User:Useight/RFA Subjects/No big deal

Admin actions can be undone, so? (Archive 43)

• In most of these discussions, I feel uncomfortable when people harp on the point that "all admin actions can be undone and so, it is no big deal." However, the last thing that we want to do is spend time on undoing admin actions as we have several urgent and important things to be done. I guess that is the reason why some voters vote "better-safe-than-sorry" the idea being that it is a cost avoidable. --Gurubrahma 05:26, 13 December 2005 (UTC)

  How many admins have gone rogue so far? These are new requirements people are adding, so what problem is being solved? Wearing a football helmet all day might be 'better safe than sorry' too... - CHAIRBOY (☎) 05:31, 13 December 2005 (UTC)

• Even if fully 1% of users trusted according to my lax criteria stated above went amok, they'd still be outnumbered 99:1 by responsible admins. Sure, undoing their actions would take a bit of time under the current software, but not too much more time than it would take if they went amok in a non-admin fashion (such as by moving pages instead of deleting them). They would just be slightly more harmful, if the admin were truly malicious (range-blocking the Internet, anyone?). But either way, I'm not aware of a single admin out of several hundred who have done any such thing so far, and if we went my way and doubled or tripled the admin numbers overnight, I doubt we'd see that rate increase much. —Simetrical (talk) 02:47, 14 December 2005 (UTC)

There has been one case on en where a missunderstanding resulted in an adminin appearing rogue. There have been two cases on other projects. In all cases the admins was delt with in pretty short order.Geni 09:41, 14 December 2005 (UTC)

• The idea that everything an admin does (except image deletion) can be undone is also flawed from another perspective. The impact that a well-meaning but errant admin can have on other users is not so easily undone. Whether we have rogue admins or not isn't really something we can state with certainty. We have essentially no feedback mechanisms for admin performance. --Durin 13:41, 13 December 2005 (UTC)

We have the same feedback mechanisms for admin performance as we have for anything else: RFCs, RFArbs, etc. Those aren't 100% effective, maybe, but what could a rogue admin do without being spotted? I still feel that I had a run-in with a well-meaning but errant admin once, but I started an RFC and the community judged the admin to be in the right. —Simetrical (talk) 02:47, 14 December 2005 (UTC)

That's not precisely feedback. An admin can intimidate, bully, or bite subtly without crossing the line into Rfc territory - remember 2 users have to have tried to work the same problem out with the Admin prior to Rfc - and what newbies or even middlebies (yes horrid neologism, late for work humor my bad phrasing) know about it? Feedback implies good as well as bad, positive as well as negative experiences, and does not have to fit such narrow criteria. Once and admin hits Rfc (legitimately) he's often annoyed quite a few people. One puppy's opinion. KillerChihuahua^?!? 12:59, 14 December 2005 (UTC)

However it is quite hard for an admin to behave seriously badly without clashing with other admins.Geni 13:29, 14 December 2005 (UTC)

Not quite true. If you take a look at the Deletion log, you'll find dozens of articles being speedy deleted each day for reasons which go well beyond WP:CSD. CSD:G1, for example, is applied very liberally by some admins, and when the article is recreated it is promptly removed (incorrectly!) based on CSD:G4, often by the same misguided admin. Actions like that do not result in clashes with other admins, but I know at least one case where it resulted in a loss of a well-intentioned editor; I have no doubt there were many others. This is one of the reasons why I believe bureaucrats on en-wiki should have the power to de-sysop. With over 600 admins, many of which are new, we need a sizeable group of experienced users to watch over their actions, looking for systematic mistakes or abuse of power. I don't see de-sysoping becoming an everyday event, but the presence of more bureaucrats inspecting admin actions, and having the power to desysop, will surely raise admin standards. Owen× ☎ 03:49, 16 December 2005 (UTC)

It is also the case, I have recently found, that 24-hour blocks commented as "vandalism" are unlikely to receive much scrutiny, if any. If an admin were to ignore blocking policy, defining "vandalism" in a broad manner and blocking new accounts without any warning or discussion, the community does not neccessarily pick up on this quickly. —Bunchofgrapes (talk) 04:05, 16 December 2005 (UTC)

What is RFA for? (Archive 71)

This talk page is ... puzzling. Adminship is supposed to be "no big deal", as pretty much all admin actions are reversible; the idea is that if you're not going to go batshit with the tools, there's no reason for you not to have them. (I've actually lowered my personal bar for adminship after a couple of cases where people I had severe qualms about got their admin bit and proceeded to do okay with a bit of peer pressure the collegial advice of others.) The social and technical skills required are about those of a message board moderator, and whereas there are adults you'd never want to have that job, the skills are easily acquired by e.g. a sensible teenager.

And particularly this year, the ArbCom has had no compunction in removing the admin bit from those who need it removed. Removal is a big deal, but the AC is there for the social decisions that are big deals.

So the ever-expanding lists of requirements don't make sense. The lists above appear to be marks of the ideal admin, who is a bit like the ideal editor with added technical powers (and similarly doesn't exist). And this stuff really doesn't have a lot of bearing on whether they are likely to go batshit with the admin tools. They make adminship into a much bigger deal than it should be.

I personally think most people need three months' experience to get a feel for the place. But beyond that, in an ideal world every Wikipedia editor who's been around enough to get a feel for the place would have admin powers.

Please enlighten me as to why it is good for adminship to be a big-deal artificially scarce commodity - David Gerard 09:56, 6 October 2006 (UTC)

See my suggestion along a similar line on WP:BN. 3 months and say 1000 edits. And a presumption that the admin bit should be granted, unless "material" issues are identified. Plus a simple recall mechanism. -- ALoan (Talk) 10:34, 6 October 2006 (UTC)

It's important that:

1. The RfA-dwellers look like they're doing more than rubberstamping RfA candidates. They need to look like they're "protecting" us from unsuitable admins.
2. The trolls and idiots can always hide behind WP:CIVIL when somebody points out to them, in eloquent and factual language, that they're batshit insane.
3. We constantly have backlogs, so the current sysops can feel as if they're relied upon.
4. The RfA-dwellers feel they have some control and influence on the project
5. The RfA-dwellers don't have to write articles.
6. The current admins can feel important as they're of fewer number.

QED. — Werdna talk criticism 10:24, 6 October 2006 (UTC)

I'd challenge points 3 and 6, possibly caveat them with 'The admins who continue to hang around navel gazing central after being flagged'. Other than that I think it sums it up ;) ALR 10:57, 6 October 2006 (UTC)

The practice the Arbitration Committee has adopted of using "administrators" as administrators of mechanisms such as probation is probably part of the problem. Considerable maturity and discretion is required. These mechanism were adopted to give us some alternative to banning, but have the effect of transferring day to day responsibility to administrators. With more power comes the concern that those who exercise it are responsible. Fred Bauder 11:12, 6 October 2006 (UTC)

Administrator_s_ - the college of admins. Individual admins are constantly subject to peer review, so errors can be quickly spotted and dealt with (althougn the scars may not be so easy to heal). There is plenty of noise about admin abuse, but WP:AN and WP:ANI are reasonably effective, I think. A recall mechanism would help too. Adminship is not a badge - it is a set of tools. -- ALoan (Talk) 11:29, 6 October 2006 (UTC)

I'm trying to think of a recall mechanism that wouldn't be a troll magnet - David Gerard 11:55, 6 October 2006 (UTC)

I guess using Wikipedia:Requests for arbitration is out then. Fred Bauder 12:00, 6 October 2006 (UTC)

With that logic, we don't even have a promotion mechanism that wouldn't be a troll magnet. --badlydrawnjeff talk 12:08, 6 October 2006 (UTC)

Absolutely. I think we need to set the bar reasonably high - 5 (or 10) admins agreeing to the recall. Then we dump the person back into my new version of RfA (i.e. only material issues cause a failure). -- ALoan (Talk) 12:08, 6 October 2006 (UTC)

"troll magnet" -> "wikilawyer playground" - assume any process that can conceivably be gamed will be gamed to destruction - David Gerard 12:20, 6 October 2006 (UTC)

Show me any process that can't be gamed. -- ALoan (Talk) 13:17, 6 October 2006 (UTC)

There are of course none :-) But some are more prone than others. If there is to be a mechanism, it needs the Hard Decisions Committee at the far end. Therefore it needs to be something to make their lives less busy, since committees don't scale. (When the AC was created because Jimbo doesn't scale, Wikipedia was #500 on Alexa. Today it's #12. Out of all websites of any sort.) - David Gerard 13:26, 6 October 2006 (UTC)

We already have an up-and-running recall mechanism which is not a troll-magnet and (so far) is not gamed in any way. In my opinion it runs far more smoothly than the ArbCom. Haukur 13:25, 6 October 2006 (UTC)

As discussed above on this page, it doesn't seem to mean a whole lot - David Gerard 13:26, 6 October 2006 (UTC)

• The problem with voluntary recall is that the (few) admins for whom recall would actually be necessary could simply refuse to volunteer. AOR is weakly worded and easily gamable: an admin whose standards are "if at least twenty people with at least 7000 edits each complain, the admin will voluntarily submit to ArbCom proceedings" could be said to be open to recall even if his standards are so ludicrous as to make it a practical impossibility. >Radiant< 13:31, 6 October 2006 (UTC)

• Yes, and in theory a freely editable encyclopedia wouldn't work. The recall process works fine, it has a bunch of high-profile admins in it with real committments. Would it be even better if it was a bit tighter? Sure, I think so, anywhow, but even just what we have is proof that recall procedures work and are not just troll-bait. Haukur 13:53, 6 October 2006 (UTC)

As I see it RFA has become a focus for those who appear to be social networkers more than content producers. Actually achieving the standard now seems to involve; lots of low risk edits, clearly lots of minors rather than anything substantive and running away from anything contentious because any conflict regardless of its management because conflict itself is am mistake. It's also becoming the classic vicious circle, as more of the social networkers get through then they're more likely to hang around rather than move onto actually working on WP.

In some ways the more effective potential admins aren't going to get through, it's clear that being an effective admin will rub some people up the wrong way (hence the issues around re-adminning), if someone hasn't done that before coming up then they haven't actually demonstrated the capability. Catch 22.

It may be that in some areas WP has now reached the critical mass where a quasi-democratic consensus approach is wholly inappropriate and should be more positively gripped to ensure effective delivery of output.

I'm sure there are reasons why all this navel gazing is useful, I'm damned if I can think what they are right now.

ALR 13:28, 6 October 2006 (UTC)

Going back to David's original question, I think he's making several incorrect assumptions:

1. Admin actions are not entirely reversible. (Or, more precisely, they're reversible only in the technical sense.) Any number of recent incidents have shown that people will get very upset—and justifiably so—if an admin mistreats them, even if the actual use of the tools (the block, the protection, the deletion, etc.) is quickly undone. This goes double for blocking; being blocked is a big deal, and having everyone shoot from the hip because the blocks are "reversible" doesn't help things at all.
2. Admins not only need to avoid doing bad things with the tools per se, but also to avoid doing bad things in general (c.f. "Administrators are held to high standards of conduct, as they are often perceived as the "official face" of Wikipedia. Administrators must be courteous, and exercise good judgment and patience in dealing with others."). Some people seem to do fine with the tools themselves, but are so abrasive otherwise that they don't really need to use the tools to cause conflict.
3. And, finally, we have had some cases of admins going "batshit with the tools"; the extent to which the ArbCom's solutions in those cases have been successful is open to debate, I think.

Kirill Lokshin 13:36, 6 October 2006 (UTC)

In response to (2), admins are said to be held to very high standards, but the mechanism by which they are held to these standards and consequences for breaking these high standards are rather unclear, and both seem to boil down to being chided by other admins who in turn get chided by yet other admins for chiding the standard-breaker. >Radiant< 13:42, 6 October 2006 (UTC)

I've been thinking about the whole 'admin as leader' issue and whilst I see it as a valid argument in the current climate, it's one that's easy to change culturally quite easily. By reverting to a state where adminship is no big deal by reducing the ridiculous numbers of hoops that candidates have to go through then admins will be less likely to be perceived as having some form of special status.

There is the additional benefit that increasing the availability of admin facilities should encourage a more reasoned level of debate in some areas, because scurrying off to get an admin to arbitrate will be a less viable option. It might actually encourage use of the mediation and advocacy schemes that bubble along.ALR 16:52, 6 October 2006 (UTC)

Possible causes

• First, while it is obvious that we should have some standard for admins, it is equally obvious that it is unclear what that standard should be, and there are widely different opinions on that. The lack of a clear standard encourages people on RFA to make up their own standard; since it is a large wiki, people will want to vote for candidates they do not know, and need a standard since they cannot fall back to familiarity with the candidate. The result is that some people tend to pick an arbitrary amount of months or edits (etc) and oppose everyone who falls below.
• Second, I think it's safe to say that some people have raised their standards because they perceive the unlikelyhood of admins being demoted if they turn out to be (perceivedly) unsuitable, except in extreme circumstances. As an example, RFA will strongly oppose a candidate for mild incivility, but the ArbCom does not generally do anything about incivility except warn against it. (I'm not saying either standard is correct, I'm pointing out the discrepancy). See also this comment about a perceived discrepancy between the ArbCom's standards and the RFA voters' standards.
• Third, it is obvious that factions exist on the wiki, e.g. groups of editors with contrary POVs, who have a tendency to oppose candidates from the other faction. Since every oppose-vote counters four support-votes, this is reasonably effective.
• The first could be solved by establishing a clear line for experience (e.g. 3 months, 1000 edits) and make it clear that arbitrary criteria are discouraged. The second could be solved by drawing up a consensual standard of adminship, and indiscriminately promote all that pass it, as well as demote all that fail it (this is the tricky part). The third could be solved by dropping the perceived 80% bar to something more closely resembling other consensus-based processes.
• >Radiant< 13:31, 6 October 2006 (UTC)
  • An additional point - the lack of an "official" minimum requirement for admins (e.g. x months y edits) leads to many nominations by enthousiastic but inexperienced novice users, which are (rather obviously) opposed a lot. This could easily lead people to think that RFAs in general are opposed a lot. >Radiant< 13:41, 6 October 2006 (UTC)

I'd suggest that another cause is that people in general will be less likely to vote support (and purlease don't start into its not a vote). People frequently seem to oppose for the most trivial of reasons, including perceived incivility, but won't support if they see no reason not to oppose. Huge number of support votes only appear to come on widely pimped candidates.

It just occured to me that the whole issue of civility or perceived incivility is a minefield given the number of social cultures, and age differentials, within which editors exist. As a Scot I'm cynical and use sarcasm and irony a lot I also have an extremely dry (arid) sense of humour, frequently I think that my SoH and use of irony/ sarcasm is lost on a great many editors and I've had one or two apparently teenage US colonials get really upset about flippant offhand comments. tbh if I could be bothered I feel sure that I wouldn't get through RFA.ALR 13:44, 6 October 2006 (UTC)

It's pretty overreacted to, too. Administrators need to be able to communicate calmly and in accordance with policy. That's All. Manytimes, an editor will be opposed based on one incident, or a couple of harshly-worded comments. Civility is to make sure we get along, not to give trolls and idiots policies to hide behind when somebody points out that their arguments are crap in a decent and factual manner. I personally believe that the only incivility that's an issue is incivility that rises to the level of a block. The word is thrown around far too much, and this needs to stop. — Werdna talk criticism 14:38, 6 October 2006 (UTC)

I've thought about a self-nom on RfA. Then I actually read the page. No way I'll put myself through that crap. Why bother? I can revert vandalism and warn vandals manually, though I guess the shiny buttons would help. And it would be nice to have the good housekeeping seal of approval. But otherwise it just isn't worth the irritation of some...uh, user (I'll be civil)...voting me down because I don't have 300 edits in Wikipedia talk or whatever.

A little while back, I put up an imaginary RfA on this talk page as a test. Sure enough, a frequent RfA voter took the bait and told me I didn't have enough edits in "process" or something. (You can read the unintentionally hilarious details on my user page.) Since the imaginary RfA was just a goof, I could smile at such silliness. But if it had been a real RfA, I probably would have gotten ticked. And who needs that?

Not to mention that some people would start pulling diffs from my user page on my comments about Signpost stories, or the Cyde quote, or something else somewhere in my thousands of edits, and then I'd really get sourpussed. So I'll just edit my articles and skip the aggravation. Casey Abell 16:12, 6 October 2006 (UTC)