Jump to content

英文维基 | 中文维基 | 日文维基 | 草榴社区

User talk:Tony1/AdminReview

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Coordinators

Just a brief idea/question on the matter of coordinators. First, coordinators will serve for some defined period, yes? A year, three months? Also, as to the selection, here's my proposal:

Coordinators are made up of two sitting arbitrators (selected at random from willing participants) and three non-administrator editors with a history of participation on Wikipedia.

My proposed definition of "participation" would be at least six months active as an editor on Wikipedia (possibly even a year) and at least 1000 edits (with possibly 50% of those being in mainspace). Editors would be able to opt out by placing themselves in a category (via their userpage) or via a list maintained as a subpage of the process page. Coordinators would also be free to reject their random nomination (and be included in the pool for selection next time, unless they opt out prior to the next random drawing) or resign at any point during their tenure (in which case another editor/arbitrator (as appropriate) would be randomly selected). Once a coordinator has served out their allotted time in the process (a year, three months, whatever) they would be ineligible for selection until the random pool had been exhausted. While we could have an RFA-style system of selection for coordinators I think it would be better to do it at random. Yes it's very instruction creepish, but any other method (especially RFA style) would lend itself to abuse I think. —Locke Coletc 10:30, 19 December 2008 (UTC)[reply]

  • Yes, because issues of police misconduct should be investigated/decided/resolved by... police[this is sarcasm]. Having said that, I wouldn't object to one of the 5 coordinator positions being an administrator (2 arbitrators, 1 administrator, 2 editors). —Locke Coletc 10:40, 19 December 2008 (UTC)[reply]
  • Wow, go off line for a couple of days and you never know just what's gonna happen. On the one hand I can't believe that anyone would nominate this project for deletion… on the other hand, it is all too believable (sigh). I'd like to go on record here stating my opposition in the strongest terms against any admins also serving as coordinators of the soon-to-be-renamed "AdminWatch". Absolutely positively no sitting admins, for crashingly obvious conflict-of-interest reasons. Having been an admin or arb in the past should not disqualify, however. Not any editor can be a coordinator of this project. If the consensus is that an editor requesting admin status must have upwards of 3000 edits and a minimum of one year on the Wikipedia, the requirements for a project coordinator should be upwards of 5000 edits and a minimum of two years on the WP. The type of Wikipedian that I am picturing for a post as coordinator would be, for example, The Fat Man Who Never Came Back, who recently garnered more than 50 percent supports in his (unsuccessful) candidacy for ARBCOM.
  • Just a note, most of you reading here already know this, but Malleus Fatuorum is the main author of WP:WIKISPEAK, my numero uno favorite page on the entire Wikipedia. Go check it out if you haven't already (smile).--Goodmorningworld (talk) 12:35, 19 December 2008 (UTC)[reply]
  • So, at the first approximation, none of our existing admins can be trusted to be neutral? And all other editors can? And this kangaroo court would be run based on that basic assumption? Wow. ➨ ЯEDVERS takes life at five times the average speed 12:44, 19 December 2008 (UTC)[reply]
  • Speaking here strictly as an administrator, and thus as a second token member of The Enemy, I'd just like to say (i) that yes, WP:WIKISPEAK is intercoursingly good, and also (ii) "Wow" (which I haven't said much recently), though not about anything in particular. -- Hoary (talk) 12:49, 19 December 2008 (UTC)[reply]
  • Seems like Locke is suggesting a jury service model. Whilst remaining in brainstorming mode, I feel that being an administrator per se should not be a criterion for disqualification. I think good Admins ought to be interested in seeing that "bad boys" do not tarnish the good overall standing of admins, so they also have a stake in the process. It is important that the process must be reasonably free of any "them and us" sentiment, and should certainly avoid people power evolving into a lynch mob. Determining a set of desirable qualities for coordinators may be a better way forward than picking people at random. If appointments will take a 'random' route, I'd like to consider how to avoid appointing known trouble-makers as coordinators. Ohconfucius (talk) 13:41, 19 December 2008 (UTC)[reply]
  • Ohconfucius, while I don't think that originally Tony meant it to be so, this page has taken on the "them vs. us" mentality. If you go through the discussion there are several references to "civilians" that would have been more neutral as "non-admins". Locke Cole, if the coordinators are drawn at random what happens when you get an editor that hates, for whatever reason, admins as a group, even if they have never had any dealing with the particular admin they are investigating? What happens if I am brought up on the page and then discover that one of the coordinaters is someone that I blocked several months ago? There are other problems with getting coordinators at random but without mentioning names it's a little difficult to discuss. I would think that a general vote on the coordinators would be a good thing with a pool available so that any with a conflict may recuse themselves leaving others to take their place. CambridgeBayWeather Have a gorilla 16:00, 19 December 2008 (UTC)[reply]
  • I'm afraid I must echo the comments made by Cambridge and Redvers; both have very valid concerns. ScarianCall me Pat! 17:06, 19 December 2008 (UTC)[reply]
  • Presumably the other four coordinators would balance it out if one were drawn that "hated all admins". As for individual issues (such as having a coordinator who had been blocked by an admin) we could either allow them to recuse themselves, or just assume good faith that the other coordinators will again balance it out. I certainly wouldn't have any issue with potential conflicts of interest being brought up (so if a coordinator had been blocked by an admin being discussed it would be appropriate to note that and ask if they wish to recuse themselves). I'm not sure if forced recusal in such a situation would be a good idea or not. —Locke Coletc 21:31, 19 December 2008 (UTC)[reply]
  • Agreed that this page has taken on the "them vs. us" mentality. However, we are all grown-ups hoping to make WP a better place, so let's all try and put the MfD and related acrimony behind us. Ohconfucius (talk) 02:00, 20 December 2008 (UTC)[reply]

The issue about 'someone that I blocked several months ago' is valid but not unique to Adminwatch as a judicial process. How is it currently handled? Lightmouse (talk) 17:10, 19 December 2008 (UTC)[reply]


Tony1's comments: Redvers, I'm disappointed you have such a negative take on this. A kangaroo court, according to Encarta (quick on my desktop) involves judgement "without good evidence". WP has grown up and needs proper checks and balances to maximise goodwill among its participants. I can assure you that people will be doing their best to make it function in a professional manner, including the NPOV assessment of evidence; otherwise it will fail. Please consider supporting it rather than constructing yourself as "the enemy". Far from it: we rely on you and your colleagues to protect us from chaos. I know many admins who would do a good job as coordinators, but that's not the point: justice has to be not just done, but seen to be done, and unfortunately, that means most or all coordinator positions should be occupied by non-admins, to minimise perceived conflict of interest.

Locke raises valuable points above, and has brought us closer to determining a viable system of recruiting coordinators so that the initiative can be put to the community. (I concede that this needs to be done and the process moved to main space at an earlier stage than I'd first planned.) The following basic questions arise from Locke's post, comments by Ohconfucius and Goodmorningworld, and my own thoughts. I seek input to these questions below. Tony (talk) 17:18, 19 December 2008 (UTC)[reply]


How many coordinators should there be?

  • Six Six seems like a good number. Three-to-two non-admin majority is too slender, IMO. This process has to be guaranteed not to be taken over by admins, although admitting their participation at the same time. I would be more comfortable with four-to-two; there are 165 active users for every active admin. If people think six is too many, I'm OK with five, if the majority is four-to-one. Tony (talk) 17:18, 19 December 2008 (UTC)[reply]
    • NB, the idea is that single coordinators take on each case, not the whole five or seven judge each case by vote: that would be impractical on a number of levels. Can you imagine the bureacracy and cross-communication required to make every decision in concert for every case? Impossible. This is why I don't understand the need for an odd number of coordinators. They won't be voting as a block. Tony (talk) 03:30, 20 December 2008 (UTC)[reply]
  • Five. We are looking to recruit the best and brightest, and there won't be that many who meet the requirements and are willing to serve; remember we also need to keep a few people in reserve to replace coordinators who resign partway through their terms. Also you want an odd number so as to prevent deadlocks.--Goodmorningworld (talk) 17:31, 19 December 2008 (UTC)[reply]
  • An odd number would make hung decisions less likely. Lightmouse (talk) 17:32, 19 December 2008 (UTC)[reply]
  • (ec) Must be an odd number. Also there should be a backup pool in case one is required to recuse themselves. CambridgeBayWeather Have a gorilla 17:36, 19 December 2008 (UTC)[reply]
I was expecting that the coordinators would take cases on themselves, and that while they might talk with each other from time to time, they wouldn't vote as a block.Tony (talk) 17:40, 19 December 2008 (UTC)[reply]
That was also my expectation. My feeling now is that this process is being railroaded down a potentially rather unwholesome path. The MfD failed, let's try instead to destroy it from within. Colour me cynical if you must. --Malleus Fatuorum 23:18, 19 December 2008 (UTC)[reply]
I note that as I write this there have been seven people who have commented on the number of coordinators. Of those seven Redvers and myself are admins. That means that Goodmorningworld, Lightmouse, Locke Cole, VX and Realist2 are all part of some scheme to railroad the proposal? I think that is rather insulting to those users, along with Redvers and myself, especially as the majority of them expressed keep comments at Wikipedia:Miscellany for deletion/User:Tony1/AdminWatch. I would suggest that rather than 5 non-admins and 2 admins trying to railroad the process that they were not clear on whether the coordinators were to work alone or together. Read through this page and there are comments that indicate both possibilities. CambridgeBayWeather Have a gorilla 06:59, 20 December 2008 (UTC)[reply]
  • Your use now of the word "euphemism" appears to be deliberately provocative, as is your continued use of the term "kangaroo court". However, you are of course perfectly entitled to your opinion, so long as you continue to state it civilly. --Malleus Fatuorum 23:39, 19 December 2008 (UTC)[reply]
  • If I'm not being civil, please let me know where I've slipped up and I'll apologise immediately - it certainly wasn't deliberate! And I didn't mean to be provocative: my opinion is that the word "judge" applies for the role under discussion. Others prefer "coordinator". We both know what the role involves, so we need a title (a better euphemism, if you will) to apply to our new (er, overlords? Overseers? Oversighters? Magistrates? Rulers who I for one welcome? ;o) people-in-charge-of-this-process-however-they-are-selected. ➨ ЯEDVERS takes life at five times the average speed 23:54, 19 December 2008 (UTC)[reply]

I think on a standard issue there should be 5. If the issue is of utmost community importance 7 should participate, giving the outcome more authority. — Realist2 01:28, 20 December 2008 (UTC)[reply]

We could call them 'AdminWatch jurors' 'AdminWatch adjudicators', I s'pose. But I also tend to believe that repeated use of the word "euphemism" is a bit pointy if not downright derogative. If we have defined majorities required (as suggested by Tony above), the odd number criterion would have little importance because deadlock would be impossible if 4:2 (for example) was not reached. Ohconfucius (talk) 01:56, 20 December 2008 (UTC)[reply]
Again, they won't block vote at any stage. This is a multistage process for each case. Coordinators will manage single cases. Tony (talk) 03:39, 20 December 2008 (UTC)[reply]
Tony, I think that you need an extra section here. It appears that there is some confusion over the coordinators working alone or together and you need to ask what others think. I for one had assumed that it would be like Arbcom and the coordinators would work together. CambridgeBayWeather Have a gorilla 06:59, 20 December 2008 (UTC)[reply]
After reading the "cab-rank" process down below, I think we'll need more co-ordinators, possibly even 10! Although I still think we should have an admin/non-admin share for diplomatic reasons, I can't see the practicality of it now if the cases are gonna be handled by just 1 co-ordinator. There might be a conflict of interest there, and if there isn't, then one of the parties will probably still see it that way anyways. Ryan4314 (talk) 17:39, 20 December 2008 (UTC)[reply]

Status qualifications?

Should there be status qualifications for the positions (i.e., non-admins/admins)? If so, what should be the composition of non-admins and admins?

  • This is the one deal-breaker for me: ordinary users should be in the majority, and thus status qualification is an essential criterion (later insertion: I now favour a five-to-one or four-to-two non-admin majority; out there it's 165-to-one). It seems risky to reserve a position for an arbitrator when none of the 13 may be willing to take on a double task; there may also be a conflict of interest if ArbCom makes rulings on AdminWatch when one of its members is in both groups. If there are six coordinators, I'm thinking of 4/2. It seems important to be include at least one admin as a show of good-will, and because admins tend to be good at process and tend to know the system well.Tony (talk) 17:18, 19 December 2008 (UTC) Tony (talk) 00:24, 20 December 2008 (UTC)[reply]
    • I could see two seats out of five being reserved for former admins and no limit on the number of former admins elected, but I would not want even a single serving admin to do double duty so as to avoid conflict of interest. To current admins: Do not take this personally, it is a separation of powers kind of thing. The same goes for arbitrators, who anyway have enough on their plates already.--Goodmorningworld (talk) 17:35, 19 December 2008 (UTC)[reply]
  • I'm repeating my initial suggestion here for the sake of discussion: my composition would be two arbitrators and three regular editors. An alternative to that would be two arbitrators, two editors and an administrator. But keep in mind that most arbitrators are already administrators as well. —Locke Coletc 21:35, 19 December 2008 (UTC)[reply]
  • Depends on the number of coordinators there are. Admins should be in the minority (2 out of 5, or 3 out of 7). All others should be non-admins. VX!~~~ 23:16, 19 December 2008 (UTC)[reply]
  • I don't mind if it's Arbs or Admins, only that they wont have the majority. So, if we have a group of 5, only 2 should have the tools. I don't think the ex-admin thing is a good idea, that'll be a conflict of interest i.e. how do they lose their tools? Ryan4314 (talk) 10:22, 20 December 2008 (UTC)[reply]

Other qualifications?

Should candidature involve an explicit acknowledgement of the appropriate skills—namely, the analytical, NPOV, interpersonal and writing skills? What minimum number of edits is appropriate, and what minimum time as a WPian?

Method of selection?

  • If being an administrator is (according to Jimbo last I recall) "not a big deal", then being a coordinator here should be even less of a big deal. As soon as you start voting on things though, it becomes a "big deal" whether you wanted it to or not. —Locke Coletc 22:53, 19 December 2008 (UTC)[reply]
  • As below, I can't imagine a viable system of put-your-hand-up and be randomly selected, as much as I agree with Locke's motive in making it "a small deal". Three issues come to mind: (1) the skill-base and NPOV stance of candidates needs more prominence and scrutiny than this; (2) the system would need to be re-run every time a new coordinator was needed (relief, replacement, expansion ...); (3) randomness is not easy to apply, and someone would have to be trusted to apply it. At least an election is open. My analogy is not with the dreadful RfA system, but with ArbCom elections, but much much scaled down. Tony (talk) 23:57, 19 December 2008 (UTC)[reply]
  • Again, that indicates that approval voting is the way to go. Everyone stands and the one(s) with the most votes becomes a judge/coordinator. That will mean that there are issues with competence (for want of a better word) but then we just need to vote on what level of competence we require in a judge. As I've said, I'd go for anyone being able to stand. Let the people here decide on who they want to run the court. How would it harm? ➨ ЯEDVERS takes life at five times the average speed 00:08, 20 December 2008 (UTC)[reply]

Terms?

Should coordinators have fixed or indefinite terms; if fixed, for how long?

OK, initially two years for half and one year for half, staggered ... I hadn't thought of it as a board. Tony (talk) 17:43, 19 December 2008 (UTC)[reply]

Dismissal?

Should coordinators be subject to dismissal by ArbCom?

Reserve list?

Should there be a reserve list of candidates in case a coordinator leaves the project, needs to reduce their commitment for a period, or is dismissed, or if there are insufficient coordinators for the task?

Coordinator powers?

What powers will the coordinators have to sanction administrators found to have acted contrary to WP:ADMIN policy?

  • None, at least not in the way of sanctions imposed. The authority of coordinators stems solely from the judgment exhibited by them in holding up administrator conduct to WP:ADMIN policy mixed in with a healthy dose of pragmatism.--Goodmorningworld (talk) 19:40, 19 December 2008 (UTC)[reply]
  • None, obviously they would be able to initiate an arbitration case for circumstances that require it, but this is true of any editor (coordinator or not) and would be subject to the same acceptance by the ArbCom. —Locke Coletc 21:49, 19 December 2008 (UTC)[reply]
  • None. They are kangaroo judges taking part in a kangaroo court. Giving them powers of any sort would be asking for trouble. ➨ ЯEDVERS takes life at five times the average speed 22:49, 19 December 2008 (UTC)[reply]
  • None. All points above are valid. VX!~~~ 23:23, 19 December 2008 (UTC)[reply]
  • Redvers, just as giving all admins blocking power without proper oversight isn't asking for trouble? Riiiight. The coordinators would have no tools or formal power, but would derive their authority from the system in which they work (structured evidence, analysis and recommendations), their skill and NPOV stance, and consequently the trust of the community. (That, by the way Redvers, is the only way individual admins can gain trust and respect.) There are analogies; for example, featured-content reviewers who consistently show skill and NPOV gain considerable authority in those processes. It's a very wiki thing. Tony (talk) 23:30, 19 December 2008 (UTC)[reply]
  • The question is surely rather: "Do none of the 1,500 current admins just bluntly use their tools without judgement?" If you believe that none do, then there is no need for this AdminWatch process, I quite agree. That's not what I believe though. --Malleus Fatuorum 23:45, 19 December 2008 (UTC)[reply]
  • ArbCom is itself a very blunt weapon, and one of last resort. The intention of AdminWatch, as I understand it anyway, is to avoid the need for ArbCom requests for desysopping whenever possible, by giving the community a forum in which legitimate grievances can be recognised and hopefully dealt with, with as little unnecessary drama as is possible. It may also make ArbCom's job easier, in that a case can be more easily and quickly assembled if that ever becomes necessary. --Malleus Fatuorum 23:56, 19 December 2008 (UTC)[reply]
  • Exactly. WP needs a smarter system of nipping in the bud any trouble between admins and other users, before it blows out of proportion. ArbCom should be spared, wherever possible, fine-grained stuff like this, to concentrate on the bigger picture stuff. And just a small point: IMO, the most successful admins are those who value the "R-word"—rapport; the trouble is, rapport takes time, effort and skill. While some misbehaviour our admins have to deal with is idiotic and doesn't deserve the R-word, some does. When it happens, it's magic to the project. Tony (talk) 00:12, 20 December 2008 (UTC)[reply]
  • Well, ArbCom can be used as both a sledgehammer and as a scalpel. Historically, the community has been uncomfortable with either, but up in arms with judgments that try to do both. But we're trying to establish a court to judge admins, not a court to judge ArbCom, aren't we (I may, in my defence, have missed something here)? Which are the admins we have that use their tools without judgment? They need to go in front of ArbCom now, regardless of this debate. ➨ ЯEDVERS takes life at five times the average speed 00:14, 20 December 2008 (UTC)[reply]
  • To characterise AdminWatch as some kind of "court", kangaroo or otherwise, is disingenuous. It would better for you to think of it more in terms of an arbitration process, the purpose of which is to avoid the need for any court. Admins misuse their tools every single day, and as a result sometimes drive good faith editors from the site. But nobody is interested in the occasional or sporadic abuse of tools; it's easier to get that camel through the eye of a needle than it is to have an admin desysopped for what many might perceive as minor abuses of power. AdminWatch is simply an attempt to redress the balance, to support those editors who have legitimate complaints, and to hopefully help resolve those complaints before anyone has to be dragged before ArbCom, or another good editor is driven from the site. --Malleus Fatuorum 00:32, 20 December 2008 (UTC)[reply]
  • It might work, and if it doesn't, then all it means is that Tony's wasted some time. If there's one lesson Esperanza and the Award Center have taught us, it's that Wikipedia is generally very good at getting rid of processes that aren't working as intended. Cue a long string of "but what about Arbcom? RFA? FAC?" replies, I know… – iridescent 00:36, 20 December 2008 (UTC)[reply]
  • I see what you're saying, Malleus, but you - and others - have said clearly that there are certain admins who are abusing their tools now. Why are we talking here about setting up processes to manage a process, to manage the process to pass on processes to another process, when there are admins who are abusing tools now? We should be naming them in front of ArbCom. This is very serious. ➨ ЯEDVERS takes life at five times the average speed 16:39, 20 December 2008 (UTC)[reply]
  • If it is your counter assertion that there are no administrators abusing their tools or their position, then if true that will lead to the best possible outcome, no need for AdminWatch to do anything at all. --Malleus Fatuorum 20:34, 20 December 2008 (UTC)[reply]
  • Answering my question with a question is a great debating technique, but it's also not answering my question. You, and others, have said that there are some admins that are abusing their tools now. This is serious. Very serious. You and others have asserted it several times. We, the community, need to stop them now. Please, name them in front of ArbCom. Really: this serious accusation needs to be dealt with immediately for the good of the 'pedia. ➨ ЯEDVERS takes life at five times the average speed 21:34, 20 December 2008 (UTC)[reply]
Sorry to bump an old thread but Wikipedia:Deadminship suggests at least 45 cases where the problem with admins has been so severe they've received the ultimate sanction. In a population of 1600 over several years, I'm happy that's not very many, but arguing that as a class, admins are infallible seems folly, and I'm sure there are other problems there which might not be so severe as to be cause for desyssoping, but remain nevertheless unchecked in the current system.--Joopercoopers (talk) 16:36, 22 June 2009 (UTC)[reply]
  • Oppose: I can see this as being the biggest point of contention over getting this project accepted in the community. Lets get it up and running first, and if it works the community might wanna enable powers, they might not. Ryan4314 (talk) 10:47, 20 December 2008 (UTC)[reply]

Conflict with other venues?

How will the new project avoid straying into the territory of other venues for resolving editor complaints about admins, i.e., ARBCOM, Administrators' Noticeboard, Administrators' Noticeboard/Incidents, RfC/U on individual admins?

  • As to ARBCOM, conflict is minimized because the project has no power to impose remedies. For the other three, either a written guideline delineates respective competences, or there is a partial merger, or the respective competences are left to sort themselves out over time.--Goodmorningworld (talk) 19:40, 19 December 2008 (UTC)[reply]
  • AN and AN/I were always, as far as I know, about reporting incidents or issues relevant to administrators, not necessarily reporting administrator conduct. ArbCom is for situations which have spiraled out of control or for ongoing conduct issues. This would serve as a good first step in dispute resolution. —Locke Coletc 21:55, 19 December 2008 (UTC)[reply]
  • This kangaroo court duplicates RfC's powers and remit, so must defer to an RfC as the process immediately above it. RfC comes below AN/ANI, where administrators and editors deal with rouge admins without the shrubbery and AN/ANI come below ArbCom, which comes below Jimbo who comes below the Foundation. ➨ ЯEDVERS takes life at five times the average speed 22:53, 19 December 2008 (UTC)[reply]
  • I'm sorry to cut across your über-hierarchical view of the world, full of its notions of "must defer to" and "deal with" and "comes below": this process doesn't need to be slotted into such a pyramid—it works through trust and skill, just as does good collaboration in content writing on WP. Tony (talk) 23:34, 19 December 2008 (UTC)[reply]

I've moved Redvers's topic on whether coordinators will need "judge's clerks" to his/her talk page as not relevant, IMO. Tony (talk) 00:03, 20 December 2008 (UTC)[reply]

No, you moved it to an obscure talk page one of my subpages, where it was not relevant and I had to delete it. You clearly don't want to talk about this subject (I don't know why not - others did and clerking issues are relevant) but that's fine. I won't edit war with you. If you don't want discussion, just say so. It's your project in your userspace, after all. ➨ ЯEDVERS takes life at five times the average speed 00:20, 20 December 2008 (UTC)[reply]
Oh, and it's "his" talk page. But that's fine too :o) ➨ ЯEDVERS takes life at five times the average speed 00:21, 20 December 2008 (UTC)[reply]
  • It wont, this project has niche, in that it will be majority non-admins. Tony, post a link to where Redvers removed comment is, if anyone does want to discuss ir, then they can simply follow that link, that should shut him up ;) I would presume if this project is accepted by Jimbo etc, they'll move it to it's own page. Ryan4314 (talk) 10:51, 20 December 2008 (UTC)[reply]
Ryan, he says he's deleted it. That is why I haven't posted a link. I'm happy for him to do so right here if he re-instates it on his preferred talk page (I did click the "five times" link to relocate the text, which was reasonable, I believe). Tony (talk) 11:25, 20 December 2008 (UTC)[reply]
Ooh, not quite: the "five times" bit leads to an essay, so you then had to click a talk link, then cut-and-paste to a page marked in really big letters User talk:Redvers/Say no to Commons. It's a remarkable bit of editing-without-reading! :o) But I understand, you don't want to talk about this subject, you don't want it discussed, so it must go. I don't agree, obviously, but this is your baby, your userspace, your process - I'd be churlish to argue too much. I'm sure you know what you're doing :o) ➨ ЯEDVERS takes life at five times the average speed 16:44, 20 December 2008 (UTC)[reply]

The "cab-rank" model: clarifying and seeking your comments

The design hasn't changed in essence, and I apologise for any misunderstanding; it shows how devising a system can blind you to potential ambiguities. As soon as people started talking of avoiding even numbers of coordinators, I should have realised my failure to spell it out properly.

The two-stage design for the passage of each case is designed to be as simple as possible; but it is too lengthy and involved for committee decisions and block voting at each stage. Requiring all five, six or seven coordinators to liaise at each point and write group determinations (per ArbCom decisions) is not practical—ArbCom takes ages to make decisions, and it's a big deal, because they're all supposed to be involved. Involving every coordinator in every case here would be utterly impractical, inflexible, lengthy and onerous for what needs to be a simple and prompt process. What if some coordinators were too busy in a particular week, or had a conflict of interest that precluded their involvement in a case? No parties would be satisfied with an ad hoc committee-like membership on this basis.

No, AdminWatch is conceived as a cab-rank. Coordinators drop by regularly and take on a notification where they feel there is no perceived conflict of interest, and have the time and capacity to handle it. There may be room for some specialisation, e.g., in page deletion issues, block issues, where emotions are particularly fractious and need to calmed—but that is up to the individual coordinators, and should not be essential to the success of the process. This model bears an uncanny resemblance to the way judicial advocacy profession works ... and WP's mediation system, if I'm not mistaken. It's not rocket science, and to me seems like the inevitable design. The only difference that certain circumstances may require a change in coordinator during the course of a case.)

The core of the multistage design has not changed:

Stage 1: Is the complaint:

  • (a) trivial or vexatious (delete promptly), or
  • (b) prima face worth proceeding with (move to Stage 2)?

Stage 2: After assessing the comments and rejoinders of both sides in Stage 2, is the complaint:

  • (a) not a breach after all (place in the temporary "Recent cases" section with a note exonerating the admin—regrettable and will require a sympathetic message to the admin, but it may occasionally happen), or
  • (b) a breach, and if so, how significant?

For both 2(a) and 2(b), place in the temporary "Recent cases" section with the result plus conclusory comments and recommendations.

Minimal communications by the coordinator via the parties' talk pages will be necessary as a courtesy, to alert the parties to milestones (mostly by standard paste-ins). Coordinators may, rarely, need to seek further information or point out where the rules of AdminWatch are at issue (personal comments, irrelevant tracts of commentary). On rare occasions, third-party comment may be admissible with the approval of the coordinator (on application via the coordinator's talk page).

I'd appreciate comments on the structure, now that the "single coordinator" approach has been clarified. I'm hoping that this will be accepted as the practical way to go, but I remain open to criticism and suggestions. Tony (talk) 11:25, 20 December 2008 (UTC)[reply]

This is certainly a simple model. I like it. Lightmouse (talk) 12:05, 20 December 2008 (UTC)[reply]
Like it, nice one. I'm gonna change my above comments. Ryan4314 (talk) 17:31, 20 December 2008 (UTC)[reply]

Eligibility to vote

The latest draft of the project page states, "coordinators are elected by popular vote, in which any Wikipedian with at least 1000 edits and an account at least one year old may vote". That seems too restrictive to me. If I remember right, the requirement for voting in the ArbCom election is 150 mainspace edits. I can't see why this should be different for electing the "AdminWatch" "coordinators" (and by the way, we really need to decide on a better name for the project than "AdminWatch" which is too juvenile-sounding, and a better title than "coordinators": if they manage a case by themselves they are not really coordinating a team, are they?)--Goodmorningworld (talk) 13:49, 22 December 2008 (UTC)[reply]

Let's try to think of better names for both. The ideal time to make the change to the page-name is when it's put into public space. "Coordinators" should be changed as soon as possible if a more appropriate term is available. Tony (talk) 14:12, 22 December 2008 (UTC) PS Eligibility: I was lost for numbers and will insert your suggestion, pending comments by others. Tony (talk) 14:21, 22 December 2008 (UTC)[reply]

Proposal: a new protocol for incivility warnings

See my talk page. Tony (talk) 04:14, 7 April 2009 (UTC)[reply]

Discussion at ANI regarding an Admins Noticeboard

In the above discussion I noted the existence of these pages. If there are any (former) participants in the drawing up of this model still watchlisting, this may be an opportunity to revitalise this process and possibly expand it into something with more teeth. LessHeard vanU (talk) 20:22, 12 June 2009 (UTC)[reply]

Statement here. Tony (talk) 05:41, 14 June 2009 (UTC)[reply]

Please sir, what's the difference between "civil" and "very civil"?

Question is in relation to the following from the proposal, namely

"Notes:
  • (3) Your statement
  • Be succinct: state your case in 300 words or less.
  • Include only facts that are strictly relevant.
  • No personal comments; be very civil."

Very civil? How will complainants demonstrate they are very civil and not just civil? Incidentally, the intro proposes seven coordinators but this says six. Kaiwhakahaere (talk) 21:52, 13 June 2009 (UTC)[reply]

Good points. I changed to 250 words consistently. "Very" is not only included, but italicised; I think it's necessary to stress the need for civility and the minimising of emotional input. Tony (talk) 03:53, 14 June 2009 (UTC)[reply]

Tony, I thought it was obvious, but guess I need to ask again.
Very civil? How will complainants demonstrate they are very civil and not just civil?
You didn't answer that question. You specified "very civil". What is the difference between civil and very civil. If there isn't a guideline, how will the six coordinators (or is it seven?) assess that a complainant is only civil and not very civil? Will a complainant who is civil but not very civil have their complaint ignored? C'mon, you wrote the "very civil" stipulation so please define it. Kaiwhakahaere (talk) 11:01, 14 June 2009 (UTC)[reply]
My understanding of the term is that very civil is standard civility standards being vigorously applied - i.e. that no incivility will be tolerated. I really do not understand, however, your concern with the wording - it a simple emphasis of a core conduct issue. LessHeard vanU (talk) 11:08, 14 June 2009 (UTC)[reply]
Civil means no incivility, so your point, "no incivility will be tolerated", is moot. My understanding of the English language is that the word very is an important qualifier. Why specify "very civil" when "civil" would suffice? If the coordinators ever say to an editor "you are not very civil", and then are challenged, how will they explain the difference between "civil" and "very civil"? Would it make a difference to the required standard if the word "very" was deleted? If yes, then please explain the differences in standards. If not, then please delete it. Kaiwhakahaere (talk)
This is Tony's page, so he has the final word, but I suggest that the use of "very" is not as a qualifier, but as an emphasis. Deletion of the four letters may, or not, occur when a consensus forms. LessHeard vanU (talk) 12:08, 14 June 2009 (UTC)[reply]
  • Strictly speaking, Kai is right; this is exactly what I say to people who use "very". I guess here it's what is known as an interpersonal item. I'll remove it, since it will be clear that coordinators will remove anything uncivil or emotive. That in itself will be a first for a WPian process. Tony (talk) 16:27, 14 June 2009 (UTC)[reply]
  • I just thought that since I have this watchlisted, I might drop off a few thoughts. I noticed at WP:CIV on the talk page, it's mentioned that civility is "subjective". I personally don't agree with that. I think that often people mistake WP:CIVIL with WP:NICE, they aren't the same things. In a sense, I see a distinction between uncivil, and incivil - that being the "un" part simply is not extended any exceptional friendliness, and the "in" part be a "breech" of civility. Well, that's just my 2-cents worth, I see I have a lot to read up on here - there's a couple things going through the twisted cobwebs of my mind, so I'll try to sort them out and become somewhat active with this. Cheers to all. — Ched :  ?  06:30, 21 June 2009 (UTC)[reply]
Thanks, Ched (congrats on your successful RfA, too). Can you and others advise on my proposal for codified and cleaned-up civility categories; I feel as though it's a ghost town over at that talk page. Should I post a centralized-discussion note to drum up feedback, or is it such a no-brainer that no one is bothering to respond there? Tony (talk) 13:37, 22 June 2009 (UTC)[reply]

Progress

Tony, what's needed to get this on it's legs? What help do you need? --Joopercoopers (talk) 13:13, 22 June 2009 (UTC)[reply]

Tony was held up by the Arbcom date linking debacle. Hopefully, he should be back on track soon. Dabomb87 (talk) 13:23, 22 June 2009 (UTC)[reply]
Good grief - what a waste of everyone's time! --Joopercoopers (talk) 13:44, 22 June 2009 (UTC)[reply]
Yes, there has been a time-sink of a distraction. I need to develop two things to put to users here for feedback: (1) how to deal with multiple complainants/admins; (2) the voting method for electing coordinators.
Then, we need to review the whole thing and think about whether it needs community-driven support to launch (in project space, of course). A respected ArbCom clerk informed me a while ago that, in his/her view, technically speaking it doesn't need an RFC. Tony (talk) 14:07, 22 June 2009 (UTC)[reply]
Not sure about that, there's a bit of flap on at the moment about self-elected groups without consensus. I'll have a think about multiple complainants. Presumably there's an 'out of the box' solution for voting method in a similar vein to our other functionaries. --Joopercoopers (talk) 14:14, 22 June 2009 (UTC)[reply]
  • Thanks. Peter D's proposal was self-elected, of course, whereas coordinators here would be community-elected. On the one hand, AdminReview has to have community support, and does need at least not to have the active disapproval of ArbCom. But I'm determined that it not be bulldozed over—without a chance to prove itself—by people who just don't like it. Multiples are a hard think, aren't they; it's already elaborate. Voting method: yep, it should be off the shelf, with the uniqueness that candidates will have to declare that they have the requisite skills. I'd like to know that coordinators can be removed—by ArbCom? But this goes to the heart of whether AdminReview is to have a formalised relationship with ArbCom (it doesn't have to). Tony (talk) 15:08, 22 June 2009 (UTC)[reply]

Thoughts about multiple complainants

  1. Look for harm. If user:Daisy Duke and user:Jesse Duke complain the admin user:RPtrain unfairly blocked Daisy. Only Daisy might have a case because no harm was caused to Jesse. Can cases be brought by anyone even if they are not harmed? If they are blocked, how will this be achieved? (see below)
  2. Look for Community harm and related timescales. During an edit dispute at Dodge Charger, admin user:RPtrain unfairly blocks user:Daisy Duke for misconstrued incivility and admin user:BossHogg protects the page contrary to policy (say amidst accusations he's paid by Dodge to edit). user:Jesse Duke complains that both actions are abusive. Jesse has no case against RPtrain because the harm was to Daisy (that's her case). Jesse does though have a case against BossHogg, because of the wider harm to the community. However, because the incident is related in time and place (Dodge Charger and within 24hrs say) and shares parties with a prima facie case, it's efficient for the review board to look at both actions together.
  3. Hazzard County articles are plagued by rival POV pushers identified in previous Arbcom sanctions as The Dukes and The Law. user:RPtrain and user:BossHogg (both identified with The Law faction) use admin tools in a multi user dispute that involves blocks to both sides, page protection, accusations of wheel warring, potentially de-sysopping offences and wild accusations of impropriety from both sides. All coordinators convene and sift the complaints, dismiss the vexatious, and present a full RfAR case to Arbcom, less the vexation.

Thoughts about the process

I like the idea of a lightweight process, but it needs some checks and balances.

  1. An awful lot of responsibility and power is concentrated with the sole managing coordinator at the moment. It might be better to have a system where, as cases come in, whichever coordinator picks up the case, then becomes the managing coordinator for that case.
  2. The idea of delegating decisions to one or two coordinators is good, but other coordinators may feel that a particular case needs the scutiny of the whole group. Should we imbed the idea that coordinators have a right to 'call-in' any case at any stage so that it may be discussed as a group?

Thoughts about blocked users

  1. Unblock requests are a separate set of processes at WP and this review board should not conflict with that. For blocks of up to a week, I'd amend the main page to require complainants to submit a complaint within 7 days of expiry of a block, rather than the admin action. For those blocks over 7 days duration, if unblock requests have been unsuccessful, then the blocked user may request a strict conditional unblocking to particpate in these proceedings at the 7 day mark. Absolutely any violation of that agreement will incur not only a reblock for the full duration again, but the immediate dismissal of the case.

--Joopercoopers (talk) 15:10, 22 June 2009 (UTC)[reply]

Jooper, thanks heaps for this input. I can't do it justice until tomorrow. Tony (talk) 17:36, 23 June 2009 (UTC)[reply]

Dispute resolution and archiving

Tony, thanks for all the time, effort and thought you've put into this proposal. I hope to see it reach fruition in the near future. There are a couple of small points I'd like you to consider:

  1. Dispute resolution has a hierarchy of processes that are clear for content disputes, probably less clear for behavioural disputes. Where do you see this process sitting in the DR process for behaviour? Should this position be spelt out explicitly in the proposal?
  2. The fact that old cases are to be archived is spelt out in "Recent cases", but the text in the three stages (at B, D, E, G, H) doesn't quite say that - in four cases it says "deleted", so would it be better to explicitly replace "deleted" with "archived" at those locations?

Good luck with taking this forward - I think this proposal addresses an area of DR that is presently sorely lacking - and, as always, please let me know if there's any way I can help with your efforts. --RexxS (talk) 22:09, 8 July 2009 (UTC)[reply]

RexxS, thank you very much for your supportive comment. I must put the effort into moving this forward soon. Tony (talk) 00:39, 9 July 2009 (UTC)[reply]
I think that would be a good idea Tony - additional to the above I think the idea needs to be either a. simplified or b. explained better - probably with some kind of flow chart. I know the idea is to achieve an efficient process, but it does read a little in the vein of "forms must be stamped in triplicate by the departmental under-secretary before being passed for scrutiny and approval to the relevant section overseer.........." --Joopercoopers (talk) 09:47, 11 July 2009 (UTC)[reply]
If it's any help, I'll list here the processes that cover dispute resolution of behaviour as I understand it now:
  1. Discuss the problem with the other user
  2. Wikiquette alerts
  3. Requests for comment/User conduct
  4. Administrators' noticeboard/Incidents
  5. Arbitration/Requests
It seems that Mediation is aimed primarily at content disputes, but maybe it should be included.
I believe that AdminReview would actually span/replace numbers 2–4. Since RFC/U requires two users to certify, it is not immediately applicable to a single user with a grievance about an admin. Similarly, it appears difficult in practice for a user with a grievance about an admin to see AN/I as a neutral venue. Therefore I suggest that AdminReview would be a parallel path for those who find 2–4 inappropriate. I guess this is all obvious, but perhaps it needs to be spelt out for those who are new to DR? --RexxS (talk) 12:40, 11 July 2009 (UTC)[reply]
The above links seem to have to relate to user conduct, but reminder that there is also third opinion. Dabomb87 (talk) 16:24, 11 July 2009 (UTC)[reply]

Name redux

I think the new name is good to address concerns that arose from the previous "watch" (=stalk) associations. Unfortunately, if this ever goes live (I personally have previously commented in favor of such a mechanism, so I hope it will), there will be confusion with Wikipedia:Administrator review which exists already. As such, I fear another name change is needed. Something equally neutral and short, like "AdminCheck" or "Administrative Action investigation" (similar to WP:SPI's naming) - you people probably can come up with a better name than I can. Regards SoWhy 15:26, 11 July 2009 (UTC)[reply]

Re Stage 3, Option H

Where admin abuse has been determined there is a further potential option that may present itself - please see User:Roux/RFA-reform where a process currently named XfA (a method of desysopping) is being drawn up for consideration. This could be an addition to the option of recommending RfC/RfAR in cases of findings of extreme or large scale admin abuse. In my brief comment on the talkpage I suggested that a case forwarded from this page may be both an approved method of initiating the process, and also certain criteria determining the validity of the request might be dispensed with. People watching this page may wish to comment there in any case. LessHeard vanU (talk) 20:22, 18 July 2009 (UTC)[reply]

I think there is an obvious synergy here. As part of the XFA proposal, I have suggested (and feel strongly) that some outside party needs to certify a complaint before it can move forward to an actual request for desysopping. The proposal currently requires two uninvolved members of any of ArbCom, AC clerks, crats, MedCom, admins to certify. Philcha raised a concern (invalid IMHO, but whatever) that somehow the 1700 people who qualify to certify could/would sweep issues under the rug. Having AdminReview as an alternate avenue to certifying a desysop request would obviate that concern. → ROUX  20:51, 18 July 2009 (UTC)[reply]

Comments

Although I am yet undecided on whether or not this process is a good one or not, one flaw the page has is that it is too crowded with information and very difficult to navigate. This is before there are even any reviews on the page. I do recommend that the page is broken off into subpages. Not every piece of information needs to be here. Seddσn talk|WikimediaUK 00:38, 23 July 2009 (UTC)[reply]

I agree; in fact, I'd decided similar already. It's a matter of knowing what to put where. In addition, should each case be put on its own /subpage, or would it be enough to put each section on a /subpage? Tony (talk) 02:53, 23 July 2009 (UTC)[reply]
I think the subpage-with-transclusions (possibly collapsed?) method works well enough in so many other areas (SPI, AfD, RfA, RfB, MfD...) that it would make perfect sense here. Also guarantees persistent permalinks. The 'how to' should also be /subpaged probably. → ROUX  03:00, 23 July 2009 (UTC)[reply]

In case you missed it...

... there is now Wikipedia:Appeal Committee. A proposed body that seems to duplicate this proposal. Enter CambridgeBayWeather, waits for audience applause, not a sausage 13:08, 30 July 2009 (UTC)[reply]

Not exactly a duplicate, more a parallel operation. AR looks more at Admin abuse of the flags in relation to an editor or event(s), where the Appeal Committee is to determine if a particular sysop action was done properly (that is, well or badly but in good faith). LessHeard vanU (talk) 20:36, 30 July 2009 (UTC)[reply]

A new initiative by admin Ched. Participants here might consider taking a look! Tony (talk) 04:09, 5 October 2009 (UTC)[reply]

Plans for this project?

I'm not sure, Tony, if you were still planning on following through with this project. If you were, then I was hoping that, once established, to push for formal recognition of this forum as an oversight committee over administrators. Cla68 (talk) 01:50, 6 October 2009 (UTC)[reply]

Cla, (1) I think it should have teeth; (2) I think it should be an official subcommittee of ArbCom (with coopted members); (3) I'm nervous about finding the right people to be coordinators via election, and about the potential pitfalls in setting it up.
I almost trust myself to run it, but I don't want to—it's very administrative, rather like being an arb, which is a necessary but awful job. ArbCom deals with behaviour (only), so would be the natural place to locate it. I think ArbCom should appoint one or two of its number as coordinators, and admins and non-admins for the rest. ArbCom has the ability to delegate and is moving in that direction WRT other matters. Your thoughts on the matter? Tony (talk) 03:21, 6 October 2009 (UTC)[reply]
I agree with you completely. I think the Committee has the authority to unilaterally designate this as their official admin oversight sub-committee. If you've completed the draft for the form and function of this forum, and all that's lacking is an election, then perhaps the Committee might be willing to supervise the first election, much like they did recently with the oversight and checkuser audit boards? Cla68 (talk) 08:06, 6 October 2009 (UTC)[reply]
Cla, don't you think elections easily become political / popularity-based and are not very good at sorting out candidates with the right skill-base? (I think WP has demonstrated this on occasion ...). My attempted solution for AdminReview was to list the skill-base (see overleaf) and have candidates formally declare that they have those skills before standing (on which basis they could be probed by the voters in the run-up to the vote). To digress slightly, I'd like to see this at ArbCom elections—watch that space <clears throat>.
One alternative to an election is for ArbCom to simply appoint the personnel, including one or two arbs to supervise: 18 arbs (which Jimbo has said will be the top-up number) are enough to spare a few for other duties in rotation. After all, we do elect ArbCom, and apparently trust them to be the peak body for judging our behaviour. The critical component for the AdminReview Subcommittee of ArbCom would be that non-admins be reasonably well-represented. The challenge would be that non-admins tend to be less experienced in the process side of WP, while the best admins have proven experience in sorting out behavioural issues. However, my hope is that there are non-admin users who are capable of doing the job: both categories are needed to gain the confidence of both categories of users, I think.
I guess the issue of taking on AdminReview, or something similar to it, as a delegated function might be a talking point in the run-up to the December Arb election. Tony (talk) 09:16, 6 October 2009 (UTC)[reply]
Sounds good to me. If you or anyone else starts advocating this somewhere in project space, please let me know and I'll help get your back. Cla68 (talk) 00:50, 7 October 2009 (UTC)[reply]
Proposed. Cla68 (talk) 07:06, 7 October 2009 (UTC)[reply]
I would point out that I am working up a proposal for an admin desysopping process that would have this page as a primary method of investigating admin conduct, leading to a admin desysop process which then decides on any length and conditions of flag banning. My talkpage, but more dedicated User:LessHeard vanU/Dead minimum is where I am working up the proposal based on this and a few other pages (mainly dealing with the desysop argument process). LessHeard vanU (talk) 12:45, 7 October 2009 (UTC)[reply]

←Good news, LHVU. If I get a chance, I'll try to make this page seem less voluminous. As Roux suggested above, a few open/close banners might be in order for the process. Tony (talk) 13:39, 7 October 2009 (UTC)[reply]

Draft RfC

Please see Wikipedia:WikiProject Administrator/Admin Recall and Talk:WikiProject Administrator. I hope this is self-explanatory. Please feel free to amend or comment. No idea if it has legs, but I think its worth a try. Ben MacDui 19:18, 22 October 2009 (UTC)[reply]

A little discussion on AdminReview

Here. Tony (talk) 11:08, 23 October 2009 (UTC)[reply]

small mistake

Under coordinators, it says "The six coordinators are:", and then lists seven names. (Well, placeholders, but yeah...) J.delanoygabsadds 23:02, 24 October 2009 (UTC)[reply]

Vote on community-driven desyssopping

Wikipedia_talk:Community_de-adminship/Draft_RfC#Community_de-Adminship_.27Proposal_Finalization.27_Poll Tony (talk) 12:41, 18 January 2010 (UTC)[reply]