User talk:Cenarium/Administrative Appointments Commission

User:WereSpielChequers/RFA_reform
Copying my notes from: User:WereSpielChequers/RFA_reform My general concern with an RFA committee is:


 * A potential solution would be to make adminship an indirect community decision by electing a committee or individuals to appoint admins, but unless we have a community discussion as to the criteria for adminship this masks the real issue which is that we haven't agreed the criteria that admins should be judged against. We'd have to set a criteria if we elected people to make these choices for us, either explicitly or implicitly by the views of those elected. So why not set a criteria and then judge candidates against it?

My particular concerns with this proposal include:  Ϣere Spiel  Chequers  21:48, 10 November 2014 (UTC)
 * 1) Timing. One of the few things that works about RFA is that the start time is entirely the candidate's choice. You could have a standing committee that hears applications as and when they come in. But a committee that only meets for certain times of the year is hopefully going to have a glut of work for a short period, but that may not be the right time for some candidates.
 * 2) This would be bureaucratic, especially in two 6 month appointments prior to someone becoming a full admin.
 * 3) 6 months and one year probation periods would be fine if our problem admins were ones who caused problems that quickly. But last time I looked at desysoppings it was more typically three years before admins went rogue.
 * 4) You need more detail as to how you are going to elect a committee, and if you use the method used for Arbcom you risk a very different sort of RFA. If deletionism/inclusionism is the main divide when electing the committee then the Arbcom election method risks electing a committee where one end of that spectrum got less than 40% of the vote and didn't make the committee. So the risk is that the appointments committee could be more extreme than RFA and either appoint a number of deletionists or inclusionists depending on who won the election. A proportional system could prevent this, and of course there is the possibility that the electorate would elect some very mainstream members. Though on an issue where feelings run strongly that is unlikely.
 * I don't see that as an issue since this would be a long process, of about two months, mostly of email correspondence, so the candidate could take his/her time to give answers. This would be nowhere near as rushed, stressful and intensive as RFA.
 * That can be changed to one reappointment required / six months term if desired. I've put this requirement since otherwise there would be automatic oppositions for candidates who wouldn't have gone through the commission appointment and 'trial'. You would still be a full admin though, you'd only have to be reappointed by the commission each time. The biannual RFC + election may be bureaucratic, but there is already so much talk going on all the time about RFA, etc. This would have the distinct advantage of formalizing and focusing all this talk in a constructive way. In the end it may save community time.
 * This is true, but this proposal isn't intended to solve this issue, it's up to ArbCom or some other proposal. The reappointment requirement is more a way to lower the threshold and have some sort of trial, to increase our pool of admins, while still keeping the decision on permanent status in the community's hands.
 * Yes, I'll think about it. Cenarium (talk) 22:31, 10 November 2014 (UTC)
 * An idea with regard to conduct issues of permanent admins, this proposal would open up a new sanction for arbcom, not as strict as desysopping : removal of permanent status. The admin would not be desysopped straight away, but would have to be reappointed by the next commission. Whether the admin would be reappointed or not would depend on community standards, the elected commission and community input. This could be employed more liberally than straight desysopping by arbcom, and would become a sort of forced reconfirmation. Cenarium (talk) 22:56, 10 November 2014 (UTC)
 * On point #1, "two months" can easily mean "the two busiest months each year at work" or "the two months when I was moving house" or "the two months before my thesis was due". WhatamIdoing (talk) 20:55, 11 November 2014 (UTC)
 * Yes, I've increased the commission duration to four months, so in total 8 months out of 12 months. Cenarium (talk) 21:06, 11 November 2014 (UTC)
 * At this point I think the specified process is far to specific. There are enough details so that the whole idea will be knocked back because some detail has a disagreement.  I would prefer a multistage approach.  Before this amount of detail is voted on we need a much simpler RFC, that decides if there is to be an administrator appointment commission or not.  After that is agreed then you can think about what if any term they appoint to.  I would support the whole idea of such a commission, even if I don't like the details.  The way that such a commission would be selected should allow a diversity of opinion to be represented, and not just one camp of opinion that happens to be the majority at the time. Graeme Bartlett (talk) 01:49, 11 November 2014 (UTC)
 * Ok, the general idea is a commission which appoints new admins for a limited term, and reappoints them for further terms, with the possibility to ask the community for a permanent term. This could work as a permanent commission too, for example, so indeed there are details that aren't absolutely necessary for it to work exactly as proposed, and the RFC could be first on this more general idea, as you suggest. Then we could decide whether the commission should be permanent or temporary, the term duration, the conditions for requesting a permanent term, etc. Cenarium (talk) 02:26, 11 November 2014 (UTC)
 * I think that might be a good approach. User:Beeblebrox has some experience with structuring proposals like that.
 * Also, I think that a two-month election process for a two-month job is unrealistic. Aim for committee terms of at least a year.  WhatamIdoing (talk) 20:55, 11 November 2014 (UTC)
 * I've increased the duration to four months (from 3 months, not 2). So 1 month RFC, 1 month election, 4 months commission, and repeat. This may be increased to 5 months by having the RFC and election at the same time but this would mean less time for analysis, thoughtful reflection, discussion, and I don't think that it would be that useful. A year would mean a 10 months term duration, with 1 month for RFC and 1 month for elections, but it would mean less frequent community input on the standards for giving out adminship and may put off some commission candidates. I'm also unsure if the admin term should rather be six months or one year in that case. Cenarium (talk) 21:26, 11 November 2014 (UTC)
 * I disagree that we need to institutionalize twice-annual RFCs to discuss adminship. If we needed public discussions about our views on the criteria for admins, then we'd be having them now.  That has bureaucracy written all over it, and I fully expect that any such process would produce two basic sets of responses:  (1) same basic standards as the last time [now quit bothering me!  I've got articles to write!], and (2) same soapbox as the last time [and it won't get consensus this time, either].  I could maybe see the value in doing it once at the beginning, and certainly in doing it whenever someone (the commission and/or anyone else) feels like it, but I don't see the value in requiring people to have an RFC twice a year.  Honestly, I think that you should remove that part of the proposal.  WhatamIdoing (talk) 22:38, 11 November 2014 (UTC)
 * I agree, this doesn't have to be automatic. Cenarium (talk) 22:46, 11 November 2014 (UTC)


 * This is how I see it : the community does not want to give any single committee the power to make permanent admins, the community does not want to require community reconfirmation of admins because it's already hard enough to pass RFA, so what of having a commission making and reconfirming term limited admins, who can subsequently ask for permanent status to the community ? Cenarium (talk) 02:42, 11 November 2014 (UTC)
 * I like the general idea, but I do think there are some things that need to be worked out. In particular, I think the prospect of a month-long, highly-scrutinized election process for a 1-2 month job is just as much of a turn-off as a long-term commitment. I would also be concerned about workload. If the commission only appoints a couple dozen people a year like the current RFA process, it won't be much of an issue, but it also means the commission probably isn't achieving all of its goals. However, if it gets back to the promotion rate of 2005-2007, or if few people ever decide to seek permanent adminship, you're potentially talking about reviewing several hundred people in a month or so. Mr.Z-man 04:37, 11 November 2014 (UTC)
 * I think the first issue can be solved by extending the commission's term, to 4 months. It would also address WereSpielChequers' concern about timing. Candidates could come forward at any time except during the last 3 weeks (so there's enough time to review). Successful candidates would be granted adminship one at a time. The candidates to reappoint would be listed from the start. So 4 months of commission, 1 month for RFC, and 1 month for election.
 * Reappointment should be the default option, so unless there are significant issues brought up by users, this shouldn't take much time in most cases. But this does add up so it's indeed a potential problem. Maybe the commission should recommend requesting a permanent status to worthy admins and nominate them itself (if they agree). The assurance of being recommended and nominated by the commission should be more inviting to candidates. Cenarium (talk) 16:57, 11 November 2014 (UTC)
 * I'm not convinced that "the community" opposes the idea of an elected committee being able to appoint permanent admins. It might need to start off with 6- or 12-month appointments, but I am not sure that community-wide confirmation votes are actually necessary.  If the committee posted a list of temp-admins that they were contemplating moving to permanent status, and requested that people with concerns contact them with evidence that this was a bad idea, then I think that might be adequate. WhatamIdoing (talk) 20:55, 11 November 2014 (UTC)
 * I think that starting with a time limited term is important, but indeed giving out permanent status does not necessarily have to be a form of community based process. The advantage is that it would be similar to the current RFA process, so it wouldn't be too drastic a change, and the proposal would be less likely to be opposed by users who want a strong community input or who prefer gradual changes. Cenarium (talk) 21:26, 11 November 2014 (UTC)
 * This proposal is no good. It creates complications where there are no need for complications. Such frequent elections would be a nightmare, and likely to quickly fall in obsolescence. Second-class administrators are useless. Administrators need to be free to act without being subject to repeated public shamming. RGloucester  — ☎ 13:45, 12 November 2014 (UTC)
 * It's certainly not public shaming, since it's a commission that would reappoint them, and let's just say that free admins isn't a welcoming prospect to a substantial part of the community. I've already addressed all this here and below in answer to John Carter. Two elections a year isn't that frequent, and we could do with one if necessary. Unfortunately, there is a need for complication, there has been numerous proposals for a committee to appoint admins on a permanent basis for ever, none succeeded. Look, go ahead and propose ASC, and if it fails I'll propose this. Cenarium (talk) 18:49, 12 November 2014 (UTC)

Second class administrators? Bad idea
We really don't need to establish a corps of second-class administrators. I can only see that causing problems. RGloucester — ☎ 17:34, 11 November 2014 (UTC)
 * This isn't a second class, this is a form of trial adminship, made necessary in order to lower the barriers for entry. Which problems could it cause ? Cenarium (talk) 17:54, 11 November 2014 (UTC)
 * The fact that they would be only "interim" administrators would make it much more probable that (1) they would be less likely to act in potentially controversial ways as interim admins than they would as regular admins, which would make the review of their generally extremely safe interim admin actions much less effective than it would be if they were acting the way they would as "full admins", and (2) it would make it much more likely that their admin actions would be taken to AN or ANI for review, which could easily make them feel like they aren't trusted (because they aren't, basically) and quite possibly make it less likely for them to seek full permanent adminship if the treatment they received as temps was even less respectful than regular admins receive. John Carter (talk) 20:44, 11 November 2014 (UTC)
 * (1) This is one of the main reasons I've introduced the new arbcom sanction of removal of permanent status. Permanent status would not be a given for life, and arbcom could use this sanction more liberally than straight desysoppings. (2) This would quite likely be the case if it was easier to remove adminship for them before their term is up, but this is not the case. I've said it was a form of trial, but it's not a trial in the strict sense, it's certainly not 'easy granting, easy removal' like rollback and such. It would be just as difficult to desysop a term admin before their term is up than a perm admin. There would be little point for frustrated users to cause drama at AN(I) because they would still have ArbCom to convince for any straight desysopping. There may be some threats of derailing their reappointment or potential RFPA, but the nature of the commission would make it insensible to worthless accusations. In the worst case scenario where they don't want to apply for perm adminship, they would still be reconducted in their term by the commission if they have a good track record.
 * Some of those concerns may still persist somewhat, but I still think that this provides a lower barrier to entry which would greatly increase our pool of admins. Cenarium (talk) 21:51, 11 November 2014 (UTC)
 * This sort of scheme might result in a lower barrier to becoming an admin, or it might not. You won't know until or unless you try it. But I don't see anything in it which would make things easier for candidates.  Ϣere Spiel  Chequers  22:40, 12 November 2014 (UTC)
 * Indeed, there are many things that may not turn out as I would expect, still a trial may be worth it. As for not making it easier for candidates, don't you agree that being evaluated by an elected commission based on an email correspondence at a relaxed pace would be less taxing than RFA ? Cenarium (talk) 05:29, 13 November 2014 (UTC)
 * Whilst I don't agree with this proposal for several reasons, most especially the problem that we would need to agree a criteria for adminship before appointing an appointments committee; "2nd class admins" would not be useless, far from it. The vast majority of admin actions are uncontentious blocks and deletions and as long as they are done correctly no-one will complain. Closing contentious RFCs and AFCs and blocking experienced editors are a very small part of the total admin workload. Many new admins avoid doing such things until they've found their feet - which is one of the reasons why a probation period is the wrong direction.  Ϣere Spiel  Chequers  22:33, 12 November 2014 (UTC)

Request for clarification and proposed semi-alternative
As some people who haven't been keeping up with all matters admin and otherwise around here might not know this immediately, I think it might be useful if somewhere on the page it specified exactly which particular functions are available only to administrators and which are available to other levels of user-rights.

Also, as one possible option, maybe, and this is clearly just a maybe, maybe such a group could also be active in perhaps seeking to give out other levels of user rights, perhaps on a cumulative basis, to editors who would be interested in using those particular user rights, and perhaps after successfully using other rights packages for a while nominate potential candidates for adminship specifically. John Carter (talk) 21:42, 11 November 2014 (UTC)
 * Do you mean by that unbundling and having the commission grant different rights packages, such as only delete/undelete for those interested mostly in deletion, and only block / protect for those interested mostly in anti-vandalism ? Unfortunately, many of the common objections, mainly interdependence between those rights, would still hold. In my opinion, the best option would be to have a form of soft deletion that can be granted to non-admins; as for anti-vandalism, I'm working on that with the pending changes blocks proposal. What we have already is the template editor usergroup for those interested in templates and technical matters.
 * As for your first point, there's Special:Listgrouprights. There has been an accumulation of new userrights in the last years, and the purpose of many of them is unclear, such as spamblacklistlog (allows to watch blacklist log, only admins), titleblacklistlog (not a single entry in this log, so ?), jsonconfig-flush, transcode-status, and all of those related to the flow extension. I would also prefer that the admin usergroup doesn't have all those course-related userrights which clog up the list, and are already affected to course coordinators, let's keep them separate. There is in fact a general lack of information on many userrights, even at mediawiki.org, for example I couldn't find out if noratelimit has any use besides preventing more than six account creations. Cenarium (talk) 03:59, 12 November 2014 (UTC)
 * I was more thinking that there are other lesser rights groups, like you indicated, including template editor, and the rather longish list of additional groups you mentioned. It might to my opinion make most sense to have individuals who have an interest in the delete/undelete and block/unblock/protect rights to maybe go through what might be the less controversial rights like template editing first, give them a chance to familiarize themselves with it, then maybe expand to other rights, and maybe taking on the "big two" of deletion and protection/blocking rights last, after they've demonstrated that they are competent in following the procedures associated with the lesser rights first. John Carter (talk) 16:40, 12 November 2014 (UTC)
 * I see what you mean, it's similar to the quasi-admin usergroups that I had envisioned here. I see three of them, in those areas : (1) technical, and template editor is a good start, actually I just suggested granting it more userrights here (2) anti-vandalism, with a sort of 'moderator' usergroup, which is essential to my proposal on pending changes blocks and (3) page creation and deletion, with some kind of 'curator' usergroup, but unfortunately we'd need better page creation / deletion software for this one. It could indeed be an occasion for potential admins to prove themselves.
 * To get back to your first question on userrights, there is a table at User_access_levels, but it's rather succinct, I've added some more details. Cenarium (talk) 07:17, 13 November 2014 (UTC)