a subheading - perhaps inadvertently - and thought that these would help. Obviously this shouldn't get out of hand - I'm suggesting only three (Support, Oppose, Neutral). They'd make the TOC longer, but editing (much) easier, I think. Am I missing something? John Broughton | Talk 15:40, 9 January 2007 (UTC)

Seems like a good idea. It has been tried before, but someone reverted that user. I cannot recall the time period. — Nearly Headless Nick 15:45, 9 January 2007 (UTC)
It was reverted before due to messing up the bots, but the style used on Wikipedia:Requests for adminship/ProtectionBot seems to work for both humans and bots; it displays headings when viewing the RfA individually, but not on the main RfA page. --ais523 15:47, 9 January 2007 (UTC)
(edit conflict) I remember reading Tim Rogers at Next Generation (a gaming site) about how Famitsu (a japanese magazine) conducted his poll: they gave a long list of games, a long, completely unsorted, list. They weren't sorted by release date, system, not even alphabetically. You needed to read every entry (maybe over 500 titles) once to know which ones have been chosen, and then again to locate the one you were voting at. In other words, if we create subsections, people would just click in the subsection and add their opinion. While they won't read every opinion until finding the place for adding theirs, by forcing them to load the full debate you are at least making him scroll down to the section. I would prefer removing the sections like the old proposal, and forcing everyone to discuss and !vote in the same subsection. -- ReyBrujo 15:52, 9 January 2007 (UTC)
I see your point. I think the idea, though, is to make it easier to vote for those who have trouble loading large pages. I think. I don't have problems with large pages, but have heard that some people do. Carcharoth 17:36, 9 January 2007 (UTC)

I dislike it when the subheadings are introduced. The TOC becomes far too long, and it does not help the readability of the individual RfA if one likes to read the whole thing. Navigating a particularly long RfA isn't difficult. If scrolling is too much, there's always CTRL-F to search for the point in the discussion where you want to insert. Agent 86 19:03, 9 January 2007 (UTC)

Please observe the current ProtectionBot RfA. Subheadings are hidden on the main page, thus not making the WP:RFA table of contents eleven miles long, and yet allow for reduced edit conflicts on the RFA itself. I preferred the orginal idea/solution wherein individual sections got edit headers without expanding the RFA TOC, but this works okay in this case. I don't think it needs to occur in every single RFA from here forward, but it's an option when things start to become cluttered, and doesn't seem to be breaking Dragon_flight's or TangoTango's bots. -- nae'blis 19:19, 9 January 2007 (UTC)
If the 2 bots are still working, then I've no objections at all. --Kind Regards - Heligoland (Talk) (Contribs) 19:47, 9 January 2007 (UTC)

I endorse subheadings on overly-long RFAs. It really does make it easier to find the place in the edit text that you want to insert a response to. --Cyde Weys 19:40, 9 January 2007 (UTC)

I like the idea, but it doesn't work very well in practice. See ProtectionBot's RfA for an example of what it would look like in action. It works fine, but people then screw it up because they put their comments right in the middle of the noincludes. See [1], [2], [3] and probably more that I'm too lazy to find. --Rory096 20:56, 9 January 2007 (UTC)

  • Further to my comment above, if we can hide the subheadings on the main page of the RfA, as with the ProtectionBot RfA, I can get on-side. I'm not too worried about people mistakingly entering their comment on the wrong place. It won't take long for people to catch on; besides, there are already times in RfA comments under the existing format that require another editor to go in and do a format fix. Agent 86 21:08, 9 January 2007 (UTC)
    • Perhaps when the next RfA hits (say) 100 opinions, if it still has at least a couple days to go, it might be good to do (another) pilot (test) to see how people react. John Broughton | Talk 01:22, 10 January 2007 (UTC)
I like this idea, and I agree with Agent: if subheadings become more common, people will become accustomed to them. They would also be helpful to me since I often get edit conflicted when !voting on RfAs, and editing a separate subheading would make these conflicts less common. Dar-Ape 01:28, 10 January 2007 (UTC)

I think subheadings are a good idea. A much larger TOC is worth living with for the sake of easier editing, I'd think. Oleg Alexandrov (talk) 03:37, 10 January 2007 (UTC)

We could remove the TOC totally and have some other form of list. I know this week is rather mad at WP:RFA but 12 RfAs with 4 sections each is 48 sections on a TOC which would be too long. A standard Wikilink in a table or some such thing would do. We don't need to assume an inability to edit Wikipedia with RfAs, candidates really should be capable of that sort of basic editing. --Kind Regards - Heligoland (Talk) (Contribs) 03:57, 10 January 2007 (UTC)
What's wrong with the noinclude/includeonly headings like on ProtectionBot right now? Headings on sub-page only, normal TOC on main page. A win for everyone right? -- Renesis (talk) 05:33, 10 January 2007 (UTC)

I entirely agree with the idea of subsections - and in fact suggested exactly this not so long afgo, as can be seen here. If the bopt problems can be negotiated, then it still seems like a very good move. Grutness...wha? 05:48, 10 January 2007 (UTC)

[edit] This is not even remotely working

User statistics

We have 3,214,867 registered user accounts, of which 1,087 (or 0.03%) belong to administrators. [ Source ]

At one time I was very proud: Of all the wikimedia projects, en.wikipedia had the highest admin to non-admin ratio. But now, only 0.03% of all registered user accounts belong to admins.

Admin abilities are really quite trivial. Like a drivers licence is a "won't destroy the city in traffic" licence (if even that), an admin bit is more like a "won't blow up the wiki" licence. It probably should have been renamed long ago, but "admin" happens to be an old misnomer, and it seems to have stuck. In reality, Stewards are the people who actually have some measure of administrative ability.

Currently we have situations where people are running sub-projects, running sub-organisations or even doing mediation... but they still don't even have that measely admin bit!

Due to the limitiated number of admins, a collosal bureaucracy has been created in places. These procedures are there so that at least some amount of fairness can be retained. However, bureaucracy is much less empowering than consensus. As a result, we get stuck in a continuous negative spiral of more rules and less power.

What I would like to see is an admin/non-admin ratio closer to 90% (or at least 90% of active contributors, see notes below). Everyone should have a licence that says they're a trusted editor!

So something is rotten in adminship country. Do people have suggestions on what to do about it?

notes:

  • Of the 3.2M registered users, many are single use accounts, due to certain current not-so-smart policies. If only all the most active wikipedians get the admin bit, that would be ok too. We would need to promote perhaps another 10000 people or so?
  • Don't argue that you've been here for x months and don't see anything wrong because it's always been that way. It wasn't this way before you came here x months ago. How do you think wikipedia was when practically all regular users had that admin bit? Would you still need DRV? Would AFD need to exist? How about PROD? If anyone who is moderately trusted can review and undelete, do you need so much bureaucracy? How about blocks? Consider! :-)
  • If we only have a small exclusive club of 1000 admins, isn't that essentially just Esperanza by another name?
  • Several times when dealing with RFA nominees, I came across people who didn't know or understand The policy trifecta or our m:Foundation issues. We need some way to enforce a *minimal* standard that is somewhat sane. We could tell people to come back in 3 months, or we could tell them to come back after just reading those, and prove they understand them now. The lattersaves a lot of time! :-)
  • You can easily learn to wisely use the abilities granted with an admin bit in under 30 days, if you are willing to practice and read a lot. Give it twice or three times that if you really want to play it safe. I think we are dealing with some kind of requirements inflation, where people just set harder and harder requirements, without thinking about what that means, or what should be achieved at that level of experience. Instead of basing your ideas on experience on raw seniority, consider what a person should actually *be able to do* (like be able to pull apart two other editors who happen to be fighting, or to be able to personally solve a case before it comes to arbitration, or that this person is trusted by developers and can ask them to make small changes to the mediawiki engine, to name some skills of advanced users. Note that in the past, people only learned these skills after becoming an admin, and adminship was considered a prerequisite. Has that now turned around?)
  • This is a scaling problem. We might want to consider creating a new admin-creation system that addresses scaling issues rather than merely applying further band-aids.

-- Kim Bruning 17:07, 10 January 2007 (UTC)

Of the 3.2M registered users only a few thousand are what would typically be considered active. The latest stats are from the end of October 2006, when 4,300 registered users made over 100 edits in that month. While this doesn't take into account edits by anonymous users (IPs), the admin percentage is surely far higher than .03%. However, I do agree with many of the points you made, especially concerning doing things a certain way just because that's the way they've always been done. SuperMachine 17:19, 10 January 2007 (UTC)
I just want to agree wholeheartedly with Kim's direction, and take the opportunity to remind people that I made some statistics which show that there are only 500 active admins and in the last two years over half of all the admin actions have been made by just 59 admins. That is a recipe for a cabal. Jim182 17:20, 10 January 2007 (UTC)
  • The percentage is meaningless. The number of vandal and throwaway accounts is always increasing at a faster rate than the number of active users. Even if the admin bit were automatically given to every single user account that had more than 50 edits, the percentage of admins out of registered accounts would still approach 0.
Hence the caveat "or perhaps only just the active users", I think the percentage is 10%-20%, it's still low. Kim Bruning
"Active" is defined by the statistics simply in terms of edit counts. I know of users who have 2000+ edits who are here purely and simply to push a particular point of view on a small set of articles. There were only 4,700 users who had more than 100 edits in October. That is, we already had 20% admins, though that would be a misleading statistics, and a not insignificant portion of those user accounts with 100+ edits are POV pushers and sockpuppets. —Centrx?talk • 18:29, 10 January 2007 (UTC)
Are you saying that being an active user is more likely to indicate a POV pusher or vandal? If so, would you be in favor of removing the "community" entirely, in favor of anon editors? Kim Bruning
No, I'm saying that liberally giving admin tools to get 10,000 more admins, or even 500 more admins, would result in a significant number of them being vandal sockpuppets or POV pushers, and if that laxity were to continue would result in a significant number of admin sockpuppets. —Centrx?talk • 02:11, 11 January 2007 (UTC)
Straw man. I would like to see both actual minimum standards, and better means of evaluation. Kim Bruning 02:29, 11 January 2007 (UTC)
I don't see any better means, and an admin ratio of 90% of "active contributors" is not the same thing as 90% of "active contributors I trust through some unknown better means of evaluation", and having sysops after less than 30 days really does mean vandals won't happen to be caught up in a checkuser request, or that someone really can decide they want to become a sysop and mess up Wikipedia over school vacation. —Centrx?talk • 02:54, 11 January 2007 (UTC)
I can see several directions that can be taken. I've also seen other people in this discussion hint at them. :-) Just because you don't see it doesn't mean other wikipedians don't. But before we can try to figure out something better, it helps to explore the issues with the current system. You're expecting me to take all the steps in a single leap. :-)
In fact, several of the issues at hand do have known solutions. It's a question of finding the issues and putting the solutions together in some elegant way. We'll get to that! Kim Bruning 16:46, 11 January 2007 (UTC)
I meant I didn't see better means being proposed, or any means being proposed to get 10,000 admins. I myself have proposed the Discuss first, vote second idea several times. —Centrx→talk • 15:30, 12 January 2007 (UTC)
  • Unless we get Wikipedia:Administrator's noticeboard/Chess games and unless all administrator decisions have to be referred to a governing board before being implemented, being an administrator will be nothing like Esperanza. Almost no admins at all have an "Admin" link in their sig—and such a link might actually be useful for helping people—but that was the most common thing among Esperanza "members". There is no "club".
Admins can be seen as a different kind of club, but a club nevertheless. See the whole giano and #wikipedia-en-admins situation, for instance. Kim Bruning
Any such club exist separately from whether a person has an admin bit. There are clubs with no admins, and there are a large number of admins who have never used IRC. —Centrx?talk • 18:29, 10 January 2007 (UTC)
*Sigh* Yes, but groups are being divided into admin, and non-admin, with one of the groups being considered "better" than the other. Sound familiar? Kim Bruning 19:54, 10 January 2007 (UTC)
And now you've gotten to the crux of the issue. It isn't the sysop/non-sysop distinction that's the problem. It's the regular user/better user divide that's harmful. In theory, sysops are regular users who are trusted by the community to carry out certain tasks on its behalf. They are not supposed to be super-users whose opinions outweigh others' and who needn't be held accountable for their actions. It's this preferential treatment (not the sysop bit) that fuels talk of a "cabal," and it's the air of superiority that makes for an Esperanza-like clique. —David Levy 23:42, 10 January 2007 (UTC)
And if admin tools were given to all "established" users or any "trusted" user, there remains that split, maybe even more so because there would less gradations of trust. If all established users had the air of superiority, all the newcomers would be bitten. —Centrx?talk • 02:14, 11 January 2007 (UTC)
Yeah, it wouldn't help. Right now, you can honestly say "No, we're not a superior cabal; look over there where there are some respected editors who aren't admins!" If all respected editors are admins, you really have no grounds to say that. -Amarkov blahedits 02:17, 11 January 2007 (UTC)
And that's why I was talking about my own admin bit with Cyde Weys, somewhere in this section. :-) Kim Bruning 02:29, 11 January 2007 (UTC)
  • Banned vandals can easily game the system to become an admin in thirty days, too. There was no one actively trying to screw up Wikipedia, or use it for political manipulation, when admin bits were handed out easily. If you added 10,000 admins, and made it so trivially easy to become an admin, you would need to add another tier of users who can desysop quickly or who otherwise over-ride the other admins. In other words, the current or slightly higher standards for admin-ship would just become a level of "super-admin". What is the cost if these un-vetted admins ban an established contributor, who gets pissed off and never comes back? What is the cost when admins protect pages in an edit war, and edit war on protected pages? This happens currently. What happens when becoming an admin is so easy that a vandal or edit warrior has 10 admin accounts?
The point is that people are looking at raw statistics. You re-iterate that point. I think we're in agreement? Instead, we need to train people so that they meet some minimum requirement (see above), and we need to check for actual skills and abilities we'd like to see, as opposed to using arbitrary, unrelated statistics. We need to pursue both higher quality and higher throughput all at once. (typically if you want both at the same time, you're going to have to put in more work. So be it. ) Kim Bruning
What are the problematic statistics used? The statistics people use are things like Wikipedia-space edits: Is someone who hasn't done anything in any policies or processes going to be familiar with anything about those policies and processes? Why does someone who has hitherto not been interested in any Wikipedia processes want to suddenly adjudicate them? Why should that person not learn about the processes before becoming an admin? Article editing and being a user for a long-time are a heuristic for deciding whether a person is trustworthy and is interested in Wikipedia as an encyclopedia. That can't be done if a person has never discussed anything with anyone or has only done edits that can be automated by anyone. —Centrx?talk • 18:29, 10 January 2007 (UTC)
Have you ever considered the amazing possibility of - oh for instance- actually looking at the edits themselves, and judging them on quality? If you go "that's no longer practical because some people are requiring 3000 edits", then bingo, you got it in one. :-) Kim Bruning 19:54, 10 January 2007 (UTC)
Yes, that's exactly what I do, but there is no need if the candidate says they want to "help out at AfD" yet has never participated in an AfD discussion. Similarly, if a person has even read any policies, they should have at least a few edits to the policies or the talk pages. Did they seriously completely agree with everything in every policy they've ever read and did not think any changes were warranted, or have they never read a policy at all, or come across a gem of wisdom to add to it? Did they not even find any spelling errors to correct? These edit counts people use are just bare minimums to cut out the chaff. —Centrx?talk • 02:27, 11 January 2007 (UTC)
Now there's something that can be corrected quite rapidly with (for instance) some kind of training or community introduction, no? Kim Bruning 16:46, 11 January 2007 (UTC)
It is simple to figure out how to get involved with these things. No training system is needed. People can just get involved, and if they have any questions they can just ask others. There are numerous venues for asking questions, we don't need any more. —Centrx→talk • 15:32, 12 January 2007 (UTC)
  • Have the project managers and mediators you mention as not being admins applied for adminship? It seems right now that it is pretty easy for someone to become an admin, but a lot of people might not apply at all. The problem, then, would be getting them to apply.
Since the failure of the Amgine RFA, consensus is that project coordinators might want to consider *not* running, in fact. Kim Bruning
As far as I can tell, this person had little activity on the English Wikipedia, which raises questions as to why they would need admin tools or whether they would make mistakes. While they can be trusted, it doesn't help Wikipedia much if they just come by every few weeks and make a mistake that has to be corrected, and it's not much of a loss if a project coordinator who is never going to use the admin tools anyway doesn't have them. —Centrx?talk • 18:29, 10 January 2007 (UTC)
International coordinators might find it useful to edit particular site-notices and protected templates from time to time, and might find it useful to not have to bug a steward or what have you to do it for them every time. :-) But that said, either you trust a person to not make mistakes, or you don't trust them to not make mistakes. You can't have both at once, which is it? Kim Bruning
They can be trusted to be honest and have good will, but not be trusted not to make mistakes. —Centrx?talk • 03:04, 11 January 2007 (UTC)
—Centrx?talk • 17:32, 10 January 2007 (UTC)
Now that we've already limited scope to such a degree, I'd be very interested to hear what classes of error can be made within that scope? Kim Bruning 16:46, 11 January 2007 (UTC)
--Kim Bruning 17:46, 10 January 2007 (UTC)
-- Kim Bruning 19:54, 10 January 2007 (UTC)

I think that too many people when !voting at RfA are prepared to !vote oppose on technicalities for people that they actually would trust with the tools. --Dweller 17:46, 10 January 2007 (UTC)

Perhaps the problem lies in the !voting itself? Kim Bruning 17:49, 10 January 2007 (UTC)

The problem with your driver's license analogy, is there aren't people out there who want to get one so that they can destroy the city. That is naturally not in their best interest; but people have nothing to lose when it comes to vandalizing Wikipedia. -- Renesis (talk) 17:49, 10 January 2007 (UTC)

The stated reasons for adminship and for drivers licences are the same. We are concluding that this person won't accidentally or deliberately cause damage. How we should come to that conclusion is the matter at issue. Kim Bruning 17:57, 10 January 2007 (UTC)
However, there are several significant differences between granting someone a driver's license and adminship. At least in the majority of the US, drivers are required to purchase auto insurance as well as pass written, vision and driving exams. Quite often, they're required to attend driver's education classes where they have instruction in both the classroom and vehicle. Also, driving infractions usually result in fines, suspension of licenses or even criminal charges. Compared to this adminship is quite easy to get and maintain. SuperMachine 20:35, 10 January 2007 (UTC)
That's because with a drivers licence, you can go out and *kill someone* with a vehicle massing at least a metric ton. I did actually find it easier to get my drivers licence (as expressed in hours of work), than the current RFA requirements. (6 months, and 3000 edits... that's an awfully large number of hours of work!) Kim Bruning 21:01, 10 January 2007 (UTC)
You're right about the number of hours, that is an interesting way of looking at. ;) Still, there's huge bureaucracies (RMV) and industries (auto insurance) built around licensing, insuring and holding drivers responsible should they do something wrong. There are also traffic courts specifically for determining if a driver did something wrong. That's a whole lot of overhead that we don't really want or need on Wikipedia. SuperMachine 21:10, 10 January 2007 (UTC)
And yet that's what we're building :-) Kim Bruning 00:48, 11 January 2007 (UTC)
If you go out and kill someone in a vehicle, you are likely to be caught and imprisoned or killed yourself. If you maul page histories with an admin-bot or steal passwords using the MediaWiki namespace or javascript files, you just have to move to another account, and you keep your stolen passwords and someone else cleans up after you. —Centrx?talk • 03:12, 11 January 2007 (UTC)
Does it kill anyone? How much damage is done to society? How much damage is even done to wikipedia? Kim Bruning 16:49, 11 January 2007 (UTC)
See [4]. —Centrx→talk • 17:51, 11

January 2007 (UTC)

Some of those points are not possible in this universe. Others assume a proposal I have not made. The remaining points do not constitute very large damage to society or wikipedia. The conclusions do not follow. Kim Bruning 00:20, 12 January 2007 (UTC)
No, they are quite possible, and if a sufficiently intelligent person were doing them, all the new admins would have to be desysopped or it would go on regularly for weeks. I won't illustrate it though. Yes, this is in response to a proposal of adding 10,000 admins or 90% of active contributors after less than 30 days. I see no system whatsoever to reliably do that. —Centrx→talk • 15:37, 12 January 2007 (UTC)

I fully agree with Kim Bruning's comments. In my own case, I'd be glad to serve as an admin and to help out where I can, and believe that I'm qualified to do so. But I haven't even considered an RfA because the requirements for passing with a favorable number of !votes seem so demanding -- I don't have a gazillion edits with the correct percentage in each space and so on. Raymond Arritt 18:00, 10 January 2007 (UTC)

Kim, if we are so short on admins, why don't you become one again? You didn't leave "under a cloud", so you can get it back at any time just by asking. It's kind of weird to see you calling for thousands of new admins when you won't even do it yourself anymore. --Cyde Weys 18:12, 10 January 2007 (UTC)

Actually there's another side to this whole story where I'd also like to make it possible for people be released from their adminship-sentence after a while. Of course, this will only be possible if we have enough replacements! (So it's two sides of the same coin)
I'm currently working on that. Apparently the loss of ability to review deleted revisions is a stumbeling block. Kim Bruning 18:46, 10 January 2007 (UTC)
Are you saying that admins that go inactive (as admins) don't bother handing the bit back because they like to be able to look at deleted pages? I liked Kelly Martin's idea of enforced 1-month breaks for admins to either have a complete break, or do more editing. Enculturation really just needs people to talk more and demonstrate how they do things (rather than just doing them). Takes certain types of people to be role models. Actually following someone's edits, or meeting in person and watching how they do things, can be very instructive. Carcharoth 23:32, 10 January 2007 (UTC)

There's no reason a new adminning process needs to be exclusive of this one. All we need is something reasonable that will be honored by one or more of the bureaucrats; both can run for a while. A few points:

  1. Making adminship more plural will increase inter-admin conflicts, which, bad as they have sometimes been, have not been nearly as bad as they can be.
  2. No one can really agree on what exactly admins are supposed to do, and for what reasons. There's a significant number who think that admins are entitled primarily to unquestioned personal judgment. This is the reason RFA seems so broken--it's almost the only control point anywhere in the admin career cycle.
  3. While you increase the admin:non-admin ratio, you need to increase the ratio of checks on admins to admins as well. There is almost nothing there besides, in theory, the arbcom. But the arbcom isn't a solution to anything

Demi T/C 18:15, 10 January 2007 (UTC)

I made a proposal for a replacement process some months ago on my blog. I'm sure that any number of people will have objections to it (Demi, for example, would almost certainly object based on his comments above, and I know Kim objects to the mandatory retirement clause). However, I still think that something in this general vein would be a net long-term improvement, even though it would likely be very disruptive in the short term to the portion of the Wikipedia community that is preoccupied either with getting adminship or deciding who should be an admin. As I am not, and have not been, a part of that portion of the community for quite some time now, I'm not really invested very much in this, but I offer the proposal linked above as a point from which to begin discussion. Enjoy. Kelly Martin (talk) 19:36, 10 January 2007 (UTC)
Interesting read, Kelly, and an interesting reply by David Gerhard. While I see a need to get away from evaluating a nominee based on edit counts and would appreciate getting back to adminship being "no big deal" I think Wikipedia has out grown this ideal. Sure, admin actions can be overturned, but a bureaucracy has grown around reversal of admin decisions. Deletion is still one of the most disruptive things an admin can do and yes it can be reversed. But to do so you have to wade through the morrass that is WP:DRV. This may work on Meta, but if we made everyone an admin on Wikipedia after one month without much review I think we'd see massive wheel warring. However, I do think we should find a way to make the RfA process more of an evaluation, less political, and less seeming of a popularity contest. —Malber (talk • contribs) 20:02, 10 January 2007 (UTC)
If everyone is an admin, then you don't need DRV, because you can overturn an admin descision yourself. Wheel Warring would be less disruptive as well, since the collateral damage to people who are other admins anyway would be minimal. This is an expansion on one of my original points here. :-) Kim Bruning 21:03, 10 January 2007 (UTC)
I don't understand your "don't need DRV" comment. Sysops aren't entitled to undelete stuff on a whim; DRV applies as much to them as it is to anyone else. This is true no matter how many sysops we have. —David Levy 23:42, 10 January 2007 (UTC)
It's a consensus system, and everyone needs to provide input, right? If you do not have the power to overturn a deletion yourself, there needs to be some bureaucratic system in place to allow you to petition to overturn an admin descision. If you're an admin yourself, sure, you can just overturn a deletion on your own whim. If two well trained users meet in this way, they can then solve the situation rapidly and amicably, and cut Deletion review out of the middle. (see Bold Revert Discuss) Kim Bruning 00:31, 11 January 2007 (UTC)
I strongly disagree. No sysop action should ever be undertaken on a whim, regardless of how easy it is to become a sysop. DRV serves as a means of determining community consensus (not merely as a workaround for users lacking the sysop bit). —David Levy 00:50, 11 January 2007 (UTC)
I won't go into why Deletion Review is such an awful means of determining consensus right now, because that's an entirely different discussion (which I'll have to go into at DRV at some point in time).
What I can ask right now is for you to consider how wiki-based consensus usually works. When there are no big conflicts, and things are going along well, how do people edit? Are you familiar with the Harmonious Editing Club? Why do their methods work so well? Kim Bruning 16:57, 11 January 2007 (UTC)
  • They don't work so well, actually. The HEC membership list contains several known edit warriors. FYI. >Radiant< 17:01, 11 January 2007 (UTC)
If you're edit-warring, I don't think you're being very harmonious. :-P Some people join HEC to improve their skills at editing peacefully. If you can show that the HEC rules -when applied consistently- can be used in scary edit-warry disruptive ways, that would be interesting. (And also mean there's no hope whatsoever btw... but that's a different story). I don't think it's going to be easy to show that that though! ;-) Kim Bruning 00:25, 12 January 2007 (UTC)
What do you think it would be like if it were trivially easy for any idiot who thinks Wikipedia is a game, for any vandal, for anyone at at all who is currently actively vandalizing pages or using Wikipedia as a propaganda tool, to become an administrator? What do you think it would be like if the people who currently blank pages with insane screeds could move a bunch of pages into a deletion bin-bucket? Conversely, because under such a system you would need an efficient way to quickly remove admin tools, what do you think it would be like if any administrator could be desysopped on the slightest cause? How do you determine what is wheel warring when the person "wheel warring" is actually undoing the actions of several admin sockpuppets? These are ill-formed proposals, and I do not even see what the necessity of them is. The reason we have backlogs is because the current admins aren't doing anything about them. Half of the current admins aren't active. —Centrx?talk • 20:25, 10 January 2007 (UTC)
Some of us "inactive" admins are busy writing encyclopedia articles.
The reason we have backlogs of admin tasks is that we have an increasing number of tasks that require an admin bit, but the numbers of people with admin bits is not keeping pace with the tasks. Many hands make light work. Adminship really should be no big deal, and handed out much more widely to any editor in good standing (and taken away more easily where there is a problem). -- ALoan (Talk) 21:03, 10 January 2007 (UTC)
The reason we have an increasing number of tasks is that we allow any new account to upload images and create new pages. This is not because of an increase in the highly active users who would be editing articles, it is because of an increase in the general popularity of Wikipedia so that random passers-by come here to have fun, advertise, and use Wikipedia as a webhost. You would have a harder time writing articles if anyone could delete them and block you on a whim; after all, the person has several admin sockpuppets, they blocked you with their admin throwaway account, and what sophisticated mechanism is there for desysopping anyway? It's nice to say "everyone is an admin", but how would this supposedly work? —Centrx?talk • 00:42, 11 January 2007 (UTC)
Strawman. We are advocating improved adminship procedures. Better procedures means both better checks and better throughput. It's not either-or. Kim Bruning 02:29, 11 January 2007 (UTC)
You started off saying we should have maybe "10,000" more admins, "90%" of active contributors, and perhaps admins being made admins after "less than 30 days". I don't see much in the way of what an "improved adminship procedure" would look like, except for something that would look exactly like RfA except with different voters, which could be done anyway if all the people who keep complaining about RfA and complaining about admin abuse participated in RfA. Note that most of these people with high edit counts aren't doing it to pass RfA, they are dealing with mucky areas that require a lot of edits and they have fun dealing with those areas. Here is a suggestion for a better RfA: Have a discussion period for a week prior to the "voting" period. This would at least discourage people from voting and then never coming back; if they have to come back to vote, or if there is an active discussion, they won't base decisions on initial superficial impressions. —Centrx?talk • 03:35, 11 January 2007 (UTC)
I think that some current admins are quite unqualified to do the job that they have been set. So the current system is inadequate, we *do* get too many wheel wars, even now, and I have some suspicions about admin sockpuppets and meatpuppets as it stands. In short, I think the current system is broken already. Practically anything would be an improvement. Kim Bruning 21:07, 10 January 2007 (UTC)
Inadequacies of the current system does not mean it should be replaced by a system with worse inadequacies. —Centrx?talk • 00:44, 11 January 2007 (UTC)
I postulate that that would be extremely difficult to do anyway ;-) Kim Bruning 02:29, 11 January 2007 (UTC)

Consider edit wars on a Wikipedia where almost everyone has the admin bit. Edit wars often involve fairly experienced users - certainly in the top 90% of active users, so we'd have admins edit warring on a regular basis. How would we stop them? We can't protect the page, because admins can edit protected pages. We can block them, because admins can unblock themselves (we can forbid unblocking yourself, as we do now, but that won't stop it happening - at the moment, admins are only blocked under very unusual circumstances, and there is a lot of discussion on AN/I, so admins know it's best to let that discussion take its course. We can't have that discussion for every single block.). We would have to go to a steward to desysop people every time there was an edit war. It's untenable. So, if we want to give the mop out as much as you say we should, we have to introduce a new level admin and give some powers to all admins and keep some back for "superadmins" only. Your idea then becomes the "Trusted User" proposal which has been rejected at least 100 times. --Tango 20:41, 10 January 2007 (UTC)

Actually ArbCom is tremendously more willing to desysop for misuse of admin tools (like unblocking oneself, or deliberate wheel warring over protection) than they are for POV or personal attacks and all the stuff any editor can engage in. An admin gets the same benefit of AGF on edit warring as any editor, but very little AGF on misuse of admin tools. And this is as it should be. NoSeptember 21:08, 10 January 2007 (UTC)
All you would need is one admin sockpuppet to block your enemies and, depending on how relaxed the proposal is, a person could have a dozen admin sockpuppets easily. We already have problems with people using sockpuppets just for plain edit warring to get around 3RR; that would then happen for admin actions and in combination with admin actions. —Centrx?talk • 00:51, 11 January 2007 (UTC)
I think there should be no further relaxation of our standards. It's already gone far enough! People have become lazy, so that RFA has become both slower AND they even select worse candidates. it's not either or, we need to improve both. Kim Bruning 02:29, 11 January 2007 (UTC)

When there is a sudden lack of admins available to do needed tasks, I think the floodgates of promotions will open again. Perhaps our bots and scripts are becoming so effective at doing things we used to have to do manually that we can get away with a much smaller pool of admins (relatively speaking) than was necessary in the past. This has made it easy for many to oppose anyone who they even remotely are uncertain about giving the admin tools to. A lot of the drudge work associated with many admin tasks have disappeared. Instead of using a mop, we just press a button and the MopBot does it all for us - or so it sometimes seems ;). The problem is that we can have a big backlog in one area that has not yet gotten a dedicated bot to address it, while other former backlogged areas may now be running smoothly. So, we still need more admins. NoSeptember 20:49, 10 January 2007 (UTC)

I don't think we're going to get any "floodgates of promotions", and I am already worried about the quality of people currently passing through RFA. (Specifically, there's no longer any decent influx of admins I can shanghai to do cool stuff looking after wikipedia ;-). One shudders to think of what would happen if we were to open the floodgates. Kim Bruning 21:11, 10 January 2007 (UTC)
If you think there are few new quality people, then the problem isn't RfA, it's finding quality people to join Wikipedia. Wikipedia started with a heavy usage by tech savvy people who can do "cool stuff", now it is becoming much more reflective of society as a whole - filled with "normal" people who can't get the 12:00 to stop blinking on their VCR (and who are still using VCRs). NoSeptember 21:18, 10 January 2007 (UTC)
Nah, still plenty of quality people joining, except you won't find many of them on RFA (I've gotten lucky one or two times in the past year or so, but that's not much). Kim Bruning 21:31, 10 January 2007 (UTC)
And the reason people don't particuarly want to volunteer for RfA is it has become a ritual, hyporcritical savage event where no self respecting user really wants to spend 7 days suffering the very worst of behaviour seen on Wikipedia. --Kind Regards - Heligoland (Talk) (Contribs) 21:38, 10 January

2007 (UTC)

You are right, and I have neglected to mention that aspect. Could you elaborate on that, perhaps with examples? Kim Bruning 22:39, 10 January 2007 (UTC)

[edit] subheading 1

RfA is only as good as the people participating in it. If people make bad decisions then we promote bad admins. I notice Kim Bruning hasn't commented in a single RfA in nearly 5 months. Perhaps you're misdirecting your energy... we've talked about RfA being broken on this talk page ad nauseum but the novels we write here probably have very little actual impact on RfA participation, which is what matters. --W.marsh 21:21, 10 January 2007 (UTC)

Due to edit-count inflation caused by others, I am no longer able to apply my standards to RFA. I am simply incapable of making what I believe to be any sort of meaningful contribution, and have not seen any improvement in the last 5 months. Why do you think I'm complaining? Kim Bruning 22:43, 10 January 2007 (UTC)
I'm pretty sure your voice would count as much as any of "theirs". That's the whole thing, RFA isn't really that complicated, it's really not much more than the sum of the parts, the b'crat can only close it as well as the people who participated allow him to. Make a good contribution to individual RfAs and it's a step towards fixing what you believe is wrong with them. Stuff opined about here on the talk page, the track record shows, doesn't really have a very good chance of changing the underlying standards of RfA. --W.marsh 22:53, 10 January 2007 (UTC)
No, that won't work. I am being (accidentally) sabotaged by those who have edit-count criteria. Candidates now make large numbers of low quality edits in random areas before coming to RFA. This makes reviewing their entire edit history by hand more or less impractical, because of the low s/n ratio. I literally cannot apply my standards in a way that is still practical. Kim Bruning 00:44, 11 January 2007 (UTC)
Well um, sorry you feel that way, but RFA is more about the people involved than any specific process. The same people will be involved (and not involved) no matter how we arrange the deck chairs, and the same problems will be there... you seem to have a problem more with the people of Wikipedia today than the process, which hasn't changed dramatically in years. They only have their way because they participate in RFA a lot... despite your assertions no one is really stopping you from participating and countering them. --W.marsh 00:48, 11 January 2007 (UTC)
I just told you I cannot. I am unable to gather the information that I deem nescesary to make a decision. The following phrase has been overused in a perjorative sense, but I mean it neutrally: could you explain what part of that you don't understand? Kim Bruning 02:41, 11 January 2007 (UTC)
I waited a year, and about 2300 careful edits (none of which were done using AWB, and only maybe 200-300 of which were vandalism reversion) before entering an RfA. I make heavy use of the preview button, don't often need to correct myself (except when testing templates or monobook.js), and have a low edits/page ratio (2.23 or so then). That would be quite a few edits for you to go through, and I was still opposed by some for not enough experience. I don't think it's "edit-countitis" that is the problem, nor is it ever going to be feasible to check every edit of admin candidates again; it's just the nature of a growing community. -- Renesis (talk) 01:00, 11 January 2007 (UTC)
But you did pass, and there will always be dissenters anyway. —Centrx?talk • 02:34, 11 January 2007 (UTC)
Then you are a positive example, though it is unfortunate that some still chose to oppose you based on fairly random criteria. Kim Bruning 02:41, 11 January 2007 (UTC)
Your absolutly correct, and as I seem to keep saying, it's so easy to game the system here, all to the detriment of Wikipedia, and all encouraged by some of those who reguarly !vote on RfAs. The mentality is such that if you've got 300 good quality, well thought of Wikispace entries which show a great attention to detail and fantastic ability to reason over the deletion of something or the promotion of a admin candidate, you'll almost certainly be Opposed due to lack of experience. If you make 3000 edits to the Wikipedia namespace where you simply say "Delete per nom" or "Support per nom" with reasoning and no rationale, then you'll be supported due to a high number of Wikispace edits even if they show absolutly no knowledge of policy whatsoever. I think, therefore, that any RfAs closing with support ranging from 65-85% should be looked at very carefully by the closing Bureaucrat and checked against a set of criteria, like a driving test, stuff like "User shows knowledge of Deletion policy in edits checked" and "User correctly warns vandals after reverting". A random sampling of 10 edits from each name space could be generated by a bot and checked over by the closing bureaucrat. If the candidate meets a certain score on this test, they pass the RfA, if not, they are unsuccessful and need to resubmit there nomination no sooner than 28 days after the closing date of the first RfA. I'd like to think the benefit of this type of system is that it forces candidates to look at and refer to policy, probably out of fear of making an error and being marked down, but hopefully out of interest in knowing policy and not screwing up. --Kind Regards - Heligoland (Talk) (Contribs) 02:05, 11 January 2007 (UTC)
The procedure for my own RFA was perhaps more strict: every single editor checked all of my edits, or some large sample. (IIRC There was a single edit counter, but people mostly treated him as a troll). The bureaucrat had absolute discretion. Not all changes are an improvement. Kim Bruning 02:41, 11 January 2007 (UTC)

One thing we haven't even addressed because we've never had a problem with it but most certainly would if RfA standards were signficantly relaxed, is MediaWiki space vandalism. Think: genitalia image on every page. -- Renesis (talk) 21:55, 10 January 2007 (UTC)

  • I find it hard to believe that someone who could get even 50% support at RfA would ever be a blatant vandal... it's just incompatible with people who contribute unselfishly for months or years. --W.marsh 22:03, 10 January 2007 (UTC)
    We aren't talking about percentages on standard RfA's... some of the suggested remedies here are far less restrictive than "50%" support and months (plural) of experience. -- Renesis (talk) 22:07, 10 January 2007 (UTC)
How about actually requiring the review of every single edit made by a candidate. Would that make you happy? Kim Bruning 22:43, 10 January 2007 (UTC)
Not if I was the one doing the reviewing! :) Are you actually proposing that? That's just creating busy work, and it's still not a system that couldn't be gamed. -- Renesis (talk) 22:58, 10 January 2007 (UTC)
Not just seriously proposing, I actually used to do this! :-) The thing is, people used to not have numeric requirements, so I could just skip through 750-1250 edits at a leisurely pace (that's only about 2 big pages of text, mind you), and quickly pick out some interesting ones.
With the current quick quick get some more edits in this or that arbitrary category before I get opposed mentality, there are huge numbers of crap edits which I need to dig through to maybe find one or two gems.:-P
In this way, the numeric requirements people are actually sabotaging those who would like to do a more thorough review.
Kim Bruning 00:36, 11 January 2007 (UTC)
I do check contributions, but I don't need to check every single one. I only check those with interesting (or no) edit summaries for the past 2/3 months (depending on how active they are). This reduces the amount of checking I need to do but also, if they have made a bad edit more than 2 months and a few thousand edits ago they will usually have learnt better. Usually the last 2000 edits is all I check and it doesn't take more than 15 minutes to look through 4 pages (of 500) with popups. If you don't think there are enough admins, find someone who would be suitable and nominate them. Encourage others to nominate more people aswell, and I don't just mean nominate someone you know, go looking for someone. James086Talk | Contribs 03:02, 11 January 2007 (UTC)
Once upon a time I actually did that too! ... But there are several reasons I've stopped doing that too (see above). Kim Bruning 03:09, 11 January 2007 (UTC)
This sort of thing is why I created my diff-generator (bundled with my edit counter). I agree that edit counts are a bad way to make decisions on RfA (unless the count is really low like 250); normally, I check the pages that a candidate has been editing and what they've done to them. (My interest is mostly in policy-space; I don't think activity in the mainspace has much bearing on RfA unless it shows a misunderstanding of policy, but people might disagree with me on this.) Checking every edit made by a user is probably impractical nowadays; although artificially inflated editcounts are normally obvious, there are legitimate reasons to generate a large number of edits (for instance, RC patrol or semi-automated AWB editing). I definitely don't agree with the 'generate edits to meet some edit-count-per-namespace' concept; I avoid this by looking at what the candidate has edited and comparing it with the answer to question 1.
Unfortunately, one of the main problems of RfA at the moment is the pile-on supports (most RfA supports give no reason at all...). If a candidate isn't suitable, what's needed is either some killer diffs that show misunderstanding of policy, or else some incontrovertible statistics (preferably generated automatically) that show that a user has too much inexperience in a certain area; then people will start opposing per-you. If this sort of research were more common and people actually checked their admin candidates before voting, then it would help the system. (By the way, I tend to analyse candidates in detail, and then still be unable to make up my mind and not vote, except in the case of obviously unsuitable candidates. I would prefer a WP:DFA variant to RFA, to avoid the need to 'vote' at all, but that's been rejected more than once in the past and is anyway slightly off-topic to this conversation.) --ais523 10:12, 11 January 2007 (UTC)
"Pile-on supports" exist to help combat the people who oppose over trivial nonsense (not enough mediawiki talk and portalspace edits) or personal grudges (he called my garage band non-notable in 2003, the dick) and are a reflection of the adminship is no big deal mindset: if you don't see anything fatal that he's done, why not support? You can always change your vote later. --tjstrf talk 10:27, 11 January 2007 (UTC)
If people are voting support just to cancel out oppose votes, this is really a sign that WP:RFA is in trouble (I had wondered about this, but wasn't sure). Not that it isn't necessarily a good idea, it's just that in other non-vote discussions like AfD the spurious oppose !votes (and spurious supports, for that matter) would just be ignored by the closing admin (or 'crat, in RfA's case). (I use '!vote' in most contexts, but 'vote' in the context of RfA; one of the main problems is that it is a vote, rather than a consensus-building exercise. Unfortunately, it seems unlikely that there will be a reform proposal that enough people will support and that would actually work.) --ais523 10:33, 11 January 2007 (UTC)
Here's my two cents: I know it's already been said, but we do need to somehow scrap the voting. We have:
  • people who vote against admins if they don't need the tools vs. people who vote against admins because they only need the tools (e.g., Wikipedia:Requests_for_adminship/BigDT).
  • users who vote against candidates because they vandal-fight, and users who vote against them if they don't fight enough.
  • users who vote against candidates on the basis of mistagging (only once or twice) articles with {{db}}, as if current admins are perfect in the area (keep in mind, I'm an inclusionist).
  • users who vote against candidates without sufficient time on the project but with high edit counts (e.g., Wikipedia:Requests for adminship/MER-C, 4 months, 18000 edits) but for admins with <4000 edits but several years - Wikipedia:Requests for adminship/Ben W Bell).
  • And, the worst: users voting against candidates who they have grudges against (see Wikipedia:Requests_for_adminship/BostonMA, with a suspiciously high amount of opposes from people who disagree with his politics, though certainly not all votes).
As a result, the only kind of admin who gets approved is one who
  1. refuses to get involved in any disputes before becoming an admin, whether policy or article disputes (if even a small community of people holds a grudge, it can easily paralyze the RfA). But this is exactly the kind of admin we need.
  2. one with pretty candidate submissions
  3. users who are experienced but not too experienced, because users with higher edit counts are more prone to have made a damning mistake at some point.
In all, there's a serious need to either have a bureacrat treat it like we do all other issues (Template:Not a ballot), or cut down on the 80% figure that is just making it too hard for well qualified users to get the bit. I realize these ideas will have some resistance, but I really hate to have seen so many good candidates passed up on personal grudges and spurrious grounds. Patstuarttalk|edits 19:49, 11 January 2007 (UTC)
I have said this before, and got shot down in flames, but I will say it again, and take the consequences. In commenting on an RfA, an editor should in my view look at the candidates edit summary, edit count, answers to questions and nothing else at all. If !votes, either opposing or supporting, were not visible to subsequent !voters, we would achieve this, and any pile-on that occurred would be on the basis of the merit or otherwise of the candidate, not on the basis of pre-existing comments. Given that RfA is not a vote, it in my view should not be discussion forum either. Editors give a personal and un-influenced opinion, and a bureaucrat determines consensus or otherwise. Tell me why not?--Anthony.bradbury 00:43, 14 January 2007 (UTC)

[edit] Contradictions

(Since this is becoming impossible to track, I'll start a new subsection) Kim, you can't advocate both having 90% of active users as admins, and making standards tighter. It doesn't work both ways. You say that improving the process will increase throughput and increase the quality of applicants, and that those two go hand in hand. These two thoughts imply that you think that some 15000-25000 of our active users are more qualified for adminship than many of our admins. I just do not see how this can be true, plain and simple. You are building an impossible scenario here. Where are you going with this? -- Renesis (talk) 03:07, 11 January 2007 (UTC)

Well, we need to both improve throughput, and improve admin quality.
I don't think either is unattainable.
Since enculturation is such a major problem for wikipedia, we should take more time to actually teach skills to new users, especially those who we're nominating for admin. That's step one. This will ensure that more people will meet our quality standards, and they will do so more quickly. Learning to be an admin is not really hard.
More attention should be paid to direct review of edits, as opposed to edit counting. This helps in many different ways, but most importantly it is more strict. Since the number of edits required for this to work is actually lower than for pure edit counting, people can still become admins sooner.
As is at the very least indicated by the above, methods that are both quicker and deliver higher quality certainly do exist. Perhaps you agree or disagree with either of these, but at any rate, perhaps you can now think of further methods with similar properties? Kim Bruning 03:18, 11 January 2007 (UTC)
  • There are two issues here. One is about how RFA works, and I found that the best way of dealing with it is simply to find more good candidates and nominate them. The second is about the admin/non-admin divide, which is mostly a result of (1) people misunderstanding Unction's Law ("If enough people act independently towards the same goal, the end result is indistinguishable from a conspiracy."), and (2) a wrong direction fallacy, in that it's not adminship is not the cause of greater trust from the community, but the effect of that trust. The best way of dealing with this is education; but educating users is Not Easy. >Radiant< 10:50, 11 January 2007 (UTC)

[edit] New proposal / new criterion

I know that criteria for RfAs have been discussed many times and in details indeed sometimes. However, i've been thinking lately about a new criterion. I've been noticing that most of the nominees have never ever discussed issues at WP:ANI. I just find it odd and weird that a would-be admin had never been seen at ANI. I believe participation at the noticeboard is essential for a would-be admin in order to have a better idea about how stuff works. OTH, It would be easier for voters to have an idea about the nominee reactions and suggestions to solving problems prior to their nominations. Please discuss. -- Szvest - Wiki me up ® 17:23, 10 January 2007 (UTC)

Strangely enough, I just opposed someone due to not enough participation at the admin's noticeboards (along with other obvious problems such as having less than 200 edits). I can imagine some Q1-answers which wouldn't require it, but most would. Of course, we don't want the noticeboards cluttered with people merely trying to get their AN-edit-count up... --ais523 17:44, 10 January 2007 (UTC)
Please don't. People don't need to have contributed to WP:ANI to be aware of its existence, and we can't ask admin candidates to know every page they're "allowed" to work with later. For example, we give admins the protect button, so do you think we should require edits to WP:RFPP? Kusma (討論) 17:52, 10 January 2007 (UTC)
I would expect edits to RFPP or at least entries in CAT:PER if an admin said 'I want adminship so I can protect pages'. If there aren't any, it means that person didn't really need to be able to protect things after all. (The exception would be if they wanted protection purely to clear out RFPP backlogs or to salt deleted pages; the first would require knowledge of policy but not an RFPP report, and the second doesn't use RFPP at all.) --ais523 18:00, 10 January 2007 (UTC)
I've never been a part of the formal protection process. I don't know if I have a single edit to RFPP. What I do know is that I already understood everything about protection long before I became an admin, and that I was already involved in it. How? By using lines of communication much faster than RFPP — yes, that means IRC. I never really did any anti-vandalism stuff on-wiki; I found IRC early enough, and that was definitely the best and fastest way to handle the problems. --Cyde Weys 18:15, 10 January 2007 (UTC)
The thing is that ANI discussions include many protection/unprotection issues (mainly conflicts between admins) as well as other admin processes. Adminship is not limited to singular tasks anyway. But at least ANI can give a better grasp of wikipedia daily issues to would-be admins. We do need more and more admins but we need quality admins who know how to deal w/ other admins at least. -- Szvest - Wiki me up ® 18:08, 10 January 2007 (UTC)
Right, but whether somebody has made edits to WP:ANI neither proves nor disproves their ability to work with other admins. Kusma (討論) 18:14, 10 January 2007 (UTC)

Also, I find this terribly unfair, as the mere naming of the admin noticeboard gives the impression that it isn't for "common" users to comment at, and to this day, most people only get active on it after their successful RFAs. --Cyde Weys 18:20, 10 January 2007 (UTC)

Frankly I find this suggestion and the responses to it profoundly depressing. Many of the non-admins I have seen participating on WP:ANI have been border-line trolls using a heaven-sent opportunity to wind-up an admin who might have been getting on their case. Is this the kind of person you would want wielding the mop? As for protection issues, my main encounter with this has been to protect high-risk templates, and to be able to to fix protected templates when they break. What kind of discussion would you want to see in connection with that? Ofttimes there simply is not time to thrash something out in long discussion: you simply have to be bold and do it. Requiring participation in long-drawn-out debates over the minutiae of policy-wonking will not get you good admins: it'll get you people who are good at long-drawn-out argumentation. The only genuine test of whether you're good enough is to get out there and do it and see if you can pick yourself up when you inevitably fall over. Unfortunately there are also people who will watch you fall over and then use that as an excuse to remove your mop, because that is how they like to exercise their "power". One looks at the best admins who have been around the longest, and wonders whether they would manage to jump through all the hoops that have been installed since. Quite frankly, it's a surprise anyone consents to be subjected to the kind of abuse one sees on the average RFA, and one wonders what kind of admins we are hoping to gain by it. In answer to the original question, this kind of instruction creep is what is destroying the adminning process: we need something which isn't some kind of "tick the right boxes for me and I'll vote for you" process. HTH HAND —Phil | Talk 18:25, 10 January 2007 (UTC)

  • If we don't have any fixed objective standards, including a mandatory minimum number of edits, I can't see making this a criterion. In fact, I don't think it would be a useful criterion. It'll just encourage editing ANI for the sake of editing it. Furthermore, I don't think contributing at ANI makes you any better or worse a candidate. A candidate's overall editing record should speak for itself. I am sure I'm not the only person who Anecdotally, I read ANI frequently enough, but for the life of me I can't recall ever editing there (I'm sure my contrib history will confirm that, if I took the time). I'm sure I'm not the only one. Agent 86 18:36, 10 January 2007 (UTC)
    • Okay, I just proved myself a liar. I found one contribution that I made to ANI. However, looking at how huge ANI is and the massive volume of archives for it, I don't think we want to encourage any unnecessary postings on that page. Agent 86 23:12, 10 January 2007 (UTC)

Another bad bit of criteria creep I'd like to see addressed soon is this ridiculous requirement that the person must have written a featured article. I still don't see how that is at all related to typical administrative tasks. --Cyde Weys 18:41, 10 January 2007 (UTC)

I thought that had gone out of fashion last year. A current popular and not very relevant criterion is "no experience with images". Kusma (討論) 18:49, 10 January 2007 (UTC)
Nope, it's still going on now. Just look at BigDT's RFA. It seems to be the same people doing it, too. --Cyde Weys 18:56, 10 January 2007 (UTC)
  • This is rather disturbing. First, Cyde is grossly misrepresenting things. Most of the demurrals on BigDT that I saw were that he simply didn't do much with articles. Given that most of Wikipedia is articles and that Wikipedia itself is known to the universe as a collection of articles, rather than images, templates, boxes, and troll wars, it's not shocking that people might want administrators to have experience there. The "one FA" is a straw man. Second, I'm seeing people seeming to say that they don't want non-admins at AN/I. That simply can't be correct. Unlike Phil, most of the non-admins I see at AN/I are pretty helpful folks voicing opinions about matters that concern the site. I suppose it depends on where you look, but all users interested, either through violation or desire to improve, policy ought to be at WP:AN. AN/I is more for emergent situations, and it's the reason we don't need IRC channels for administrators, but the noticeboard itself should be a regular stop for everyone interested in policy and operation of the project. Geogre 19:12, 10 January 2007 (UTC)
    • By the way, I should say that I don't much care for this proposal as a criterion. I think anyone considering an RFA should get active on AN and possibly AN/I, if appropriate, but it's meaningless as a criterion, because going "me too!" would make a person active there but not actually contributive. Also, I think it sets us up for one of our worst biases in administrators: making vandal fighting a silent prerequisite. It's true that we all go on to the rank because we want to do some defense, but we also go on to the rank to do some cleaning, some moving, and other things that don't involve fretting all day about whether Willy pulls another fast one or not. When we let ourselves adopt the "vandal fighters only" model, we end up with "vandal hunters," and that leads us to abuse excused by the "dangers" of "bad people" and "idiots." It also leads us to administrators who have so much time spent in the dreck that they get convinced that all the users are vandals in hiding. "All is yellow to the jaundiced eye," as Pope said. Geogre 19:24, 10 January 2007 (UTC)

One of the optional questions on my RfA was about my participation on WP:AN and WP:ANI, but I have rarely seen the question asked since then. NoSeptember 21:36, 10 January 2007 (UTC)

Just a thought but it seems to be easier to become a steward than it is to become an admin. m:Stewards/elections_2006-2 --Kind Regards - Heligoland (Talk) (Contribs) 23:33, 10 January 2007 (UTC)

Um... 12 promotions over a year? Don't we get at least 20ish? -Amarkov blahedits 23:36, 10 January 2007 (UTC)
75% success rate, but the people who actually ran were probably a bit more self-selective than the 120 edit self-noms we get here. However, the absolute standards for stewardship application are lower than those we regularly enforce at RfA. It's probably easier to become a steward than a bureacrat, that's for certain. --tjstrf talk 00:46, 11 January 2007 (UTC)
Doesn't being elected as a steward require the ability to communicate in at least two languages? That may be natural to most of the world rich enough for basic education, but it eliminates most Americans and Britons (except the Welsh). --Sam Blanning(talk) 03:03, 11 January 2007 (UTC)
Yay for the Welsh. Diolch yn fawr Samuel ;) Proto::► 18:54, 11 January 2007 (UTC)
Not that I can find on the meta:Steward policies page. It would of course be helpful. --tjstrf talk 23:19, 11 January 2007 (UTC)
Don't get me wrong, I have no idea what that means. I just like Wales. --05:06, 12 January 2007 (UTC)
  • We don't need arbitrary criteria for adminship. For instance, there's one user at the moment who opposes any candidate who has 'no experience with images', but that is arbitrary as most admins don't ever do anything with images. >Radiant< 10:51, 11 January 2007 (UTC)
    There are some, frankly, ludicrous reasons for opposing. I am going to start keeping track of those who oppose for unfair or ludicrous reasons and make sure to bear them in mind if and when they go for RFA themselves (unfortunately, some of the stupidest reasons for opposes have come from admins recently). Proto::► 18:54, 11 January 2007 (UTC)
    At least we haven't gotten opposes due to lack of MediaWiki talk edits. Yet. -Amarkov blahedits 04:28, 12 January 2007 (UTC)
    Hey, if we had more people contributing to mediawiki talk then maybe the messages wouldn't have egregariously bad grammar in them.[5] --tjstrf talk 06:26, 12 January 2007 (UTC)

[edit] The Big Problem™

The Big Problem with the adminship process is that people are allowed – even encouraged – to come up with their own ridiculous criteria that they think an admin must fulfill, and then they vote oppose on everyone who doesn't fulfill them. That way the role of adminship is given only to those people who fit all of those ridiculous criteria – a selection process that comes no-where close to selecting the people actually needed for the role. — Timwi 18:11, 10 January 2007 (UTC)

Agreed. Here are my own requirements, just for the record. Not only do I think this abomination should remain inactive, I think it should be deleted. --Cyde Weys 18:18, 10 January 2007 (UTC)

Cyde, please restrain yourself! The abomination is a piece of Wikipedia history. Future generations will look on it in awe. :-) Carcharoth 23:43, 10 January 2007 (UTC)
That page and I just didn't get along. I originally tried to transclude my criteria to it, but got reverted because someone accused me of "not being serious" (because all of those criteria were really helpful, amirite?). Then I later added "Every candidate is a unique person, so I perform a unique analysis on each one", which is frankly the only thing on that entire page that makes sense. I guess I'm glad it's all been shut down by now, but I think it deserves a worse fate. --Cyde Weys 19:18, 12 January 2007 (UTC)
I didn't realize you were joking until requirement 8. Shows just how ludicrous some of the criteria for voting are/have been. --tjstrf talk 18:31, 10 January 2007 (UTC)
Joking? Who said anything about joking? CSCWEM is still pissed about not passing his first RFA because he failed his SCUBA certification test. --Cyde Weys 18:40, 10 January 2007 (UTC)
For another moment of absurdity, the easiest requirement to pass there is 17. The only unique criteria for becoming a steward that anyone with a chance of passing RfA here wouldn't necessarily have is that they give proof they are 18. (meta:Steward policies#Steward elections) --tjstrf talk 19:31, 10 January 2007 (UTC)

i don't even understand seven out of the twenty-eight.--Anthony.bradbury 23:14, 10 January 2007 (UTC)

At FAC, we divide between "actionable" and "inactionable" objections. Perhaps we should keep in mind, when we answer questions or pose them, the difference between a criterion that has to do with being an administrator specifically and what does not. For most people, the criteria come down to "time, experience," and "like me on an important issue." The various ways that time and experience are counted are less or more elegant and precise, but the "like me" is where death and terror lay. I remember when "inclusionists" wouldn't vote for "deletionists," when "fair use" wouldn't vote for "free images only," etc. So long as these things remain anchored to "will you use administrator's powers to enforce your will on this issue or not," they should be ok, but they rarely are so anchored. Geogre 19:20, 10 January 2007 (UTC)

After reading half the RfA archives, I've seen arguments like this over and over. The real problem is, RfA is a naturally subjective process (and rightfully needs to remain so) and people will be people and !vote for whatever they want to, whether you introduce hard requirements or not, whether you disallow certain types of objections or not, whether you try to change the process to circumvent human nature or not. The best we can do, I think, is to ask the bureaucrats to maintain a strict interpretation of the criteria listed in WP:GRFA do determine if the community has rightfully decided whether the person would make a good admin or not. This is partly why I will be running for bureaucrat, because I believe I can bridge the gap somewhat between people who object to the current process and others. Of course, if you think another process would be better, go ahead and propose it, and try it out; I'm always willing to listen. :-) Grandmasterka 20:29, 10 January 2007 (UTC)
People have tried to fix this. For an example of how impossible it is to actually come up with what the criteria should be, look Wikipedia:Qualified adminship voting. It seems reasonable, until you see my absurd mockup scenario on the talk page. The problem with any system to make sure only rational objections are put out is that it necessarily implies that some people (i.e. the policy creators) are more intelligent than others (i.e. the people who comment on RfAs). -Amarkov blahedits 23:05, 10 January 2007 (UTC)
Don't forget another type of voting prejudice: when "article writers" won't vote for "non-article writers". Even though the "non-article writers" are actually involved in a lot more of the tasks that need admin rights anyway. --Cyde Weys 23:41, 10 January 2007 (UTC)
I haven't; being a non-article writer, it's not something I can just ignore. Of course, people are opposed for only being article writers too, but that's much more defensible; admin tasks simply do not involve article writing. -Amarkov blahedits 00:40, 11 January 2007 (UTC)
  • Indeed, silly criteria (such as editcountitis) have been a long-standing problem in RFA, although it's less bad now as it was three months ago. But there's no feasible way of preventing people from doing this, other than educating them better. Unless you want to make RFA based on committee rather than community (another can of worms), the best you can do is lead by example. >Radiant< 10:56, 11 January 2007 (UTC)

The problem with published standards is that someone might hold you to them. —Malber (talk • contribs) 12:53, 11 January 2007 (UTC)

  • Hm. Any chance of convincing people who write stuff like this that their bureaucratic viewpoint isn't particularly useful? >Radiant< 13:43, 11 January 2007 (UTC)
    Those criteria completely fail to take into account the fundamental difference between talk and MediaWiki talk edits. More seriously (although the first point is serious enough if you take such criteria literally), the criteria couldn't deal with the sort of crazy one-off situation that creates a massive oppose (just a look at WP:RFDA shows that desysop-worthy offences tend not to be consistent, and not-sysop-in-the-first-place offences are likely to be just as varied). --ais523 13:54, 11 January 2007 (UTC)
    (edit conflict) Now, notice that I actually do allow for a bit of flexibility. Not meeting say, one or two of my 'to pass' criteria does not necessarily indicate automatic failure. I don't purely go by having 10-20 criteria and if any one of those are not met, then I oppose. Instead, the flexibility bit (which I've thought about prior to writing my criteria, is to have plus points for good indicators that the candidate would make a good admin, and negative points for bad indicators). It's not clear to me what the problem is, I suppose it's the edit count/numbers thing that may raise concerns. The reason why I write criteria is for consistency, and before I had those, I don't think I was as consistent as I am now with voting. Insanephantom (my Editor Review) 13:59, 11 January 2007 (UTC)
    I created that quite recently, and I still have room to improve on it. Remember, it is only intended to give a general idea of how I vote. Perhaps removing the 'Interpreting the score' section would be best, as I do not always follow the result according to the score. Insanephantom (my Editor Review) 14:03, 11 January 2007 (UTC)
    • The point actually is that you base your votes not on whether or not you trust the candidate, but on a bureaucratic interpretation of arbitrary measurement. The fallacy in that is that you are trying to quantify the unquantifiable. >Radiant< 14:09, 11 January 2007 (UTC)
    Most of my criteria are constructed based upon how others vote after some observation (and possibly a bit of other peoples' criteria) and why they supported/opposed for reason X. The more of the criteria is fulfilled, the more likely I am to trust the candidate. I know that my criteria may not be the best way to judge a candidate, due to a lot of numbers coming into play there, but it is good enough to provide a sufficient outline of what I might do, and then go from there and check if there is any issues that may alter my trust of the candidate (such as the candidate being involved in an Edit war, or anything similar). Insanephantom (my Editor Review) 14:21, 11 January 2007 (UTC)

[edit] Accounts created just for RFA

What is the current policy on accounts created solely for RFA?

Dragons flight 20:55, 10 January 2007 (UTC)

Block as probable sockpuppets? --Deskana (For Great Justice!) 20:58, 10 January 2007 (UTC)
Mark as SPA's, I think. Yuser31415 21:03, 10 January 2007 (UTC)
That's why I ask. Sometimes such votes are simply marked, sometimes they are crossed out, sometimes outright removed, and whether or not the account is blocked (and for how long) is another wrinkle. For the record, I am not going do anything about these myself since they "voted" on the ProtectionBot RFA. Dragons flight 21:15, 10 January 2007 (UTC)
Then I will. :-) --Deskana (For Great Justice!) 21:17, 10 January 2007 (UTC)
I just explained to them that their "votes" won't be counted. I think it's important to AGF here, they could presumably have stumbled on RfA and contributed there first, and their involvement doesn't seem disruptive. --Deskana (request backup) 13:28, 11 January 2007 (UTC)

[edit] Somewhat distressing outcome in Wikipedia:Requests for adminship/ProtectionBot

I find myself a bit saddened at the outcome in this RfA. As I said in my initial support I wasn't convinced there was a need to run a bot through RfA this way if the BAG approved and if the owner was an admin, but given that we were doing it that way, we were.. I'm glad the developers implemented an improvement that made the bot less necessary but I'm saddened at the time and effort and writing spent on this process which in the end turned out to be "needless". I guess this is more of a comment and commiseration with Dragons flight than anything like a specific complaint, because I certainly don't want to discourage developers (and specifically, Werdna, who did this one) from making improvements, but... it seems a bit of a mismatch in processes (between development requests and how they are prioritisied and decided and bot development)... is there anything that could be done differently if in future a need is identified? We're all volunteers here of course which makes it a bit harder. ++Lar: t/c 16:26, 11 January 2007 (UTC)

  • Less bureaucracy would help. I mean come on, there were people objecting because the bot hadn't signed his acceptance of the candidacy, that's just silly. >Radiant< 16:35, 11 January 2007 (UTC)
    • I'm sure it would be possible to write an RfA designed such that when Mathbot edited it to put the summary usage on, it would appear to sign it... But bots really shouldn't need to accept their own RfAs. Either that, or when going through WP:BRFA they'll have to add 'accepting an RfA' in the list of things they were requesting approval for... --ais523 16:40, 11 January 2007 (UTC)
    • I HOPE those objects for unacceptance were supposed to be funny, but I'll reiterate, I'd prefer that community consensus form around the BAG getting the call on this one. I'm not sure I favour just running rouge the way Curps does/did... the community should have SOME say but this process was... a bit much. ++Lar: t/c 16:50, 11 January 2007 (UTC)

I'd just like to echo the comment that some people have been considerably less than helpful during this whole process. -- Steel 16:44, 11 January 2007 (UTC)

  • I think the message of all the bot RfAs is "Admins should run adminbots from their own accounts just like Curps did. Putting a bot account on RFA wastes so much of everybody's time that it qualifies as disruption". Kusma (討論) 16:47, 11 January 2007 (UTC)
  • It was one of the first, so something had to be done. The same reasons bots are in separate accounts is the same reason an admin bot should be. Segregation and control. There is no need to block an admin because their bot needs to be. - Taxman Talk 16:53, 11 January 2007 (UTC)
  • I was being sarcastic. I support adminbots, and would find them preferable to an admin running them on his own account for precisely these reasons. Anyway, please let us try something else than RfA for the next adminbot proposal. Kusma (討論) 16:58, 11 January 2007 (UTC)
  • A lot of people were less than helpful. But part of having a free system is accepting that some people will take advantage of it and not be helpful. The comment that even though there was no established process for such a thing, that since the bot didn't go through my exact preferred process, I'm going to oppose was perhaps the worst of the lot. What made it worse, is the person should have known better. That things like that will happen in a free system doesn't mean we don't need to deal with it, and in fact we should work to minimize it, but in the end, the articles are all that matter. On that front we are moving ahead. Perhaps not fast enough, but faster than anyone would have expected 5 years ago. We need to minimize side efforts that do not go into improving articles to the greatest extent possible. - Taxman Talk 16:53, 11 January 2007 (UTC)

I'm of the view that the bot approval group really made a sham of it. There's a reason that a selected group of experts, rather than the community, exists to evaluate bot proposals. It's still not clear to me why they found themselves incapable of doing that task in this case. And if RfA is well-suited to evaluating this bot, why not just run the bot request system in the same way for all bots; perhaps a "requests for bot permissions" under the RfB section. Christopher Parham (talk) 18:48, 11 January 2007 (UTC)

That's a terrible thing to say. Nothing said they had the right to grant a bot admin rights, so they most certainly did not make a sham of it. - Taxman Talk 22:29, 11 January 2007 (UTC)
Of course they had the right -- their mandate is to evaluate whether it is safe and necessary to give pieces of code special technical access levels that improve their functioning. The bot approval process is a much more appropriate framework for decisions of this type than is RfA -- simply because the distinction between +bot and +sysop is not nearly as significant as that between "human being" and "machine." While the BAG might not be 100% comfortable with considering the sysop flag, RfA and the "election" process have no apparent qualifications to consider machines. Christopher Parham (talk) 23:02, 11 January 2007 (UTC)
Very simply, they didn't. The BAG is allowed to approve bot flags, not admin rights. This bot required adding admin rights. So they made no mistake. Now, if people want to set up a policy that allows the BAG and the Bcrats to make the call, that's fine, but it wasn't in place. In any case, we need to move forward, the reason I made a comment, was your saying the BAG made a sham is not helpful. - Taxman Talk 18:50, 12 January 2007 (UTC)

I've pretty well and good lost faith with the community's ability to handle adminbot issues. I was going to open source my adminbot (which I've been running for about ~6 months now) and run it through RFA so all of those deletion log entries could show up under Cydebot's account rather than mine, but frankly, I don't see a reason to bother now. The whole process is FUBAR. What's going to happen is that all of the adminbots (and don't kid yourselves, they are out there, you just don't know of them) are going to be kept running secretively under their owner's sysop accounts, which is far from optimal. There are a lot of people who don't seem to realize that running adminbots under separate accounts breeds more accountability and better results. I don't think RFA is capable of handling this issue; I think BAG is, and I think they should be given the ability to accept or deny adminbots. Because clearly, the average RFA voter doesn't know what in the hell they're talking about when it comes to these issues. Opposing because a bot "didn't accept its nomination"? personal attacks removed --Cyde Weys 20:40, 11 January 2007 (UTC)

Most of the complains (including my neutral opinion) was that the source should be open. Had it been so, it would have collected a few less opposing opinions. While some may have complained about a bot with admin status, that will always happen (as with the 1FA opinions). I feel sorry for Dragons flight, but he should have known he would have gotten this kind of feedback, especially after TawkerbotTorA. Should BAG grant adminship to bots when necessary? Ah, that is absolutely another topic of discussion. -- ReyBrujo 21:03, 11 January 2007 (UTC)
I think Dragons flight did sort of know that that sort of feedback happens at RfAs for adminbots, which is why he wanted to get his bot approved a different way, but BAG passed the buck and insisted on an RfA. Though to be fair, the RfA was going reasonably well. The developer patch (laudable though it was) did throw a large spanner in the works though... Carcharoth 21:17, 11 January 2007 (UTC)

If there was a lot less silliness within this site (particuarly this RFA as well) it would have been promoted. —Pilotguy (ptt) 21:14, 11 January 2007 (UTC)

Huh? It was withdrawn because a software patch made it redundant, or at least the RfA would have needed to be restarted. It wasn't withdrawn because it was failing. It was probably going to sneak through at just above 80%, but the closing 'crat would probably have discerned more support than that, based on the large number of principled 'open code' opposers. Carcharoth 21:17, 11 January 2007 (UTC)

I'm surprised none of the "it didn't accept" people have noticed that it didn't withdraw either... Carcharoth 21:18, 11 January 2007 (UTC)

The community's involvement in the decision should been restricted to establishing the existence of a problem. Upon determining that Robert's bot was a viable means of addressing this problem (and was operating properly), the BAG should have given the go-ahead. Instead, they shifted to burden to RfA as a means of deflecting criticism from the adminbot-fearing segment of the community. Rather than doing what they knew would benefit Wikipedia, the BAG forced Robert—who selflessly dedicated a great deal of time and effort to this endeavor—to stand before a mob of torch-wielding villagers and try in vain to convince them that his creation would do them no harm. (Note that I'm not charactering all of the RfA opponents in this manner, as some did raise valid concerns.)
I can almost understand the BAG's position, as this obviously was a controversial matter, and I would hate to be accused of circumventing the community's will. I cannot, however, understand why they insisted on conducting the discussion at RfA (thereby anthropomorphizing a non-sentient entity). A computer script cannot be an administrator, so this "request for adminship" was an absolute farce.
I (and others) attempted to suggest that the discussion be held on a different page (with links prominently placed at RfA and elsewhere), and we were continually met with non sequiturs along the lines of "the bot can't become an admin without an RfA" and "we can't bypass the community." Never before have I encountered such ridiculous bureaucracy on this site. —David Levy 21:19, 11 January 2007 (UTC)

BAG did not shift the burden. So far in the 5 year history of the project no account has gotten admin rights without previous RfA approval or papal dispensation. To avoid that would have led to unending outcries of cabalism and subverting the community will. Bureaucrats have not in the past been given the pass to say "well there is a consensus among this small group, so ok, I'll promote." Therefore a few of the Bureaucrats said we felt it would be better if it went through RfA. The BAG (excepting Essjay) doesn't have the bcrat rights to grant admin rights to the bot account anyway. Now I do agree the BAG, along with one or a group of bcrats should be able to make the call and pull the trigger, that has not been established to be an acceptable solution. I think it's now clearer that many people want that, but it certainly wasn't before. It's easy for you to sit here now and claim I told you so, but if you don't think there would have been outcry if we made the decision on our own, you haven't been paying attention. That would have made this look like pillow talk. In any case, the bot likely would have been approved. - Taxman Talk 22:29, 11 January 2007 (UTC)
I was describing how I believe this ideally should have been handled, not how it realistically could have been handled. As I said, I realize that the matter was controversial and that a dedicated discussion was a necessary evil. Again, my main objection is that this discussion was held at RfA instead of a logical venue. —David Levy 02:47, 12 January 2007 (UTC)
That may have been what you intended to do, but what came out was overly sharp criticism of people that did the best they could. You could have stated more directly and simply, "here's how to do this better next time". Instead comments like the one with "absolute farce" amount to "those that don't share my opinions don't have a brain." Bot's aren't human but because they are accounts they can have admin rights added. Therefore the bit that they aren't human isn't really germaine, since so far RfA is the place to get admin rights on an account. - Taxman Talk 01:15, 13 January 2007 (UTC)

The problem I see with this as the process is that in all of our discussions about what may be wrong with RfA, or at least with users' RfA criteria, is that the discussion always comes back to the idea that it's about community trust, and whether the community trusts the user with the tools.

If that's the goal of RfA, then what the heck is a bot doing here?

Community trust of an admin bot comes under a group of headings:

  • Trust of bots (in general) being admins
  • Trust of the person who runs the specific bot
  • Trust of the code of the specific bot (how it will perform; is it partially or fully automated; etc)

Due to there being three entirely separate entities, an RfA will rarely succeed. (It would be like nominating 3 people for RfA, and needing to pass all three for any of them to be admins.)

The interesting thing is that none of the three above involve trusting a person with admin powers (presuming the bot owner is already an admin, which I presume is a criterion of even proposing adminship to a bot).

Trust of the person who runs the specific bot - The community has already determined that they trust the admin as an admin, so by corollary, we should presume that the community should trust that admin to make admin decisions regardless if doing so through the admin account, or an associated bot account.

Trust the code of the specific bot - this seems to fall directly under the purview of WP:BRFA. Knowledgeable Wikipedians doing what the process is designed to do.

This leaves us with the question of whether the community trusts bots being admins. I think before this discussion goes any further, someone start a discussion about this. I think establishing community consensus about this would be a first step to resolving the rest. - jc37 22:44, 11 January 2007 (UTC)

Of course having a MediaWiki patch made this an exercise in futility, but some people above make it sound like this adminbot was going to fail RFA. I don't believe that was in any way a sure thing, and more likely that a few of the truly ridiculous opposes would have been voided and the bot passed.

But I continue to maintain that most adminbots suffer from a lack of ability to perceive when their account has been compromised, and a lack of need/code for all the tools they are being granted. The former can be somewhat countered by the Big Red Button approach, and the latter could be handled if the developers would implement granularity of rights.

A protection/unprotection-only bot would have sailed through even RFA, which is the community consensus place to grant administrative powers to an account. For those who would reply that the user already has admin rights, consider your reaction to an admin placing a sock up for RFA. -- nae'blis 22:54, 11 January 2007 (UTC)

"...consider your reaction to an admin placing a sock up for RFA." - I couldn't find a guideline/policy against it. And honestly, if the sock is noted as a sock (essentially if it follows the rules of WP:SOCK#Legitimate uses of multiple accounts), then I don't think really have any problem with it, especially in this era of WP:AWB. - jc37 23:07, 11 January 2007 (UTC)


I think the main controls I'd want over an adminbot (in addition to usual bot requirements) are a few very simple things, some human, some built into MediaWiki.
  1. On a human front, I'd want to have some committee or panel, who are trusted software and coding experts/specialists, who have reviewed the code, and can certify a 'clean' or safe version they're happy to see run -- and only that version can be run. (WP:BRFA?). If they're happy then I as a regular editor don't need to see the code if the owner wants closed source; if I want to check bot code then I go join the Bot Group. Since code revisions might be needed regularly or urgently in some foreseeable cases, I'd like to see a process where relatively simple code changes can be submitted for rapid approval of the change, perhaps on a "first person who gets the email checks it out" basis.
  2. I would want an exceptional degree of trust in the developer and user. More than we give a usual admin. If there is any question at all, then let them code the bot -- and let someone else who does have that exceptional degree of trust operate it. (Assumption: Most adminbots will be automated; if their actions were manually approved they wouldn't need a bot admin account.)
  3. Because an adminbot would be a formidable problem if it went wrong (wilfully, by hacking/mistake, or from unforeseen circumstances), I'd want "block on suspicion of misconduct" to be the rule, with suitable automatic standard template WP:AN/I posting built into the MediaWiki software if a user account tagged as an adminbot is blocked. (If all is okay we can sort that out at leisure.)
  4. Last, the main new risk introduced by adminbots is not "bad admin action". It's that being dependent on coding, and lacking administrators' judgement, an amok adminbot (again through coding error, malice, or whatever cause) could in principle cause a large number of seriously problematic administrator actions being pushed through the Wiki in a stream, perhaps even hundreds or thousands in a minute before someone hits the OFF button. I would want the mediaWiki software to have a built in throttle, that accounts tagged as an adminbot have an admin action limit of (say) 2 or 3 per second, and 100 in 10 minutes, or something, after which the bot is restricted from admin actions either for a given amount of time, or to a given rate, unless the restriction is manually lifted. A bit like flood control in IRC I guess. And I would want that in the MediaWiki software itself so that it can't be bypassed, with all tagged adminbots being required to be compliant (ie self limit their admin actions to the approved level).
I'm sure there's more that others would say, but until we have more experience of adminbot handling, those would be the main controls I'd like to see. Thoughts? FT2 (Talk | email) 23:08, 11 January 2007 (UTC)
Ah well, you're already hosed then, because someone already just put the particular functionality at hand directly into the mediawiki codebase instead. Nomic players 0, Wikipedia 1. Hurrah! :-) Kim Bruning 00:40, 12 January 2007 (UTC)
  • Since a lot of the circus was due to the lack of a clearly-defined process for handling adminbot requests, how should they be handled? Standardization here would be helpful to everyone. Titoxd(?!?) 06:19, 12 January 2007 (UTC)

Titoxd has it exactly right. I would have preferred that the BAG just approved it and bulled it out, to see if the consensus was there, but I can't fault them for not wanting to. Now, however, we have some data to work with, can we as a community consensus on a process for granting bots approvals for certain actions, including actions that require admin rights, that isn't exactly the standard RfA? That process when applied to bots, seems broken to me. ++Lar: t/c 13:46, 12 January 2007 (UTC)

[edit] But it did pass

  • (Resp to Cyde) Actually, despite the poorly-reasoned parts of the opposition, the protectionbot would still have passed RFA, in that it had about 82% support. So I'd say the FUBARness is a vocal minority and we can in fact obtain admin rights for bots here. >Radiant< 09:04, 12 January 2007 (UTC)
    And if Dragons flight had been willing to make it open source, it would have been more like 92%. Proto::► 13:37, 12 January 2007 (UTC)
  • I'm happy that it was going to pass, apparently, even without the need to make it open source. ++Lar: t/c 13:46, 12 January 2007 (UTC)

Hrmm, I wonder how a Cydebot RFA would go. However, I fear ludicrous oppose reasons along the lines of "But it's Cyde! He's just going to use it to game the system and delete everything he doesn't like!" --Cyde Weys 19:13, 12 January 2007 (UTC)

You're ignoring the supports you would get with the same reasoning. ;) James086Talk | Contribs 03:27, 13 January 2007 (UTC)
Quite. If you're not going to use it for that, I'm certainly not going to support. We don't need a flower delivery bot from you, Cyde. :) That's not your core competency. (I'm Lar and I approved this silliness) ++Lar: t/c 14:10, 13 January 2007 (UTC)

[edit] Vote tallies

Would it be possible to have the vote tallies of each RfA in the RfA's section header, rather than next to 'Voice your opinion' link where they are now? I.e. the section headers would look like ===[[Wikipedia:Requests for adminship/TSO1D|TSO1D]] (00/00/00)===. This would provide a nice summary of the way current RfAs are going in the table of contents at the top of the main RfA page. --Sam Blanning(talk) 23:40, 11 January 2007 (UTC)

That might not be a bad idea, also it would remind the voters to count the tally afterwards. Arjun 23:41, 11 January 2007 (UTC)
I think it's a good idea. Cbrown1023 00:38, 12 January 2007 (UTC)
Putting it in the section header would mean it would change every time an opinion was added, which would make it harder to use the page history to jump to sections. Seems overly complex; of course, I still think the tally is superfluous (and with section headers becoming more likely the larger an RFA gets, it means a dummy edit to boot). -- nae'blis 02:03, 12 January 2007 (UTC)

I think the tally was a bad idea to start with. Can we leave it off entirely? :-) (See also Articles For Deletion style. My goodness, I never thought I'd see the day where RFA could learn from AFD! ^^;; ) --Kim Bruning 00:28, 12 January 2007 (UTC)

  • I'm for keeping the tallies, but not for moving them as suggested. It makes no real improvement, and if people want a "snapshot", there's always the various analysis reports. (Oh, and the last discussion on this topic is here.)

If an RfA is not supposed to be an election, why do we even have "vote" tallies? —Malber (talk • contribs) 20:01, 12 January 2007 (UTC)

[edit] Secret voting and stuff

Just another of my wildly fanciful ideas, but how about a secret ballot, the candidate goes up for RfA and a link is left at the bottom of the RfA !vote here / !comment here which talks logged in users to a page where they can make their !vote and comment in secret, creating a !vote that is only visible to bureaucrats until after the closing time, at which point the results and all the comments made are published (either with or without the users details). Naturally, to ensure all !votes and comments are being recorded, there would still be history of a contribution in the users contributions page, it's just the comment wouldn't be visible to anybody else and no tally would be present.

I feel that it's becoming very sheepish on RfA now, if one or two well respected admins vote, it can guarantee or sink an RfA straight away and then there's little reason to vote in a manner appropriate to the candidate and it might show up any !voting cabals that could be operating.

It might also be an idea to increase the number of bureaucrats and have promotions agreed by a panel of 3 bureaucrats. !voting would, of course, be automatically closed after 7 days and questions could still be asked of the candidate, rather, it would just prevent "per whatever" type votes and bitterness developing during the RfA process where a candidate feels the need to respond to every !vote. I often wonder if anonymous promotion by the bureaucrats might help sneak a few more well qualified candidates into the ranks without fear of perennial complaints. Anyway, just a few suggestions and perhaps madcap ideas for you all to mull over. --Kind Regards - Heligoland (Talk) (Contribs) 02:20, 12 January 2007 (UTC)

How could we invent a secret page? --TeckWizTalkContribs@ 02:39, 12 January 2007 (UTC)
  • How do you propose to deal with the fact that !voters often leave comments that help shed more light on a candidate (good and bad) and provide information to support their !votes that are useful to others? After all, RfA is supposed to be a discussion. Agent 86 02:41, 12 January 2007 (UTC)

Hmm, yes, secrecy and "sneaking" is the solution to any problems RfA may possess. Opabinia regalis 02:43, 12 January 2007 (UTC)

(edit conflict) I would be completely against this-my recent RfA failed, but I found the comments left by the oppose !voters to in general be very helpful on areas I could improve in. If this were a secret process, I would just be left getting told that the nomination failed, and wondering why that was. Anyone who is going to take constructive criticism as an attack or hold a grudge for it is certainly not ready to be an admin. Seraphimblade 02:44, 12 January 2007 (UTC)
I wasn't quite clear, all the comments would be published after the end of the RfA as they are generally useful for the candidate. --Kind Regards - Heligoland (Talk) (Contribs) 02:51, 12 January 2007 (UTC)
Opabinia regalis - you don't think there's a mass of secrecy and sneaking going on behind the scenes at the moment. It happens all over IRC and I suspect Google Talk, AIM, Yahoo and MSN Messenger, E-Mail, Telephone and whatever have you. Giving people the opportunity of making comments without having to worry about how their cabal will behave when they find they've broken with protocol or they've brought up points they might be too frightened to make in public with their signature present for all to see might, just might help things. It probably won't, it'll probably be as crap as ever, but such is life. --Kind Regards - Heligoland (Talk) (Contribs) 02:55, 12 January 2007 (UTC)
So I can't read others' potentially constructive and useful comments because some people can't disagree with their friends in public? We need less tolerance of lame pseudo-cabalism, not more. Opabinia regalis 03:27, 12 January 2007 (UTC)
"I feel... if one or two well respected admins vote, it can guarantee or sink an RfA straight away". Well, if anyone is so lacking in self-confidence that they won't vote against a certain admin, that's their problem. If you see someone you respect vote a certain way, you shouldn't be thinking "I can't vote against him", you should be thinking "This guy has his head screwed on, I should look carefully at why he's voting this way"... and still be prepared to vote against him if, after looking, you still disagree.
If the votes of one or two admins are sinking or guaranteeing nominations, it's the fault of those who aren't aware that adminship is no big deal, and certainly doesn't give admins' opinions more weight. If this is really happening (and I don't think it is) it's not the fault of admins, and definitely not the fault of the process.
I think being so in awe of someone's opinions that you're afraid to voice your own is unhealthy in any circumstance. If Roger Ebert walked into my local and said that Underworld was a good film, I'd still tell him he was an idiot. --Sam Blanning(talk) 03:42, 12 January 2007 (UTC)
  • Special:Vote exists on SVN. That doesn't mean it would be a good idea. Oppose votes many times provide useful tips to someone who is undergoing RFA. I learned that in my RFB a long while ago. Creating a secret ballot short-circuits that forum for improvement. Titoxd(?!?) 04:15, 12 January 2007 (UTC)
    • Yes, this is a great idea! Let's make it so absolutely no discussion is possible! I mean, we can't have enough votecountitis, can we? Seriously, sorry for the sarcasm, but this is a terrible idea. -Amarkov blahedits 04:24, 12 January 2007 (UTC)

Let's keep wikipedia a consensusocracy, at least in theory. Making secret votes would make RFA a democratic process in fact; and our journey to the dark side would be complete. :-P

Democracy normally sucks least[1], but consensusocracy has been known to suck somewhat less yet in particular limited circumstances, which happen to be present when writing an encyclopedia.

The problem is that both democracy and consensus introduce large time overheads. (with consensus costing less time than democracy in normal cases on wikipedia where wiki-editing works smoothly)

A system of governance which could even just break even wrt time spent would likely earn the inventer a nobel prize. :-)

-- Kim Bruning 05:49, 12 January 200s (UTC) [1] "Democracy is the worst system, except for all the others" -- Mark Twain

I think that opposes from 2 respected admins can bring down an RfA because people trust their judgement, they just happen to be admins aswell. Would it be right if a "respected admin" found something that would sink the RfA (if anyone found it) but nobody else could see it? Then the candidate got through with the RfA process masking their true contributions. James086Talk | Contribs 09:08, 12 January 2007 (UTC)

You call it a "!vote" and a "secret ballot" in the same sentence - that's a complete contradiction. If it becomes a secret ballot it will be a vote, not a !vote (please excuse the double negative). It's impossible to determine concensus without discussion, and you can't discuss something if everyone is commenting in secret. This suggestion is to turn RfA into a vote, and I don't think that's ever going to achieve concensus. --Tango 15:51, 12 January 2007 (UTC)

[edit] While we're on about this...

While we're on about standards and optional questions, I must question users who publicly announce that "An optional question is not answered for 24 hours." is opposable. Whatever happened to the "optional" in "optional questions"? It's things like this that make "optional questions" mandatory. – Chacor 07:13, 12 January 2007 (UTC)

(copied from my talk page) I'm considering deleting the page altogether if it doesn't work. But it seems you're quick to jump in to criticism, which I find a bit overwhelming. I know there are a lot of problems with it, but I will either work on it to fix anything, or just request that it gets deleted. Insanephantom (my Editor Review) 08:26, 12 January 2007 (UTC)
I've noticed people opposing for not answering optional questions. Remember that my criteria is based on my observations on what people look for in RfAs. But I will delete that criteria anyway, and possibly the whole page if there's much more wrong with it (or simply stop quoting that criteria when I vote in RfAs). Insanephantom (my Editor Review) 08:30, 12 January 2007 (UTC)
  • I know you mean well by it but it really isn't the best way of going about it. You've based your criteria on people who make their votes on quantifiable things (mainly editcount), because it's kind of tricky to base anything on people who don't make their votes on quantifiable things (mainly trust). >Radiant< 09:11, 12 January 2007 (UTC)
Opposing because someone did not answer an "optional" question is perfectly reasonable. The issues the questions address are issues the voter would like to see resolved or information the voter would like to know about the candidate before deciding in favor. If they are not answered, it can be viewed as the voter not getting enough information about a candidate, or having concerns about a candidates views on WP:IAR or something. Not answering a question after 24 hours is probably not a good vote, but the current RFA system only runs for seven days and the candidate who doesn't answer after 24 hours might still be editing, or even editing the RFA itself, in which case why do they not just answer the question? —Centrx→talk • 15:54, 12 January 2007 (UTC)

[edit] Arbitrary criteria

This may be radical, but I think it would help RFA if we could indicate to people that arbitrary criteria are not such a good idea, and a plausible way of doing that would be to delete or blank those many User:Foo/admin criteria pages. The main barrier to that is some people at MFD who consider userspace to be sacrosanct. >Radiant< 09:11, 12 January 2007 (UTC)

Well in a perfect world people would only vote on users they knew or had spent a couple hours reading the contribs of. But they don't. By deleting the /admin criteria pages, you won't actually stop people from using them, just make it so that we can't see what they plan on voting based on. --tjstrf talk 09:21, 12 January 2007 (UTC)
In a perfect world we wouldn't need admins, or WP:AGF because everyone could be trusted and everything would be good faith. I think the fact that the pages exist encourages others to make their own criteria. I was thinking of making an admin criteria subpage but I decided that it would be too vague, I decide based on lots of things and quantifiable statistics don't often come into play. I suggest we encourage the users with admin criteria to delete the pages, listing the negative points of having them. It would probably cause less disruption and get it over with quicker (with {{db-user}} tags). James086Talk | Contribs 09:45, 12 January 2007 (UTC)
I don't have such a page, don't plan to have such a page, and don't think having such a page is a good idea. (Really, is this something that an editor needs to write down?) It might be better if these were checklists - things to look for, as reminders, without specifics (e.g., "consistent use of edit summaries?"; "has worked with images?"), if one assumes these criteria might fail to be mentioned by others, in the RfA, even if really a problem. Having said that, I oppose the proposal - it seems too much of a "you're not supposed to think this way, and we're going to try to force you not to" approach. John Broughton | Talk 13:50, 12 January 2007 (UTC)
I can't tell you how much I would love to forbid arbitrary criteria. But any specific banning of criteria would imply "We know better than you do, so we are going to tell you how you will think". Same with deleting the absurd criteria pages which do things by point systems; the message is "You don't think well enough to come up with criteria". -Amarkov blahedits 19:15, 13 January 2007 (UTC)

[edit] Discuss then vote

People say RfA should be a discussion, but the voting bit obscures this. The discussion doesn't obscure the voting, but some people vote and don't check back to read later discussion bits. Somewhere above there was a suggestion for a two-day discussion phase before a voting phase. I'd like to revive this suggestion to have discussion and voting separated. Discussion would continue during the voting, but should that discussion move to the voting section, or stay at the discussion section? Another consideration would be the need to find a way for the closing bureaucrats to discern which votes are 'votes without discussion', which generally count for less than votes with reasons attached (and more reasoning than 'per nom'). Carcharoth 12:09, 12 January 2007 (UTC)

PS. The talk page is currently used to post things like edit count and analysis, but could be used for discussion. In some lengthy cases, dicussions do split off onto the talk page. Carcharoth 12:10, 12 January 2007 (UTC)
That's a really good idea. A request could also be stoppable after the discussion. if the nom has less than 1,000 edits, people could explain they wouldn't get past the voting stage. The discussion could also include the questions that people ask now, so that way they wouldn't interfere with the voting stage. I support this idea.--Ac1983fan 12:42, 12 January 2007 (UTC)
I believe this has been discussed before. Basically, everything anyone suggests has probably been discussed before. --Deskana (request backup) 12:43, 12 January 2007 (UTC)
[edit conflict x3] I think this is a good idea. I can't seem to find the reason it was rejected before (looking through the archives I can't find the discussion of it). The discussion period would encourage people to make a more educated decision instead of just piling on support/oppose per User:foo. James086Talk | Contribs 12:51, 12 January 2007 (UTC)
WP:DFA. – Chacor 13:01, 12 January 2007 (UTC)
Support per discussion. :-) Carcharoth 13:31, 12 January 2007 (UTC)
  • Yep, it's been discussed before, but it's never really been tried before. So why not? I think it'd be an improvement. >Radiant< 13:57, 12 January 2007 (UTC)
More discussion is a good idea. Forcing people to not vote for two days strikes me as very artificial however. I am all for trying it, but I think we'll end up with the same outcome as now, but instead of the opinions being clearly separated into sections they would all be clumped together. I doubt it would be an improvement. Oleg Alexandrov (talk) 14:32, 12 January 2007 (UTC)
Aaron Brenneman tried to implement a 'discuss first' section in his second RFA and ended up being roundly ignored by the general population (see the "Commentary and evidence" section"), and lambasted by some others for refactoring (Kim Bruning, of course, supported because of it :). WP:DFA was an incremental improvement idea slain by a heavy-handed implementation, as near as I can tell. Do you have another idea for how to increase true discussion, Oleg? -- nae'blis 15:13, 12 January 2007 (UTC)
Responding to Oleg: You could have the discussion section itself subdivided into support, oppose, neutral areas of discussion, though more constructive, and non-voting, would be something like: "good points", "bad points", "ways to improve", "really bad points", "technical objections", and "other" (for stuff that people think of themselves). That way, at the end of the time period, it is easier for people to assess what has happened, and for the candidate to improve. Think of it like a review process. Even if the candidate is successful, there will be room for improvement. Carcharoth 15:19, 12 January 2007 (UTC)
Another point is that nominations could be withdrawn in the discussion phase if the candidate felt persuaded to by the comments. A more tricky question is whether the candidate might be allowed to fix a malformed nomination, or iron out various objections before voting started. That would seem to be one of the points of a discussion phase, but I know some people like to vote on the basis of the 'raw' candidate, rather than a candidate 'polished' by a discussion. Carcharoth 15:44, 12 January 2007 (UTC)
This is becoming unnecessarily complicated. The current system is still a discussion, let us not forget that. And people are allowed to change their votes, and routinely they do. And if they don't change their votes, sometimes it is out of sheer stubborness, not because they failed to follow up with the RfA page. Also, practically speaking, I doubt that delaying voting for two days would change the final outcome. All in all, the current system is very simple and works very well I think. Experimenting should be encouraged of course, but a massive switch to a new system while strictly enforcing no-voting for two days is probably not good I would think. (Oh, and I have no idea how to increase discussion. I think that happens naturally when necessary. Think for example of the discussion (and massive number of votes) in the nomination of ProtectBot). Oleg Alexandrov (talk) 17:00, 12 January 2007 (UTC)
Several of the ProtectionBot votes were ill-informed. But I'm guessing at least some of those wouldn't have read any discussion anyway, or wouldn't have responded to correction of their misunderstandings. Then it would have been up to the bureaucrat whether to ignore the vote or not. Maybe it should be made clearer that your vote probably will be discounted if you don't respond to corrections of misunderstandings. There is currently an RfA where some people are showing that they don't fully understand the Block vs Ban distinction (an understandable mistake, but embarassing if you make the mistake of being unclear about it when opposing in an RfA). Carcharoth 17:21, 12 January 2007 (UTC)
Having bureaucrats ignoring votes on the ground that certain votes are "ill-informed" is another idea which sounds nice in theory but is bound to cause great problems in practice. How do you define "ill-informed" oppose votes? Were some of the support votes "better-informed"? Are "me too" votes "better informed"? This is a can of worms which better be stay closed, if you ask me. :) Oleg Alexandrov (talk) 17:35, 12 January 2007 (UTC)
THE BOTS WILL DESTROY US ALL is a great example of an ill-informed vote: It's completely out of touch with the situation at hand and based on sci-fi paranoia. --tjstrf talk 17:50, 12 January 2007 (UTC)

I remember someone once tried to implement "discussion" and was badly shot down. RFA was even delisted for being malformed. – Chacor 15:23, 12 January 2007 (UTC)

A problem I see is if no one participates in the prior discussion, which is the same problem as people not participating in the "Discussion" or "Comments" section in the current RfA system. Having a first vote and then a second vote (while requiring either that the candidate passes both votes or that voters must vote in both RfAs) might alleviate the problem of people voting and then leaving without reviewing objections, but there are obvious problems with this idea. —Centrx→talk • 15:50, 12 January 2007 (UTC)

Having people ask the candidates questions and having the candidate respond would probably not have the problem of low participation, though, because questions and issues are already asked and discussed in RFA. There could still be a problem of people ignoring the discussion beforehand and only raising objections once the voting period starts. —Centrx→talk • 15:57, 12 January 2007 (UTC)
Maybe call it an inquistion questioning session, then? A community interview? Plus group discussion, and then, tagged on the end, the consensual vote. Carcharoth 16:10, 12 January 2007 (UTC)

Another point is that some people may only pass by RfA "once a week" or something, and vote on the current RfAs (some will have just started, some will be halfway through, and some will be nearly finished). Should it be expected that all voters read all the discussion before voting? That would need voters to turn up at the end. Thus the need for the voting period to remain as before, and for the discussion period to be an expanded area at the beginning. Or shortening the voting period. Things that would need to be sorted out before trying this system. I agree with Radiant though that it is worth trying. I don't participate at RfA that much, mainly because I don't feel comfortable voting for candidates I don't know or haven't taken the time to review properly (don't have the time for that, realistically speaking). I know adminship shouldn't be a big deal, but gaining the trust of the community probably is a big deal. Would further discussion help that process? It would help me decide on candidates I didn't know very well. One final point, unrelated to the preceding bits, some people may arrive already decided, and discussion won't sway them one way or another. Carcharoth 16:10, 12 January 2007 (UTC)

I guess my only problem with this is that it may draw out the RfA process and make it take longer, which is not necessary. I wish that users would discuss while they'e voting though, which is what I try and do. It's certainly a good idea, but it's probbaly harder to implement than it sounds.--Wizardman 16:15, 12 January 2007 (UTC)

  • It seems to me that if we were going to implement this, we'd need to do away with the formal vote lists and groupings entirely, and just have a single section with the discussion as a whole in it. --tjstrf talk 16:22, 12 January 2007 (UTC)

This has been discussed on multiple occasions before. Barring some revelation regarding the problems this is supposed to solve, I don't see this gaining any traction. Once again we are confronted with a solution designed without identifying just what is wrong with RfA. --Durin 18:03, 12 January 2007 (UTC)

Durin, you're beginning to sound like an obstructionist. People have brought up numerous things that they posit are wrong with RFA over time, but every time you maintain that they have not been discussed/pointed out/consensed on. But when people try to discuss them widely, they are ignored/shut down by a vocal group that want to maintain RFA as is. How do you propose we resolve this logjam that keeps coming up every few months? Hint: I don't think "accept that nothing is wrong with RFA" is going to be acceptable. Your point-by-point suppositions on Wikipedia_talk:Discussions for adminship might be a good place to start collecting data... -- nae'blis 20:20, 13 January 2007 (UTC)
  • You're right. I am obstructionist. I have every intention of being obstructionist to poorly thought out ideas that do not approach modifying RfA through some form of methodical approach, for example a SWOT analysis. That's my proposal for solving it. I've said it dozens of times, and I'll keep saying it until I'm blue in the face. Plenty of people want to "reform" RfA, yet few have made any real attempt to ascertain what's wrong with it. I've already done multiple point by point assertions and responses before. --Durin 14:01, 17 January 2007 (UTC)

I disagree. I think a forced discussion period before any voting is allowed to take place will fix a rather major problem in RFA, which has already been identified above. That problem is that it is a de facto vote in many circumstances, in which people merely vote, don't participate in discussion, and certainly don't return later to see how the discussion is going any maybe change their minds. Forcing discussion at the outset would be beneficial. If we don't want to make the RFAs longer, why not two days of discussion followed by five days of continued discussion and voting? --Cyde Weys 19:10, 12 January 2007 (UTC)

  • I respect your disagreement. However, the very same proposal was made before. In fact, there was an attempt by two bureaucrats to force it upon RfA. It was soundly, roundly defeated and criticized at the time for a number of reasons. See Wikipedia:Discussions for adminship and the archives of this talk page for more background. --Durin 19:22, 12 January 2007 (UTC)
  • I am shocked RfA is being called a "vote". Hardly. It is simply more a vote than, say, MfD, but it's still not a vote. Yuser31415 20:07, 12 January 2007 (UTC)
Yes it is. I can only imagine the hysterical wave of protest from certain parties if someone got sysopped with only 69.97% support. It's a vote with a little discression, but a vote nonetheless. Proto::► 09:15, 13 January 2007 (UTC)
Yeah. With an AfD, nobody cares if the closing admin discounts the fact that WP:IDONTLIKEIT delete votes outweigh keeps 33 to 1. But in an RfA, a bureaucrat has no discretion to discount votes in large numbers. If all the opposes are WP:YOUARE(insert ethnic category), WP:YOUONLYHAVE2999EDITS, or WP:IDONTLIKEEVERYONEOFYOUROPINIONS, the closing bureaucrat would still be expected to close it as failing. -Amarkov blahedits 19:21, 13 January 2007 (UTC)
Interestingly, BigDT's RfA just closed yesterday with a nominal percentage of 79.x, which might show that the 'crat exercised some discretion after all. I haven't seen a hue and cry...yet. -- nae'blis 20:20, 13 January 2007 (UTC)
It was closed as sucessful, and "discretion range" is between 70 and 80%, so I'm not sure why that would be a problem. -Amarkov blahedits 22:01, 13 January 2007 (UTC)
75-80%, unless you want someone to yell at you... Titoxd(?!?) 04:23, 14 January 2007 (UTC)
Errr, right. For some reason I thought we'd crept up to 80-85% being the discretionary window. Oh well, there goes my shot at 'crathood... ;) -- nae'blis 17:34, 14 January 2007 (UTC)
Too true, and that's unfortunate in my view, but at least RfA involves reasoning which people have the ability to read and consider along with links to evidence. I do take into account the quality of the reasoning and the strength of the comment, such as strong support or weak oppose, among other things in close cases. Well reasoned arguments are much more valuable for building consensus than blank votes. - Taxman Talk 07:17, 14 January 2007 (UTC)

Hmmm, I think I might have some ideas that might not even change the design *that* much, at least cosmetically. :-) Kim Bruning 20:19, 13 January 2007 (UTC)

[edit] Proposal to change a question

Currently, RfA question 3 reads, "Have you been in any conflicts over editing in the past or have other users caused you stress? How have you dealt with it and how will you deal with it in the future?" I would suggest to amend this to "Have you made mistakes in the past that could have reflected on the way other users acted toward you? How did you learn from the experience?", which covers both the current question 3 and asks another important point; it's important to have admins who can make mistakes and learn from them. On the other hand, my suggestion can simply be a 5th question instead of replacing the 3rd. Cheers, Yuser31415 21:14, 13 January 2007 (UTC)

Sounds fine, but it is really two questions. --Majorly 21:25, 13 January 2007 (UTC)
So, perhaps we should add it as well? How does everyone think? (Obviously I'm not going to add it unless I get quite a lot of support for the idea.) Yuser31415 20:45, 14 January 2007 (UTC)
That's not really a yes/no question... I would humbly suggest that anyone who would answer "no" is not really qualified to stand for adminship. Accordingly, it might be better to change it to "What mistakes have you made..." I like the idea, though. -- Visviva 15:25, 15 January 2007 (UTC)

[edit] Wikipedia:Requests for bureaucratship/Candidate questions

Just for everyone's info, I've updated this page to make it look more like the RfA template. I also think it should be moved to Template:RfB. Comments would be appreciated. --Majorly 21:45, 13 January 2007 (UTC)

I do not object to the move and agree completely with your update. :) Cbrown1023 00:48, 14 January 2007 (UTC)
I moved it, it's hardly used anyway so it's not a big deal. --Majorly 12:59, 14 January 2007 (UTC)
Looks alright. As you said, not a big deal. Redux 16:21, 14 January 2007 (UTC)

[edit] Subheadings in RfA with a bot updated TOC

In the discussion above, #Subheadings in RfAs?, there was a rather wide agreement for subsections in RfAs, as long as the table of contents does not grow and the bots do not get confused. Inspired by Heligoland's comment there, I implemented a fake table of contents, in the form of a table which can be updated automatically say every 5 minutes, and which any user can update at any time by clicking on a provided link.

The experiment is at User:Mathbot/Requests for adminship. It needs more tweaking obviously, but I'd rather wait to see if using this fake TOC would be an acceptable solution. Oleg Alexandrov (talk) 04:30, 14 January 2007 (UTC)

I think that would work well, but User:Mathbot/TOC (or the template actually used) would need semi-protection so that it doesn't get vandalised which could cause the problems it's trying to eliminate. James086Talk | Contribs 05:07, 14 January 2007 (UTC)
I tried some CSS hacks, but I still can't figure out why the lines are more spread out on the fake TOC than in the "real" one. Titoxd(?!?) 05:18, 14 January 2007 (UTC)
Probably because the "real" TOC is a list with 0 padding and margin while the fake one is a table with default padding and margin... -- ReyBrujo 05:34, 14 January 2007 (UTC)

What's wrong with doing it with noincludes as has been suggested every time the subject comes up and seems to be accepted by everyone but never gets implemented? --Tango 17:33, 14 January 2007 (UTC)

I suspect what's happening is we need someone who is a transclusion wiz to help us sort out how to get noincludes into the RFA template. I swear we've consensed in theory to do this like three times in the last 7 months, though... -- nae'blis 17:35, 14 January 2007 (UTC)
This has the disadvantage that the noinclude/include code can be rather complicated and people may edit it by mistake.
Anyway, I created and semiprotected Wikipedia:Requests for adminship/TOC and transcluded it onto the main Wikipedia:Requests for adminship page. I will have it updated every 5 minutes (and it can be updated at any time by clicking a link). Let us see if this works out. Oleg Alexandrov (talk) 17:42, 14 January 2007 (UTC)
There isn't much point implementing that part of it without putting sections in the noms... --Tango 17:49, 14 January 2007 (UTC)
Well, now you can go for it. :) Oleg Alexandrov (talk) 17:51, 14 January 2007 (UTC)
It's your idea, you implement it. I think we should do it with noincludes, I think your new TOC just looks ugly. --Tango 18:04, 14 January 2007 (UTC)
You can add some style="padding-top: 0px; padding-bottom: 0px;" to every cell to slightly minimize the wasted space. Or go with a list, just like the original TOC. -- ReyBrujo 18:06, 14 January 2007 (UTC)
Sorry, I forgot about the bots. I must go offline now but will notify soon the bot owners that section headings in RfAs may be used in the future. I will also add the padding thing noted above. Tango, you may revert the TOC in the meantime if you really don't like it. :) Oleg Alexandrov (talk) 18:27, 14 January 2007 (UTC)

I made the fake TOC use exactly the same html table code as the real toc, see Wikipedia:Requests for adminship. The only thing I could not solve is how to insert <h2>Contents</h2> at the top of the TOC without making it think it is an editable section. For now it just uses '''Contents''' which looks the same on my screen. Any help here?

I wrote to Tangotango (who appears to be different than Tango :) asking if he could modify the code of Tangobot to accept headings in RfAs. My bot script adding the edit summary usage has been patched too. Are there other bots which could get confused by the switch to headings for Support, Oppose, etc? Oleg Alexandrov (talk) 21:49, 14 January 2007 (UTC)

Looks the same for me. Also, instead of having a bot regenerate this every 5 minutes, wouldn't it be the same to have one updating it whenever a change in the RFA page is made? -- ReyBrujo 22:14, 14 January 2007 (UTC)
That would not be enough. WP:RfA transcludes a few pages, like {{Wikipedia:Requests for adminship/Front matter}} and {{Wikipedia:Requests for adminship/bureaucratship}}, so changes on those pages can affect the TOC. Also, how do you tell if the WP:RfA page changed without frequently fetching it, which would also use resources. And doing things every five minutes it is rather cheap I think. All it takes is an http get request followed by an update of Wikipedia:Requests for adminship/TOC‎. And if no changes happened the update does not show up in the history. Oleg Alexandrov (talk) 22:22, 14 January 2007 (UTC)
No problem then for me. I would add a note somewhere that, after transcluding a request, the user should refresh the TOC, in case the bot is down. I am still against the concept of dividing in sections, although I understand why it is being done. No objections from me. -- ReyBrujo 23:04, 14 January 2007 (UTC)
Just wondering, wasn't all the TOC stuff made so that the different sections could be editable? Am I missing something, or just running too fast? :-) -- ReyBrujo 04:39, 15 January 2007 (UTC)

[edit] Guide to requests for adminship

I've made a proposed addition to it. Any thoughts? -- Selmo (talk) 01:21, 15 January 2007 (UTC)

[edit] Question #2

I think we should get rid of question #2 for the reason I stated on my RFA, which was: "I've written a few articles and worked on a few, however, I don't see why this is a question for adminship, as adminship has nothing to do with writing or adding to articles. A lot of my edits are vandalism reverts. Adminship helps manage Wikipedia, not write it.." Thoughts? —The preceding unsigned comment was added by TeckWiz (talk • contribs).

I actually think that question two is a good question. As in it you can actually state your whole mission on wikipedia. Such as while I have worked on "so-and-so" I tend to be more of a vandal fighter. Also there are a lot of !voters who look for strong material contribs, as it shows just how hard that side of wikipedia can be. So I personally don't mind it. Arjun 03:23, 15 January 2007 (UTC)
While in principle I would agree with you, TeckWiz, I think the question mainly exists for the purpose of deterring users who are insanely focussed on adminship; they also have to contribute to our encyclopedia. By the way, happy Wikipedia Day! Yuser31415 03:40, 15 January 2007 (UTC)
That is true, and thanks for reminding me that it is wikipedia day. I totally forgot! Arjun 03:52, 15 January 2007 (UTC)
Indeed, Wikipedia is, before anything, an encyclopedia. Check the five pillars. I guess some administrators would leave Wikipedia if, suddenly, the world becomes ideal and every user, even anonymous, are made administrators. With 0.03% of editors and the rest administrators, this place would disappear. But with 0.03% administrators and the rest editors, the place is running well enough ;-) -- ReyBrujo 04:36, 15 January 2007 (UTC)
Yes remember, "what was the reason you joined wikipedia in the first place". 9/10 users would say to help contribute to articles to make them better. Arjun 04:41, 15 January 2007 (UTC)
I've always viewed question #2 as the most important -- Samir धर्म 05:28, 15 January 2007 (UTC)
Notice that it says or contributions. I don't think it's meant to specify mainspace, or it would say so. I do agree with you that article contributions aren't important to adminship. -Amarkov blahedits 05:32, 15 January 2007 (UTC)
Agreed - I never have interpreted it to mean only mainspace contributions, and I've always thought it's the most important standard question, because it gives us a chance to hear a summary of how the user has helped Wikipedia so far. -- Renesis (talk) 07:59, 15 January 2007 (UTC)
I agree with ReyBrujo and Samir that contributions to the article namespace is the most important aspect. Admins that only stick to their maintainin will eventually develop a case of adminitis. Wikipedia is foremost an encylopedia and I believe it is crucial for every admin and for that matter every Wikipedian understand this. I am an editor first and an admin second. GizzaChat © 11:11, 15 January 2007 (UTC)
Ditto, and that is the only question where the quality of the candiadate can actually be tested. It is easy (and often necessary) to lie for Q.1 and Q.3 can always be answered picking specific cases where the candidate was right. Tintin (talk) 11:58, 15 January 2007 (UTC)
"It is easy (and often necessary) to lie for Q.1" Assuming this is true (which I would agree it is) why aren't we discussing how to change question one so that it is not necessary to lie in order to pass? That seems a rather vital change to make. --tjstrf talk 12:47, 15 January 2007 (UTC)
How is it necessary to lie in an RfA? If someone lies in an RfA, I'm going to strong oppose them just for that. If you can't pass by telling the truth, too bad, you can't pass an RfA. The solution to that is to either change so you can pass, or just not get adminship, which really should not be a huge problem. The solution to the fact that people feel you are not suited for adminship is not to lie so they think you are. -Amarkov blahedits 15:30, 15 January 2007 (UTC)
I don't think it is necessary to lie in the answers to the questions. It may look better to answer slightly different questions, though: Q1 "Why do you want to become an administrator", Q2 "What have you done for Wikipedia so far?" and Q3 "How do you interact with others who disagree with you?". Many people who take Q1 literally get opposed for naively answering that they want to work on the wikify backlog. Kusma (討論) 15:52, 15 January 2007 (UTC)
You can't prove someone has lied until he has became an administrator and some time passed, can you? -- ReyBrujo 16:33, 15 January 2007 (UTC)
And even then, you need to consider whether the candidate has really lied. I said in my RfA that I would never block an established user indefinitely without ArbCom, but changed my mind later. Should I be desysopped for breaking that promise? (Note: I didn't intend to break this promise, I just changed my mind about this, based on more experience). So I think asking people to make campaign promises in their RfA is not very useful. Kusma (討論) 16:39, 15 January 2007 (UTC)
Reviewing my own RFA, I notice the only thing I said I would do but haven't is participating at Wikipedia:Copyright problems. I had done some work at Category:Images with unknown source and Category:Images and media for deletion, but not as much as I would have wanted. Everything else I had said, I did. And even I pointed that I would not be a "full time" administrator, I enjoy doing wikignome stuff, and haven't stopped since (lately, adding images to Commons). In other words, if I were to hold a new RFA today, I would say pretty much what I said two months ago. -- ReyBrujo 16:47, 15 January 2007 (UTC)
In my opinion, the question is necessary to determine the quality of the contributions of the editor. Personally, I am quite interested in the answers to this question. ← ANAS Talk? 19:45, 15 January 2007 (UTC)

Q2 is the most important in my opinion as it shows factual info, where as Q1 and Q3 can be political and political promises. Perhaps we should go through and see how well people kept their promises. Having said that, they never promised any set number of actions...Blnguyen (bananabucket) 03:42, 16 January 2007 (UTC)

[edit] Late evidence and comments

Partly continuing the above discussion about discussing then voting, what is the procedure when something major is raised near the end of an RfA that might lead many people to change their votes (if they had had the time or opportunity to return and read the whole discussion)? Is it just a case of "the candidate got lucky and the issue was not uncovered until it was too late", or does the closing bureaucrat have the option to "continue discussion in the light of new evidence"? How often is this discretion exercised? This sort of thing applies more often to AfDs, where new evidence can sometimes (though not as often as it should) lead to the closing admin to call for a relisting of the AfD in the light of the new evidence. But what is the procedure at RfA? I'm guessing that one way around this would be to close the RfA to new voting, but to allow a period of x days for the participants to read the whole discussion and decide whether or not to change their vote, and only then to close the RfA. But that is probably too much for what is in all likelihood a non-existant problem? Carcharoth 16:06, 15 January 2007 (UTC)

Bureaucrats can reset the closing date if they feel it's necessary. -Amarkov blahedits 16:13, 15 January 2007 (UTC)
I've not been watching RfDs for a long period of time, so I'll leave it to others to say whether the problem is pretty much non-existent (my sense is that such is the case). Rather, I note that because a successful candidate needs somewhere around a 4-1 ratio of "support" to "oppose" opinions, it doesn't take a lot of changed opinions and additional opposing opinions near the end to sink an RFA. For example, an 80-10-10 tally (easy success) can become a 75-25-5 tally (failure) with just five "support" opinions shifting to "oppose", five "neutral" shifting to "oppose", and 5 new "oppose" opinions. John Broughton | ♫ 16:16, 15 January 2007 (UTC)
I am confident that our bureaucrats will fail an RfA that numerically qualifies as "pass" if late evidence shows that the user is Willy on Wheels. Kusma (討論) 16:29, 15 January 2007 (UTC)
Brilliant. Thanks! I thought that might be the case. Just wanted to check. Carcharoth 18:53, 15 January 2007 (UTC)
'crats often delay closure where consensus is potentially unclear at closing time (happened in Joturner2 and probably mine). I think they'd let consensus change rather than closing a numerical pass as fail, but the 'crats are the best to ask.--Kchase T 21:31, 15 January 2007 (UTC)

[edit] Subheadings and TOC

I'm trying to figure out why the TOC for the RfA page is so...wonky as of late. I assume it has to do with the discussions above; however, the tinkering isn't having the results I expected (I thought we'd end up with something like we had on the ProtectionBot RfA). Nothing seems to have changed except for the weird spaces and gaps. Agent 86 20:30, 15 January 2007 (UTC)

Not only gaps, but actually unlisted noms. Check the first (Yannismarou's) which doesn't appear and led me do this and that. NikoSilver 20:41, 15 January 2007 (UTC)
There is a 5-minute delay between bot runs, so at worst a candidate needs to wait 5 minutes for his nomination to be inserted into the TOC (unless he updates it himself by clicking on the link there). -- ReyBrujo 20:45, 15 January 2007 (UTC)
What is worrying, though, is this edit, when the bot delisted two candidates... -- ReyBrujo 20:47, 15 January 2007 (UTC)
Thanks. Impatient me 20:49, 15 January 2007 (UTC)

Maybe we should restore the old TOC until the new one is perfected. I also notice that the bots aren't updating, either, and one of the concerns that was supposed to have been addressed is that the new TOC and subheadings wouldn't disrupt things like Tangobot. Agent 86 21:11, 15 January 2007 (UTC)

I don't think the new TOC influenced Tangobot, since there are no subheadings in RfA for now.
I don't know why the bot delisted two candidates, it may have to do with server cache. I will think about that.
I took off the experimental bot generated TOC at least until I hear from Tangotango that his bot won't be affected by extra headings (and until then there will be no extra headings to start with). I will also think of the server cache issue. Oleg Alexandrov (talk) 03:09, 16 January 2007 (UTC)

[edit] 14 RfAs!

At the moment, there are 14 RfAs! Is that the most there has been, or has there been more open RfAs at one time? Cbrown1023 03:55, 16 January 2007 (UTC)

Nope, I recall 15 - maybe 16 - once. Picaroon 04:08, 16 January 2007 (UTC)
And a few weeks ago there were 2. :) GizzaChat © 08:50, 16 January 2007 (UTC)
Back in 2005 it was more common to have over 20 at a time, especially in October and December. See this version as an example, but certainly not the record. NoSeptember 13:04, 16 January 2007 (UTC)
  • I do note several snowballs among there, though. Does anyone have the "most promotions in a week" record? >Radiant< 13:26, 16 January 2007 (UTC)
  • The record for most promotions in one week appears to be 22 for the week of Dec. 8-14, 2005. There is a pattern that the second week of the month has more promotions than the other weeks (perhaps because a lot of nominations are made at the start of the month). Look at the hidden text at User:NoSeptember/admincount if you want to see the pattern. NoSeptember 13:40, 16 January 2007 (UTC)

[edit] History of vandalism

I am just curious to know if it is still possible for me to be able to RFA one day in the future even though when I was new I started out as a vandal and was blocked for a month. Since then I have strived to turn a new leaf (I am actually a vandal fighter now). Is it possible or common for users with poor history's to successfully RFA. Culverin? Talk 04:14, 16 January 2007 (UTC)

Um... yes, theoretically. In practice, you'll have to go for at least a year with a near-perfect record, and no blocks, before you'll have a chance. -Amarkov blahedits 04:16, 16 January 2007 (UTC)
(edit conflict):It's up to each individual looking over the situation to decide if he or she is willing to overlook past misconduct, but in most cases a long period of positive contribution will cause past mistakes to fade. If you were to run now, though, I personally would oppose-the misconduct is too recent. Give it more time, and keep up doing good work. Seraphimblade 04:18, 16 January 2007 (UTC)
If we were to get practical and down-to-earth, fortunately yes, you can. Wikipedia welcomes reformed vandals, and the community always wishes to have a reformed vandal become a responsible admin. However, such a person would be held to a much higher standard than most others, and hence you must pretty much go for a year or more with a clean slate, along with other expectations an RfA voter might expect from an otherwise (pardon the poor word choice) unmarred candidate. Keep up your current good work, and you can make the cut. Cheers, 210physicq (c) 04:29, 16 January 2007 (UTC)
Did you vandalise with your current username? I used to be a vandal as an anon a long time ago... and didn't impact my RfA at all of course! GizzaChat © 08:54, 16 January 2007 (UTC)
It seems you've not been here that long and therefore the vandalism is pretty recent. I'm concerned that although you're up front here about your vandalism, your talk page is a whitewash, as you've archived the old material but removed the index to the archives, making people think that previous posts (including all those referring to the vandalism) don't exist. This gives the appearance of being sneaky. I'd restore the archive links, do some more "good boy time" here, get an editor review and then get a nom for RfA, which may still fail because in life, some people don't think leopards can change their spots. Alternatively, ditch this user name and create a new one. Just make sure you don't use both or doing anything that even remotely looks like a repeat of your sock puppetry, as Checkuser is pretty powerful and will catch you out. Yes, your edit count and history will be back to zip, but it'll be a clean zip... and you've not been here that long anyway. --Dweller 09:40, 16 January 2007 (UTC)
I think you're not the only one, when people first join, they might vandalize but become a good editor/vandal fighter later. I somewhat regret not starting a new account after what I did to the Sandbox when I first started (and I got blocked, although this was highly questionable because I was never given a warning, and I assumed doing that to the sandbox would be OK, whereas doing that to a normal article would not be OK.), but now, I am extremely dedicated to vandal fighting. Keep up the good work you've been doing for 6-12 months, and hopefully you should stand some reasonable chance. Insanephantom (my Editor Review) 11:11, 16 January 2007 (UTC)
See Wikipedia:Requests for adminship/Aaron Brenneman (second nomination) for an example of a vandal-turned-admin. – Chacor 12:04, 16 January 2007 (UTC)
Dude. Aaron was never a vandal. He only decided to have some fun with his ol' pal Tony and the pitiful joke backfired. LOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOL Don't block, don't block :D — Nearly Headless Nick 12:20, 16 January 2007 (UTC)
Rant-back: Example of admin-turned-vandal – Wikipedia:Requests for adminship/Chacor. Argh! Don't kill me. :D — Nearly Headless Nick 12:22, 16 January 2007 (UTC)
{{citation needed}} :P – Chacor 13:23, 16 January 2007 (UTC)

(undent) - Culverin - you were blocked in September (four months ago) for creating sock puppets and for personal attacks; you're apparently now arguing for restoring Esperanza, your talk page has a number of very recent complaints about your behavior; a joke banner on top of that page is probably off-putting to newbees -- and you're thinking about becoming an admin?

Yes, I agree that it's still possible for you to become an admin, but I hope you keep in mind, per Wikipedia:What adminship is not, Adminship is not a trophy. In the best of all worlds, in a year or more, someone will come to you and ask to nominate you, and you'll have a clean record and accept, and you'll acknowledge the problems of your past and you'll pass the RfA and be a productive admin. Or you'll decide it's too much not you to avoid jokes and typing the first thing that might come to your mind, and just be a productive contributor. Whatever you do, I hope that you trust the system enough to not try to rush (force) things, because a lot of people - myself included - take that as a very bad sign. John Broughton | ♫ 16:48, 16 January 2007 (UTC)

I never said I was intenting to run for adminship in the present, I was just curious to know what the general community thinks of this matter. I never said I intend to run now! I would make WP:200 for 200 opposition votes. I would never have any intention of abusing adminship if I was one. I admit I have some problems at the moment but I will strive to fix them as soon as possible. Thank you everybody for your reply's and messages. Thanks your kind advice. Again, thank you. Culverin? Talk 09:19, 17 January 2007 (UTC)

[edit] Mathbot's tool

I'm curious, why does every RfA contain a link to mathbot's tool rather than just typing out the information? It seems unnecessary for potentially every person who considers voting in the RfA to use the tool when two lines can fully report its output. Is there any reason not just to type out the edit summary usage stats? WJBscribe (WJB talk) 08:28, 17 January 2007 (UTC)

You could try asking Mathbot's owner, Oleg Alexandrov. --ais523 08:45, 17 January 2007 (UTC)
Originally the information was embedded in. Then some people were saying that people read too much into that information and oppose or are unhappy even with a 75% edit summary usage. In order to minimize the impact of the edit summary usage (without ignoring it altogether) the compromise was that people eager for the edit summary usage should find it by themselves. Oleg Alexandrov (talk) 16:11, 17 January 2007 (UTC)

[edit] Weird delisted RFA

Wikipedia:Requests for adminship/Xunex - anyone care to figure a way of fixing this? I cannot imagine edits like this help him in any way. – Chacor 12:03, 17 January 2007 (UTC)

Well, he's enthusiastic... and I'm guessing he's also a kid. Someone should close that and give him a kind rejection notice.--tjstrf talk 12:10, 17 January 2007 (UTC)
I see that it's been cleaned up and listed, but must we really put the candidate through this? It's guaranteed to fail. – Chacor 12:36, 17 January 2007 (UTC)
Well, we could require a minimum of 1000 edits - there isn't a snowball's chance that anyone could succeed with an RfA with less than that (and, yes, I know, a candidate could have a lesser number of quality edits ... but no such person has shown up here in recent memory). It's either that or put up with these exercises in futility. (It took a couple hours for this one to reach the WP:SNOW level, so, arguably, it's not that much of a problem.) John Broughton | ♫♫ 15:00, 17 January 2007 (UTC)
  • It's been suggested before, and shot down. It's not so much the level that's a problem, though there have been people with less than 1000 edits that have passed. The problem is the slippery slope effect. Where do we stop moving this arbitrary line? --Durin 16:43, 17 January 2007 (UTC)
    • Yes, I know it's been shot down but consensus can change. I was saying, more or less, that there is an alternative to putting poor candidates through a hailstorm.
    • Nor do I think the proposed alternative would result in an fewer successful admins - when exactly was the last successful RFA with less than 1000 edits? And as for "slippery slope", that's rather often misused - if ownership of machine guns are prohibited, isn't that a slippery slope that leads to registration of kitchen knives? If Congress can pass even a single law that taxes U.S. citizens, then they can pass laws that turn the country into a socialist state, etc., etc. In short, why not just discuss the pros and cons of what is proposed, rather than dismiss it as a possible road to hell? We can "stop moving" the arbitrary line when there is no consensus to move it, that's when. Right now it's at zero - that's as "arbitrary" as any other number. John Broughton | ♫♫ 19:30, 17 January 2007 (UTC)
      • Ok, discuss it then :) I agree consensus can change. As a bit of a counterpoint, I note that your very first question is "when exactly was the last..." (emphasis mine). This points to standards changing over time. People here constantly rail against the ever increasing standards at RfA. By asking "when...", we're pushing those standards higher. Whether the arguments against a bottom level of edits are old isn't in my opinion very relevant. It's that standards keep rising, and old arguments apply just as well. You place an arbitrary level, and it will be exceeded. It's counterproductive. Lastly, whether we place the level at 500, 1000, 1500, or 20000 edits matters little.
      • Also, as always, the number of edits has little to do with the quality of a candidate. Placing an arbitrary level isn't going to stop people from applying who do not meet that level. It's already blatantly obvious that people with <200 edits have zero chance of success. There's not a single person who's at all familiar with RfA that honestly believes someone will pass who has <200 edits. Yet, it happens all the time. Roughly 10% of all applicants have less than 200 edits. Those 10% haven't a clue how RfA works, and don't care to learn before trying. They try anyways. Saying that establishing a 1000 level will change things isn't true either; we close those low edit count RfAs with abandon now as it is. --Durin 19:58, 17 January 2007 (UTC)
Well two people passed with less than 1,000 edits that I know of in 2005, and both burned out I think Jaranda wat's sup 21:44, 17 January 2007 (UTC)
I personally think the system works pretty well (see next section); the point I was trying to make is that for those people who don't like to see (brief) debacles of clueless applicants, there is an alternative. It may be a worse one than the present system (I'm not convinced); in any case, if they don't like clueless folks applying, then they should be pushing hard for an edit count floor. It wouldn't mean that everyone applying would then pass, but it certainly would increase the percentage.
The statement that the number of edits has little to do with the quality of a candidate is far too unnuanced. Certainly it's true, to restate, that beyond 1000 or 1500 or 2000 edits, the number of edits has much less to do .... Below 1000 edits, an editor simply has't been around long enough and demonstrated enough commitment to the project to be a viable candidate, if only because more than 30% of those who express opinions in RfAs are never going to support such a person. (Really, at 8 or 9 edits per day, it takes only 6 months to get to 1500 edits; is that difficult?)
And yes, standards change. When the project was younger (i.e., in 2005), a mimimum of 2000 edits (which, I've argued previously, is reality, regardless of stated policy) would have eliminated a lot of good candidates, since relatively few people had that many edits. But it's more than a year later; the number of editors has probably close-to-doubled (many successful RfAs are of people who started editing in 2006); those who started in 2004 and 2005 have had time to accumulate more edits; and so the project has the luxury (I believe) to make the standard more strict. John Broughton | ♫♫ 02:24, 18 January 2007 (UTC)
  • As I noted above, putting in a floor of say 1000 edits to apply at RfA isn't going to stop clueless people from applying. We already have 10% of our applicants having less than 200 edits. It's impossible to pass RfA with <200 edits, yet 1 of every 10 applicants tries anyway. Establishing a floor isn't going to stop that. I do not see any argument that supports instituting a floor.
  • I will stand by my statement that edit counts have nothing to do with an editor's quality. The qualities that make a person a great editor or great admin here existed in that person before they ever touched Wikipedia. The remainder of their qualifications to be an admin have to do with experience. Yet, a potential admin can read everything that an admin needs to know, review countless discussions on pertinent topics, and consider all manner of material tangentially related to the subject area...without contributing a single edit. Edit counting provides a convenient crutch to analyze something that can't be analyzed by edit counting; does this candidate understand policy with regards to being an admin? That's a function of reading, not of writing. Trying to divine the quality of a person by looking at their edit counts is akin to looking at a footprint left in snow and trying to discern whether that person like their coffee caf or decaf.
  • If you want to discern more about that person, you have to actually read what they say on various talk pages and analyze their contributions to non-talk pages. That's the only way in which having a number of edits helps; is there enough material to review? A very good contributor who is quite wordy in their statements could demonstrate that in a few hundred edits. Editors that write short comments and do stub-sorting and citation work might take thousands upon thousands. There's no way to tell until you actually get into the guts of what a person contributes. You can't do that by looking at their edit counts and saying "Hey, you've only got 382 Wikipedia space edits. Do more!" A single edit in this realm could very well show what 500 edits by another would show in that realm. I can rack up 100 edits in an hour on XfD if I like. Does it show anything? Not if I simply say Delete or Keep and sign the comment. The edit counts are meaningless and no nuance of that statement is needed. What matters is the content of those edits. --Durin 14:20, 18 January 2007 (UTC)

[edit] What's wrong with RFA

Here's something to think about for all those people who say there's something wrong with RFA... When was the last time you nominated someone? The best way to deal with the alleged problems here is to find good candidates. >Radiant< 13:50, 17 January 2007 (UTC)

True. Granted a lot of the ones people seem to have problems with tend to be the self-noms. And to anser your question, 2 weeks ago, passed :P--Wizardman 13:54, 17 January 2007 (UTC)
It's a vicious circle, why should I bother nominating someone when I have a fair idea their RfA will fail ? There's loads of people on Wikipedia who would make excellent administrators, sadly, most cannot pass the present RfA process. We know how difficult it is for an admin who has been recalled to pass another RfA. The problem with the present RfA process is that with some nominations, personal feelings and random criteria are placed before any thought of what benefits and skills the candidate brings to the project. --Kind Regards - Heligoland 15:21, 17 January 2007 (UTC)
Tend to agree. I am quite certain that if I tried for RFA now I'd not make consensus, despite me being pretty active in most areas of administration, given my desysopping. – Chacor 15:05, 17 January 2007 (UTC)
I'm sure you're correct, but aren't former admins only a tiny percentage of potential candidates? SuperMachine 15:09, 17 January 2007 (UTC)
Probably so, there aren't many former admins. But I will say that I am interested in regaining adminship, but not with this current RFA process that we have, unfortunately, it won't happen. Back to the original question: my last nomination was of Coredesat, and it was successful. But I still think the process is flawed, to an extent. – Chacor 15:12, 17 January 2007 (UTC)
Who are these loads of people? If you believe that you owe it to the project to put forth a list of them so that we can positively influence RfA. My guess is most people will disagree that these people will make excellent admins, or they just need some more time to demonstrate that they will. But Radiant's highlighted question is a great one. I'll have to go find my next one, it's been a while. - Taxman Talk 22:01, 17 January 2007 (UTC)
  • No, it's not a vicious circle, you're just assuming that. If you know good candidates, give them a chance! You might be surprised at the results, and if they pass that once more emphasizes the idea that good candidates can and do pass. That's how you break a circle. That goes for NSLE as well. As demonstrated in a recent arbcom case, ex-admins that fail RFA fail for reasons entirely unrelated to their demotion. Guanaco passed once he figured that one out; Stevertigo failed because he was incivil in his nom, not his earlier demotion; fsf failed because he was overeager, not his earlier demotion; Mongo will likely pass in a few months, and even Jimbo said that would be good. Do not assume negativity, give it a chance. >Radiant< 16:08, 17 January 2007 (UTC)
    • Radiant! - I speak from personal experience here with a RfA that was withdrawn 2 days ago. The process is screwed and it's stopping good candidates getting the bit. I know plenty of candidates that would pass an RfA, but they're not good candidates that would enhance the project. --Kind Regards - Heligoland 18:59, 17 January 2007 (UTC)
      • I'm assuming the you're referring to Wikipedia:Requests for adminship/Wizardry Dragon, which was withdrawn by the candidate at (21/24/5). I'm not sure that this particualr RfA demonstrates that the process is screwed. This seems like a case where a candidate's contributions may not (yet) reflect their true potential. I think it's likely that he'd pass in the future if he alters his behavior based on the advice given in the RfA (more article experience, dealing more effectively with opposing views, etc..). SuperMachine 19:16, 17 January 2007 (UTC)
        • It's not just that RfA, it's happening on Ryulong's third RfA. So what if someone doesn't appreciate another users tone, or a few users decide the candidate isn't doing enough work on the actual encyclopedia, the RfA process is simply not working when users that would substantially improve Wikipedia cannot pass the RfA process because of concerns largely or totally unrelated to administrative actions. Can you imagine a Reuters not hiring web designers or Encyclopedia Britannica not hiring printers just because they don't write articles ? If everybody wrote articles here, nothing would get done, not everybody is as strong at writing English as they are writing C##, Visual Basic, Python, Perl or PHP and we seem to stigmatise those who can and do contribute to the project as much time and experience as the article writers. --Kind Regards - Heligoland 20:06, 17 January 2007 (UTC)
          • It's not uncommon for reasonable people to disagree over matters such as "Will granting adminship to this person benefit Wikipedia?" Even if it's true that a person would improve Wikipedia substantially as an admin, other editors only have their contributions to judge them by. If a candidate's contributions don't reflect the candidate's true self, there's naturally going to be a disconnect. Wikipedia is simply too large of a place for editors to have personal experience with every candidate.
          • I haven't been following Ryulong's RfA, but from reading the oppose comments, there is some concern that not much has changed since his last RfA. Still, the support is at 80%, far higher than his previous attempts.
          • I do have to disagree with the analogy you presented. In the business world, there's a formal hierarchy and the web designers have specific tasks designated by the IT Director or CIO or some other manager. If the web designers or printers were allowed to work in almost any aspect they desired (like admins on Wikipedia), I would expect that knowledge of article writing would be rather high. Also, if someone is far better at programming than writing articles, they might be more interested in becoming a Mediawiki developer rather than an admin.
          • The fact is that many RfAs pass with overwhelming support. It may not be the smoothest process ever invented, but it's also not the gauntlet it's made out to be. At this moment, there's at least seven candidates who are virtual locks for adminship in the next seven days. And the flip side of "Adminship is no big deal", is that not being an admin also isn't a big deal. ;) SuperMachine 20:33, 17 January 2007 (UTC)
      • NSLE/Chacor would not pass an RFA because he abused his sysop tools, then refused to admit it when he got caught. So will probably never be trusted by a lot of people. Nothing to do with RFA being broken. Proto::► 20:18, 17 January 2007 (UTC)
      • People who fail to find out true facts before passing judgement on something also bring RFAs down. – Chacor 00:27, 18 January 2007 (UTC)
      • See also User:NoSeptember/Desysop and it's talk page for a history of RFAs after forced desysoppings.--Kchase T 20:21, 17 January 2007 (UTC)

If you believe good candidates are failing, please give examples (and more than just one - you can't prove a trend with one data point). --Tango 20:12, 17 January 2007 (UTC)

  • I have to agree with Tango. This process accounts for a significant amount of activity by users on Wikipedia. We process close to a thousand RfAs per year. This system was created by humans. It is guaranteed to have flaws. The question isn't whether we can hold up an RfA and say "See? SEE? This person should have passed EASILY!". We can, have, and will be able to do that in such a large system. The question is whether there is a pattern to suggest that certain types of candidates or certain aspects of candidates that have no relevance to admin ability are preventing their being granted admin privileges. A few RfAs isn't enough to highlight a trend. It is a small data point. Show me five RfAs following a similar pattern, and we may have something worth investigating further. --Durin 21:08, 17 January 2007 (UTC)
    • Durin, just look through the unsuccessful candidates list and look at some of the oppose reasons, people have been opposed because they actually fight vandals, because of their signature, their username or the country they come from. Perhaps the system isn't totally broken, but there isn't enough bureaucrats removing irrelevant crap and WP:POINTY nonsense from RfAs and it has been allowed to go on and on and on unchecked. --Kind Regards - Heligoland 22:26, 17 January 2007 (UTC)
      • That people come up with silly reasons to oppose has been a "problem" of RfA for a long time. It's rare that it affects the outcome of an RfA. Bureaucrats do not remove votes from RfAs. They have discretion to ignore votes if need be to help ascertain consensus. So, such behavior is indeed checked. If you feel otherwise, can you cite say five RfAs where they failed because of what you feel are poor oppose reasons? --Durin 22:37, 17 January 2007 (UTC)
        • What I consider a good candidate and what you consider a good candidate will probably differ, but if you look through failed RfAs, can you honestly say you don't see any candidate that failed the RfA process and that is to the detriment on Wikipedia, just off the top of my head I can think of the Kafziel RfA, not only a good candidate, but a victim of the present (screwed up) RfA process. Anyway, onto the more interesting point of my reply, has a candidate who has suffered enough really trivial opposes to drag their overall percentage down to say 65% been promoted? As far as I can tell, no. A candidate with a few percentage points less than 80% with a couple of suspect votes is unlikely to be promoted, let alone someone with just 65% support, no matter how much they've done for Wikipedia.--Kind Regards - Heligoland 23:45, 17 January 2007 (UTC)
          • For your second to last question - yes, at the risk of bringing back up more controversy I'll mention Carnildo, Sep 2006. Among other reasons, part of the reason for the promotion was the discounting of a number of votes believed to have been made in bad faith (those believed to be sockpuppets and those opposing based on his image work, which is roundly considered to have been excellent). For the last statement/question, there have been quite a number passed at 75 or 76% (actually I think most), making the first part of that statement incorrect. More to the point though is what others have been pointing out: a couple examples don't a problem make. Any system will have errors. To justify change you'd have to demonstrate the new system would systematically make less errors, not just that it wouldn't make the same errors. I'm not tied to the current system being the best possible, but lets at least be clear what our decision criteria should be. - Taxman Talk 01:04, 18 January 2007 (UTC)
The only "vicious circle" I see is the frequent re-occurrence of this discussion. Frankly, while no system is perfect, I don't think there is really anything wrong with RfA. Whenever the topic comes up, it seems we're only presented with anecdotal examples or vague references to a problem without providing solid facts to back it up. If a person "should have passed easily", then they should have. What facts support the claim that "most" editors cannot pass in the present process? Looking at a "snapshot" of the current RfAs, eight or nine out of twelve nominations look to be on track for success at this moment. More empirically, looking through the stats maintained by NoSeptember, I have a hard time finding anything to support the hypothesis that the system is broken. I'm sure NoSeptember can correct me if he thinks the stats show otherwise, but in such a case I'm sure it would have a sound basis in facts.
I don't think anyone is guaranteed success, and for the most part I find it disrespectful to lump together all those who oppose a candidacy; in fact, more broadly speaking, I find it disrespectful to make blanket statements about any set of viewpoints in an RfA. If someone I support fails to become an admin, or if I oppose a candidate but the nominee succeeds, I can't see how it is the fault of the system that the !vote went contrary to my expectation. Instead, I assume good faith and consider the fact that those who opined to the contrary obviously had their reasons to do so and accept that their opinion held sway. As for the two specific RfAs that are referred to above, I fail to see how the system is at fault. Just because one does not agree with the reasons given by those opposing does not mean their opinions are invalid or are to be dismissed out of hand.
As for the original proposition that sparked the latest round of this discussion, I'm not quite sure the solution matches the perceived problem. I do agree that there may be some common sense in the idea; in fact, if one can make the assumption that such candidates are indeed "good", whether they succeed or fail may provide a real basis to test the system and its effectiveness. The problem with that is one cannot simply assume any given candidate is "good" - we simply have the nominator's opinion on that point and then it's up to the rest of us to decide if we agree or not. Agent 86 22:47, 17 January 2007 (UTC)
I think there is a problem - though I don't think its with RfA per say, take a look at these admin graphs. Admin activity is barely keeping pace with article counts, and not coming anywhere near keeping pace with edits. Half the time when us "normal" folk visit admin intervention pages (especially 3RR noticeboard, CSD, Suspected Socks) there is a backlog. We need more qualified people to stand up and mop the floors. whether this problem is in the way we give admin tools or whatever, it's a problem that needs to be addressed as Wikipedia grows. --Matthew 02:55, 18 January 2007 (UTC)
We need admins who actually like deleting stuff. It seems many just want the block button handy. There's plenty of people who could be nominated as well, they just are in some corner somewhere. Blnguyen (bananabucket) 02:59, 18 January 2007 (UTC)
Bingo! I totally agree with you Blnguyen, I think that the candidates are becoming more one dimensional, and the key reason why everybody runs is the *block* button. Arjun 03:07, 18 January 2007 (UTC)
I believe it is possible with the MediaWiki software to seperate the permissions for blocking and deleting, maybe sub-admins with only deletion or blocking powers could be created. Allowing people less trusted/experianced to have some admin tools without having all of them. (Disclaimer: I have not thought this through totally, it may be a really bad idea) --Matthew 03:13, 18 January 2007 (UTC)
It's way easy - have your friendly neighborhood developer modify $wgGroupPermissions. Your friendly neighborhood developer will ask you why you want to change it, though, and will also ask you if you have gotten the support of the English Wikipedia community, which is the hard part... Titoxd(?!?) 03:17, 18 January 2007 (UTC)
The best solution might be just to seperate out all the admin powers. That way, people could get delete access, protection access, block access, and we don't have to say "I see no reason to trust you'll close XfDs right, so you can't use the blocking powers which you could use really well". It might remove some of the percieved authority about being an admin. -Amarkov blahedits 03:18, 18 January 2007 (UTC)
  • I don't think there's any evidence to support that. I did some related analysis on this a year ago. At the time, only about 55% of admins actually used their "block" ability. Also of note, in each of protection, deletion, and blocking the top 20 most active admins in each category accounted for more than 50% of all activity in that particular area. I.e., the vast majority of admins did not use their buttons all that much. In that regard I agree with Blnguyen; we need more admins who like doing that kind of stuff. Alternatively, technical solutions might be workable that reduce admin workload.
  • So why do so few admins account for 1/2 of the work? Consider what an admin does. They delete things...things other people worked to build. They also protect pages so that other people can't do what they want to do. Lastly, they also block people again so that person can't do what they want to do. I.e., everything that an admin does is really a matter of getting in someone else's way. Not surprisingly, there's considerable backlash for doing so. Take for example a message left to me by a typical irate user; User_talk:Durin#matt_harding_dance. Multiply this sort of behavior times a hundred or a thousand, sprinkle in a few RfCs here and there and maybe an RfAr just to add to the 'fun' level. Then consider; there's no a single admin out there who's getting paid to do this. Thus, it isn't just a matter of finding people willing to delete stuff. It's also a matter of finding people willing to go in and do all this dirty work, not give a darn when irate editors lambaste them, and do it with out pay just because they feel like it. That breed of admin is more rare than we might like to think. --Durin 03:26, 18 January 2007 (UTC)
  • Improving the image upload form to either prevent bad uploads which are not filled out properly with source and copyright tag, would probably take out a lot of the deletes - either that or sprotect image uploads or something. The problem is mainly in images where perhaps only 20 admins delete the stuff, because there is no culture that people should have a bit of time with image tagging, very few people learn what the image CSD are, and don't try clearing CAT:NS, CAT:NL, CAT:ORFU, etc. Blnguyen (bananabucket) 03:28, 18 January 2007 (UTC)
I looked at restricting tags on upload for a wiki I run internally at my college, the problem that arose was that users just picked a random tag. This made the image appear legitimate even when it totally wasn't, which makes it really hard to rat out the "bad" images that now look like "good" images. At least with the current system most new users seem to pick no tag or one of the trap tags and the image can be (relatively) easily flagged for attention. --Matthew 03:47, 18 January 2007 (UTC)
  • You are all really missing my point entirely. We don't need yet another debate on this. We need good candidates. I challenge everyone who participated in this thread to find and nominate someone for adminship, within seven days. It's really not that difficult, and that is the kind of attitude we need here. >Radiant< 09:25, 18 January 2007 (UTC)
I quite agree. I'm just thankful I haven't participated in this thread. Oooops. <deer caught in headlights look> OK. So I should go and find a candidate. How to start? Make a list of the people I've encountered whose edits or attitude I've been impressed with? Investigate a bit further (how long they have been here, any blocks, have they had an RfA before or declined one?). Approach them on their talk page and see if they need the tools, want to run, and think they are ready, and then nominate? Takes a bit of time, but if I can find the time, I'll give it a go. Something I read recently that impressed me was Durin's user page and his criteria for nominating someone. See User:Durin/My guidelines for admin nomination (and have a look at his other 'admin philosophy' pages as well, listed at User:Durin#Admin_stuff). Carcharoth 14:17, 18 January 2007 (UTC)
I think we did get your point. The response from some here is clear that they think your proposal isn't the solution that they think is needed. I noted above that it might be a solution looking for a problem to solve. I don't really think there's a lack of candidates at present. There certainly seems to be an excellent selection of "good candidates" right now. Overall, I don't think there is a problem with the system, and I don't think there's a problem with the overall quality of candidates. Keeping an eye open for potential nominees is always good, but I don't think there's a need for any sort of call to arms. Agent 86 21:26, 18 January 2007 (UTC)

[edit] Editor review (and admin school), and RFA

This might not be the best place for this, but the page I'm going to bitch about is as it relates directly to RFA's problems at the moment. If it would be better being discussed elsewhere, by all means move it (but let me know, please).

Has anyone else taken a look at Wikipedia:Editor review recently? It used to be a place for new(ish) editors to go and have people telling them what they could do to make them better contributors to the encyclopaedia, how they might like to stretch themselves, and point out anything they were missing, basic errors they kept making, things like that.

Now it seems to be 'staffed' by the same people who want to turn RFA into a cookie-cutter numbers game, where they assess people's edit counts purely to tie in to their personal standards for RFA, irrespective of how sane / trustworthy / sensible / etc the candidate seems to be. It's full of advice like 'you need to increase your XFD percentage to at least X% in order to pass RFA' and 'Well, RFA requires a high edit summary count, and that's high, so you'll definitely make a great admin!'

Wikipedia:Admin school is much the same thing, only at least there it's blatant (which I find a good thing, better than sneaking these awful misconceptions in via the back door).

So my question is this - is it about time we looked at giving bureaucrats greater discression (i.e., more than the 75 - 80%)? They are all smart enough to know which oppose "votes" are based on a sensible appreciation of the candidate's trustworthiness (or lack of), and which are based on a holding up of the candidate's edit count to see if it fits a filter. Either by extending the percentage downwards (70%? 2/3? 60%?) or by allowing mindless 'supports' / 'opposes' (mainly opposes) based on numbers to be stricken out and ignored. Thoughts? Abuse? Flippant dismissal? Proto::► 11:43, 18 January 2007 (UTC)

  • Perhaps it is about time we shut down Editor Review and the Admin School, then. It seems to be the month for shutting down things that aren't quite as good for the encyclopedia as they think they are. If the pages have turned into editcountitis rather than providing help, we have no further need of them, and they should be reformed, pruned or deprecated. >Radiant< 11:54, 18 January 2007 (UTC)
  • I've kind of made it a goal for the next year or so to stay out of discussions. I have not yet seen one really positive thing come out of the incessant wrangling on Wikipedia's talk pages. Not one! OTOH, I have seen much much arrogant preening, bitter & cutting talk, etc.
  • Having said that (ahem!) I'll make one and only one comment, then disappear. Giving bureaucrats more leeway etc. is just a terrible idea, and I don't mean to be impolite. If bureaucrats are circumspect now, it's because they don't have anything resembling carte blanche. I'm thinking that down the road the results of this suggestion would be terrible because:
  1. There are always and everywhere people who seek power for its own sake (not talking about current bureaucrats.. repeat after me.. down the road..)
  2. When such people see power vacuums created (by the granting of new powers to a class or group of people) they rush in to join that class or group.. so a couple years down the road, poof!, you have cabal.
I'm done talking now. Not gonna talk any more. Direct all replies to each other. :-) --Ling.Nut 12:07, 18 January 2007 (UTC)
Why we are having this discussion here, here and here at the same time? -- ReyBrujo 13:09, 18 January 2007 (UTC)
  • I agree that discussion about keeping/deleting editor review or admin coaching should be kept there. There are however some things of interest from those that are relevant here. I've been reviewing some of the editor reviews, and I find myself absolutely astonished by some of the comments. Look for example at Wikipedia:Editor review/SunStar Net 2. One of the reviewers is suggesting ***6000*** edits before going up for RfA, with 1000...1000!... Wikipedia space edits. Unreal. Another reviewer there is suggesting at least 1000 mainspace edits. That number for the mainspace is apparently is becoming vogue now. I've seen it used on a number of occasions. Sad. Very sad. I also find it very sad that some RfA contributors are increasing focus on ratio of edits by namespace. For example, the first oppose vote on Newyorkbrad's RfA. Some people have lost focus on what being an admin is about. --Durin 13:45, 18 January 2007 (UTC)
    It would help against that problem if people who haven't lost focus on what being an admin is about vote on as many RfAs as possible. Many of our RfAs have ridiculously low participation. Kusma (討論) 13:55, 18 January 2007 (UTC)
    It would also help if they responded to opposition (or support) votes that they consider spurious. There's a strange reluctance to call out people who are evaluating candidates using unhelpful criteria (wheras no such reluctance exists at, say, AFD). Responding more vigorously to the problem may help reduce it. Christopher Parham (talk) 18:02, 18 January 2007 (UTC)

One possibility is that the concept of improving as an editor (and possibly being more suited to being an admin) is becoming too focused on editor review and the concept of an admin school. There are many different ways of improving as a Wikipedia editor. Recently I tried to list some of the more interactive methods at Wikipedia:Coaching. Maybe that would be a good link to have in the "If you are unsure about nominating yourself for adminship, you may wish to try ... first." bit? i.e "If you are unsure about nominating yourself for adminship, you may wish to try one of the options at Wikipedia:Coaching first." Though I've always been more in favour of the concept of learning by yourself, and improving by experience. Indeed, I was recently complimented on this quote:

"Enculturation really just needs people to talk more and demonstrate how they do things (rather than just doing them). Takes certain types of people to be role models. Actually following someone's edits, or meeting in person and watching how they do things, can be very instructive." (Carcharoth)

My point is that there are many ways to learn. If we work to diversify the options, that might help any one area degrading and becoming unacceptable.

Right, now I'm off to post this in the other threads. Please focus this thread on specific things that RfA can do to help. Carcharoth 14:34, 18 January 2007 (UTC)

In response to Proto's original post, I agree that the !voting based on numbers is harmful. To the extent that it causes potential admin candidates to run up their numbers (e.g. by carelessly adding to XfD's) it's bad for the encyclopedia. However, I'd oppose giving bureaucrats more power to decide. They're very reasonable people, but they're still people. It's difficult for even the most reasonable people to differentiate between an argument that's invalid and one that they jsut disagree with. I realize WP is not a democracy, but making it less democratic could mess up an important balance. And if we make a standard that a certain rationale for !voting is invalid, people are just going to put down other reasons and still really !vote based on numbers. I like Kusma's idea; I've certainly considered supporting a candidate just to cancel out some opposes I thought were unfair (I didn't though, since I wasn't sure if that would be ethical). I don't know the solution to this, but I'd strongly suggest leaning toward the discussion side (e.g. if you think a !vote is based on editcountitis, bring it up) rather than the enforcement side. delldot | talk 17:42, 18 January 2007 (UTC)
You can't reply to every oppose !vote on an RfA - that just makes things worse. It's a discussion but you can't actually discuss in case more people oppose. Who said RfA isn't broken ? —The preceding unsigned comment was added by Heligoland (talk • contribs) 17:59, 18 January 2007 (UTC).
If you've already opposed the candidate for some other reason, I don't see how anyone could object to objecting to opposes for some other reason. Candidates can sometimes get into trouble for responding to opposes, but people unrelated to the candidate should be able to point out a spurious oppose without affecting the candidate's chances. On the other hand, the support votes that don't give a reason make things even worse (but if they were all discounted per WP:ILIKEIT, where would that leave RfA then?) --ais523 18:04, 18 January 2007 (UTC)

Isn't this a case of trying to treat symptom rather than cause? Many RfA !voters apply satistical criteria- numbers of edits or proportion of edits. People ask for reviews and mention an interest in adminship. Reviewers are logically going to tell people how to adjust their statistics to improve their RfA chances. The problem here is with the reasons why people vote in RfAs but there is nothing wrong with editor review per se. WJBscribe -WJB talk- 18:18, 18 January 2007 (UTC)

  • I think there's plenty of problems with Wikipedia:Editor review. I'm in discussion with people over there in an attempt to get them to revamp what editor review is about. It's editor review...not editor-review-with-an-eye-towards-RfA. Yet, it's being treated like the latter rather than the former. --Durin 18:33, 18 January 2007 (UTC)
    • And I've posted a note at Wikipedia talk:Admin coaching that they really should change the name. Really. Editor coaching is a perfectly fine thing to do, and a perfectly neutral name; in fact, renaming might encourage mid-level editors to volunteer to coach newbees who wants one-or-one coaching. John Broughton | ♫♫ 19:49, 18 January 2007 (UTC)
There's already an Editor coaching. Admin coaching isn't really the same thing. Firsfron of Ronchester 02:35, 19 January 2007 (UTC)
Pre-RFA coaching would be more accurate in every sense. Proto::► 14:32, 19 January 2007 (UTC)
No, it wouldn't. No one has suggested a name that a) makes it clear that Admin Coaching's purpose is not to just get someone to pass RFA; b) that Admin Coaching is not basic editor coaching, but rather to polish a user's knowledge of policy and administrative processes; c) is not used already; and d) is reasonably-easy to remember. Titoxd(?!?) 23:54, 20 January 2007 (UTC)

[edit] Activity on this talk page

I was spinning some data around, as most of you know I do from time to time and came up with some numbers that some might find interesting. Over the last year, this talk page has seen (including anon IPs) 689 unique contributors. 50% of the traffic on the page came from 50 most frequent contributors. Just some numbers. I thought the 689 number was rather higher than I expected, though 183 of those had 1 edit total and 380 had less than 5. Bureaucrats accounted for 5% of the traffic. --Durin 18:30, 18 January 2007 (UTC)

How do you spin these numbers? A little black box somewhere? I could probably extract the same sort of results, but do you have a quick way of doing this sort of thing? Any nifty tools? Carcharoth 01:38, 19 January 2007 (UTC)
Sounds too low. I get 930 users, with 59 anons. I get my results from Aka, who by the way says that you're the one who talks here the most.[6] Titoxd(?!?) 05:58, 21 January 2007 (UTC)
  • My report is constrained to the last 365 days. --Durin 14:14, 21 January 2007 (UTC)

[edit] Suggest changing WP:RFA to links instead of transclusions

I don't know if this has been brought up previously, nor what the result, if any, was. However, sometimes WP:RFA can become large and cumbersome, such that it would be more practical if there were just plain wikilinks to each individual RfA (ie. [[Wikipedia:Requests for adminship/Yuser31415]] instead of {{Wikipedia:Requests for adminship/Yuser31415}}. Feedback? Yuser31415 22:15, 18 January 2007 (UTC)

That depends. Sometimes there's as few as two, sometimes as many as twenty. I think the contents box should suffice. You can see any users you like to vote on and just click their name. --Majorly 22:23, 18 January 2007 (UTC)
This is just a thought, but what if we put the votes section into <noinclude>'s? It would make the length of the page much more consistent, and still useful -- you'd still be able to see the current status and the user's answers to questions. -- Renesis (talk) 22:32, 18 January 2007 (UTC)
I have implemented a test version of my suggestion in my sandbox; feel free to comment or edit it. Yuser31415 23:30, 18 January 2007 (UTC)
I like the idea of putting votes in a noinclude more than just linking like Yuser31415 did. I have no real problem with the way it is now though. --Matthew 00:40, 19 January 2007 (UTC)
I'd also be happy with that idea (even if just the headers were included). Yuser31415 01:04, 19 January 2007 (UTC)
There's no way to mock it up, so let's get some more input and then we can just try it out. -- Renesis (talk) 01:07, 19 January 2007 (UTC)

(undent) Well, if the names of candidates aren't in the TOC in the upper right corner (and they aren't on the version I just looked at), it's a non-starter (in my opinion). The names have to be "above the fold" - people scan that when they come to the page and are going to detest scrolling down. John Broughton | (♫♫) 01:34, 19 January 2007 (UTC)

[edit] Fifteen current RfAs - what is the record?

The numbers seem to flucuate a lot. Recently as low as two (or even one) and now up to fifteen. What is the record for the most number of open RfAs at one time? Carcharoth 11:36, 19 January 2007 (UTC)

This was discussed above (on this very page).--Kchase T 11:40, 19 January 2007 (UTC)
Oops. Thanks. I'll read 5 talk archive pages as a penance. Carcharoth 13:13, 19 January 2007 (UTC)
"Hail RfA, full of grace. Jimbo is with thee. Blessed art thou amongst editors, and blessed is the fruit of thy editing, Jimbo." Hmm. Just doesn't seem to quite roll off the tongue, does it. --Durin 14:32, 19 January 2007 (UTC)
And potentially blasphemous as well :D. On a more serious note, the question was aised above "how many people have you nominated recently," that is one of the reasons I took the plunge and nominated someone. I think it is fair to say that the conversation above contributed mightily to the current plethora. -- Avi 14:38, 19 January 2007 (UTC)

[edit] Suggested withdrawals

RfA will always get inexperienced users nominating themselves not realising that they're destined to be turned down. Often their RfA's are closed early and it's probably quite demoralising. When I see one I try to leave a note on their talk page explaining why but I don't see every occurance so I'm suggesting an RfA subpage or template that would contain an example note. When an RfA is closed early due to obvious inexperience it can be easily explained to the user. Note that it would not be for any RfA that is unsuccessful, just those who would get moral supports to prevent pile-on opposition.

I whipped up an example template (try substituting User:James086/RfA withdraw into a sandbox or see the effect in mine). Of course the wording and name are subject to change. Any suggestions, comments, opposition? James086Talk 06:34, 20 January 2007 (UTC)

I like it, the wording is a little rough in places, but I think it's a geat idea. It also needs a funny picture, templates with pictures are more fun. (I mean really, that stop sign on the test4 just makes it that much more fun to use right ;) ) Seriously though, I think you've got a great idea. --Matthew 07:39, 20 January 2007 (UTC)
Mentioning Jimbo to a newbie might confuse them. Carcharoth 07:44, 20 January 2007 (UTC)

I went ahead and tried to streamline the wording, reducing awkwardness. It's still a little bit rough: I'm concerned about overwhelming the user with a ton of links to admin processes. Perhaps it should be put in a bullet point list like how {{welcome}} is done? Hbdragon88 08:35, 20 January 2007 (UTC)

I'm not sure. I think that gives me a sense of "grooming" admins from scratch, telling them exactly what they would be told at WP:ER or admin coaching, both of which are currently under review. WP:ENC, and we shouldn't be asking new users to focus on admin tasks, imo, when there's an encyclopedia to write. – Chacor 08:54, 20 January 2007 (UTC)

  • Great idea. Current wording pushes all the right buttons for me. Deizio talk 11:17, 20 January 2007 (UTC)
This is an excellent idea, and the wording is fine. While obviously some RfA candidates with zero experience are making what amounts to a vandal-type application, quite a lot a serious and sincere, and do not recognise their incompetence. I do not see any way in the software that they could be warned in the actual process of formulating their application, and given that, a gentle template as suggested is clearly the best option. If the software could be changed to show a "minimum requirements" suggestion before the RfA was submitted, that would be even better. Is this possible? I recognise that this then opens up a discussion on what "minimum requirements" are.--Anthony.bradbury 13:46, 20 January 2007 (UTC)
I'm with you on this, it's an excellent idea, I don't think there's anything incompetent about a new user wishing to become an administrator, just a willingness to help which should be firmly embraced (I know, silly idea, but hey). Half the trouble is that there's not a clear source of how many months of editing or the number of edits you typically need to pass an RfA. I suspect most administrators and I suspect a lot of editors who have had an RfA were nominated, so go have a look at how self nomination works, and you'll notice there's nothing clearly stating you'll probably need to have x months and y edits to pass an RfA. I don't know if having some of NoSeptember's charts displayed would help (I think it might) and imploring the user to do an edit count first (yes, I know, urgh - editcountitis but it exists and a candidate needs to be aware of it). If the candidate gets no support or only moral support there needs to be a nice, fluffy template that embraces the candidate and let them know that some administrators they've come across needed 1, 2 or even 3 (or more) RfAs before the community promoted them and to keep contributing to the mainspace by writing articles as well as doing whatever it is they do (vandal fighting, spam cleanup, closing XfDs etc). I'll see if I can rustle something up. --Kind Regards - Heligoland 14:20, 20 January 2007 (UTC)

An excellent idea (though your tempate needs a little grammar cleanup). Of course it would be even better if we hand typed them rather than turning civility into a subst:ed template, but it's better than nothing. --tjstrf talk 13:51, 20 January 2007 (UTC)

One thought: the current list of suggested activities seems a little burdensome, as a candidate might interpret it as a requirement to do ALL of those things. Perhaps a shorter list or alternate wording could be devised that seemed less daunting? --tjstrf talk 13:57, 20 January 2007 (UTC)
I've revised the template, sharpening the wording. I added a link to the "What adminship is not" page, but perhaps that is overkill - other opinions appreciated. As for the list of activities, the "such as" in front of the five activities makes the list fine, I think, because it's harder to read it as all five being required.
More generally, I think this template is a great idea. If we're going to have 10 to 15 hopeless RfAs a month, some standard wording to such candidates, asking they withdraw, is extremely useful. -- John Broughton | (♫♫) 14:47, 20 January 2007 (UTC)
Not sure about "almost totally without exception". Is it almost or totally? Has there been a sucessful RfA of someone who didn't have that level of experience? WJBscribe -WJB talk- 14:51, 20 January 2007 (UTC)
I'd keep the 'almost totally' wording, as otherwise it would need to be changed if someone did pass with very little en-experience. The classic example would be an editor with lots of experience on another WMF project, who might just pass if someone with good standing around here made a passionate plea that such a user can be trusted here as well. Also, those who use this template should add a personalised touch, and be prepared to follow-up and talk to the person involved. The process doesn't end with the use of the template, it starts there. Just leaving a template and walking away would be impersonal and rude. Add a 'handwritten' note as well. There aren't that many inexperienced editors trying RfA that the time has to be saved this way. Use the template as a guide to what to say, and of course, subst it. Carcharoth 16:02, 20 January 2007 (UTC)

I did say I would rustle up something User:Heligoland/Sandbox1 - Any thoughts ? --Kind Regards - Heligoland 16:09, 20 January 2007 (UTC)

Are you sure putting Tawker's name there is a good idea? --Majorly (talk) 16:12, 20 January 2007 (UTC)
I like this one as well. It's a little more formal and template-ish though. As for mentioning Tawker, I think the idea is that he's a highly respected admin who makes a good example. I can't believe it took someone 7 tries to pass RfA? Who was that? --tjstrf talk 16:17, 20 January 2007 (UTC)
Shockingly enough, it was Jaranda. When I saw that I was surprised myself.--Wizardman 16:23, 20 January 2007 (UTC)
(In response to Majorly, edit conflicted about a dozen times) - I think finding a couple of high profile admins RfA profiles will really reassure good editors who have RfA'd too early, we would do those concerned the courtesy of asking though. --Kind Regards - Heligoland 16:38, 20 January 2007 (UTC)
Yeah I think it would be nice to ask first. The template looks fine btw. --Majorly (talk) 16:44, 20 January 2007 (UTC)
I've done a copyedit on that as well. I hope you'll excuse my changing the "1 to 3 months more experience before you try again" to "several months", but I'm concerned that the template will simply encourage people to apply again in what now appears to be a minimum - 30 days - figuring that the worst that happens is they have to do it six more times before they are (ta da!) an admin. -- John Broughton | (♫♫) 19:10, 20 January 2007 (UTC)
I'll make a few changes to the template myself, just to make it more accurate ... Yuser31415 22:59, 20 January 2007 (UTC)
Done. I was WP:BOLD and moved User:Heligoland/Sandbox1 to Template:RfA withdrawal. Yuser31415 23:09, 20 January 2007 (UTC)
Now let's see what it looks like in action. As for people re-applying too soon, add something to say that reapplying too soon won't work, and it is best to ask for advice before re-applying (or being nominated). Carcharoth 23:43, 20 January 2007 (UTC)
A link to an editor review might be a good idea, something like, "before reapplying, you may wish to try an editor review first" and the template should probably protected, just in case we do got someone not happy about their RfA. --Kind Regards - Heligoland 00:10, 21 January 2007 (UTC)

[edit] Talking of templates...

I've made Template:RfA talk, which can be used on the request's discussion page. It automatically links to wannabe Kate tool, the user's page and produces the date. The edit stats can be pasted in as normal. What do you think? --Majorly 13:59, 20 January 2007 (UTC)

I'd use something more like this:
View [[User:{{BASEPAGENAME}}|{{BASEPAGENAME}}]]'s edit summary usage with [http://www.math.ucla.edu/~aoleg/wp/rfa/edit_summary.cgi?user={{BASEPAGENAME}}&lang=en this tool]; to view this user's edits on Wikipedia, see [http://tools.wikimedia.de/~interiot/cgi-bin/Tool1/wannabe_kate?site=en.wikipedia.org&username={{BASEPAGENAME}} Interiot's "Wannabe Kate" tool].
However, mine has a bug: it can't cope with usernames containing a space. Yuser31415 22:39, 20 January 2007 (UTC)
Well, the edit summary link is already on the page itself. There isn't much else different, apart from the fact the edits aren't displayed, and {{BASEPAGENAME}} stays the same. Mine can deal with spaces as well. --Majorly (talk) 22:43, 20 January 2007 (UTC)
Oh, I see what you're doing now - placing the template on the RfA discussion page. Discount mine :). Yuser31415 22:53, 20 January 2007 (UTC)

Need more comments :-) Is it worded OK, should it be moved to template space...? --Majorly (talk) 23:57, 20 January 2007 (UTC)

Why not? :) Be bold, and use it! Yuser31415 00:13, 21 January 2007 (UTC)
Thanks Yuser, for being the only one to comment =) I'll move it now. --Majorly (talk) 00:16, 21 January 2007 (UTC)
That's not fair, I wasn't online so I couldn't comment. So when using it would you put
{{subst:RfA talk|James086|copied stats from wannabekate}}
or am I missing something? James086Talk 04:12, 21 January 2007 (UTC)
That's right. --Majorly (talk) 11:38, 21 January 2007 (UTC)

Yuser31415, with respect to your bug of not coping with names containing a space: If the user name is "Joe Schmoe", {{BASEPAGENAME}} will give "Joe Schmoe" and {{BASEPAGENAMEE}} (note the extra E) will give "Joe_Schmoe" to be more friendly to external links such as yours. For example: Requests_for_adminship-- Renesis (talk) 17:09, 22 January 2007 (UTC)

Oh cool :) - now I know something I didn't before :). Yuser31415 19:40, 22 January 2007 (UTC)