Wikipedia talk:Administrators/RfC for BARC - a community desysopping process

• /Archive (discussions leading to the final draft of this RfC)

Final draft finished

OK, so after a lot of work and superb advice we have come up with a proposal that instead of simply testing the community's opinion for the 'nth time, it offers us a complete solution that we can take or reject. It should appeal to those who suggest that it is too difficult to address the problem of errant admins and by the same token, lead to an amelioration of the RfA process by lowering the bar. Many thanks to everyone who provided feedback and helped us put this together. --Kudpung กุดผึ้ง (talk) 15:31, 23 July 2015 (UTC)

• As it sits, I think it is very workable. I like doing the procedures separately as well, keeps from bogging down the concept with details. Dennis Brown - 2¢ 17:22, 23 July 2015 (UTC)

One point

Hello all,

It's said on the proposal page that the members of BARC would be able to refer the case to ArbCom. Ostensibly, the two committees would both be handling user conduct issues, and obviously BARC would have a smaller remit than ArbCom. But I really want some comments on when it would be necessary to turn a case over to ArbCom. Thoughts? --ceradon (talk • contribs) 19:29, 23 July 2015 (UTC)

My priority was to ensure this was a lightweight process, and anything heavy could be passed to Arbcom. So, intractable issues, which involve admins but also large swathes of the rest of the community (such as civility enforcement for example). Also, as this is a public process, privacy issues - off wiki information, anything like that. Basically, I want the door open to pass it on. This isn't a "buck stops here" process, like Arbcom. Worm^TT(talk) 21:20, 23 July 2015 (UTC)

The question is not one whose answer is in the system proposal. It's precisely one that would be addressed by the committee on a case-by-case basis. --Kudpung กุดผึ้ง (talk) 23:03, 23 July 2015 (UTC)

That seems more a question for the procedures discussion, not the policy discussion. Here, we are simply setting up the authority, not how they will do it. That might take some learning curve and yes, a case by case basis regardless. Dennis Brown - 2¢ 01:23, 24 July 2015 (UTC)

I'd recommend that issues where a sysop's editorial behaviour is in question (e.g for NPOV violations) should be within Arbcom remit - a desysop does not remedy editing issues, only administrative ones. Jo-Jo Eumerus (talk, contributions) 09:45, 24 July 2015 (UTC)

Jo-Jo Eumerus, some recent issues concerning editing that have led to a global ban could have been adequately handled by the BARC. BARC is an admin-specific tribunal and would be empowereed to resolve most issues except those that reach stalemate or need hearing in camera due to privacy concerns.Kudpung กุดผึ้ง (talk) 10:04, 24 July 2015 (UTC)

I see a lot of confusion here.

1. Anyone can refer anything to ArbCom. They are not bound to accept the referral. (It appears they can also act sua sponte in pretty much any way they like - but that's another matter.)
2. A de-sysop procedure does not require significant investigative powers. The community has the right to make the "wrong" decision, just as ArbCom has, just as we do at RfA, AfD, etc..
3. It's certainly the case that judging consensus is what 'crats do. I see no need for complex appointments and committees for them to do more of the same. But that can be decided later.
4. A lot of "intractable" issues are very tractable, they simply require a little imagination.
5. "Private information" is rarely an insurmountable hurdle. In this case assuming it was shareable with "the elect", for example a good but non-publishable reason for banning a cabal of 30 established editors, then ArbCom, or the WMF, or any other trusted party could be asked to so certify by the admin who was being proposed for de-sysop.
6. Editorial behaviour should be and is addressed by the community. Being an admin should have no effect on that process.

On the whole, keeping the process simple and transparent is the best thing. All the best: Rich Farmbrough, 01:58, 27 July 2015 (UTC).

Queries

"Bureaucrats are authorised, on consensus being reached, to physically enact the removal of tools."

But further up it's billed as a "vote".

Could this be clarified? A vote is presumably 50% plus one. Consensus is something much more. Tony (talk) 07:43, 24 July 2015 (UTC)

I'd like that to be a firm question for the procedures phase of the RfC. However, my personal thoughts is that it should be a vote with visible comments, similar to arbcom's proposed decision page. That would mean that votes can change, and increase transparency. I'd also want 60% agreement that the tools should be removed, otherwise it should be referred to arbcom. Still, that's all for the next phase. Worm^TT(talk) 09:34, 24 July 2015 (UTC)

I am very sympathetic to the concept of "stage one, decide whether we want to do this, stage two, work out the details" but I think that there are a bunch of people who will answer "should we do this? It depends on the details." --Guy Macon (talk) 16:19, 24 July 2015 (UTC)

And at the same time, if we do it the other way "Here is a fully fledged solution" everyone who is unhappy with a small part will veto the entire solution. This way those who are interested in specific issues can debate those specific issues without derailing the entire solution. Worm^TT(talk) 16:25, 24 July 2015 (UTC)

To be clear, you see this RfC as addressing the question "Is this process, as advanced, worth fleshing out in a future RfC?", correct? That would mean a vote to support is not a vote to enact this process immediately and figure out how it works later, or even to enact this process at all if the finer details cannot be agreed to in a later RfC. Please correct me if I'm wrong, since I may need to reconsider my vote if that is the case. ~ Rob^Talk 16:57, 24 July 2015 (UTC)

Limiting complaints to admins only

One of the main issues with opening up a process by which anyone can advance a complaint is that anyone can advance a complaint. I think everyone will agree that a completely open process has the potential to create unnecessary drama over routine matters. It's a waste of everyone's time if two vandals (or worse, a vandal and their sock) bring a complaint about an admin doing routine work at WP:AIV. It seems that this process would work much more smoothly if it required two admins to advance a complaint for it to be considered. If editors had legitimate complaints regarding a sysop, they would just need to convince two admins that their concerns are substantial enough to be worth procedurally advancing. If the support of any two admins cannot be found, the complaint certainly has no merit. I don't see any drawback to this, while I see the potential to seriously limit unnecessary drama. ~ Rob^Talk 16:55, 24 July 2015 (UTC)

Another possibility would be to apply WP:INVOLVED to such complaints, i.e to forbid nominations by users who are in a dispute with the admin in question or were affected by administrator actions by the admin in question. Jo-Jo Eumerus (talk, contributions) 16:57, 24 July 2015 (UTC)

(edit conflict) Good god, no. Consider a "twelve months activity and 500 edits" threshold if you want some kind of bar to keep out the sockpuppets, but the last thing Wikipedia needs is something that gives admins even more super-user powers. – iridescent 16:58, 24 July 2015 (UTC)

I fail to see how creating a process for desysoping gives admins more power, even if that process requires an admin to advance a complaint. I don't know too much about desysoping via ArbCom in the past, but from what I've seen, those efforts generally include admins advancing the complaints anyway. ~ Rob^Talk 17:02, 24 July 2015 (UTC)

Huh? Every editor has exactly the same rights at Arbcom. Admins are probably marginally overrepresented because they're both more likely to spot controversial actions, and more likely to make controversial actions that can't be resolved by the community, but there's certainly no bar of any kind. Looking at the 10 most recent Arbcom desysoppings at Wikipedia:Former administrators/reason/for cause, five of the cases were brought by non-admins and five by admins. – iridescent 17:10, 24 July 2015 (UTC)

I know there's no bar, and was just pointing out that admins tend to advance complaints in ArbCom that lead to desysop (although I thought it was more often than 50-50). ~ Rob^Talk 17:18, 24 July 2015 (UTC)

Absolutely not, no. It seems to me that part of the reason for this proposed process is to provide a simpler means for anyone to gain assistance when they believe they see admin abuse (and I've seen what I would consider admin abuse in my own time here). Making it so that admins could only be reported by their own kind would achieve exactly the opposite and would entrench the "them and us" attitude that is so easy to see around this place. And as iridescent says above, it would make it more restrictive than the current Arb Com system rather than more open to the community of editors. Mr Potto (talk) 17:45, 24 July 2015 (UTC)

Bad idea because of the them-and-us flavour, and unnecessary because the proposed system already begins with the BARC deciding whether or not to accept a case. It surely doesn't need yet another level of gatekeeper. That's not janitorship, it's governance. And it might defeat one of the purposes: I for one would be more rather than less likely to oppose at an RfA if I thought a candidate was likely to play gatekeeper in that sort of way. --Stfg (talk) 18:09, 24 July 2015 (UTC)

If you wanted it to be easier to de-sysop then you would want more admins, increasing the chance that one would be amenable to proposing the de-sysop....

But I also have qualms about requiring a sysop (even 1 of 1700) as gatekeeper.

All the best: Rich Farmbrough, 02:09, 27 July 2015 (UTC).

Maybe making sysop be the requirement is a bad idea, but there needs to be some type of requirement. At the minimum, a six month activity and 100 edit requirement would limit socks or disruptive editors from putting forward the names of admins that have done nothing wrong. The ability not to accept a case, which occurs after a response from the admin who is the subject of a complaint, is not a sufficient step to prevent drama and prevent good admins being harassed with bad faith or meritless complaints. If an admin who routinely handles AIV or CSD nominations had to officially respond to a complaint in an ArbCom-lite process every time someone didn't like Wikipedia policies, they would quickly burn out. A process to remove bad admins must also protect good admins from harassment. Otherwise, we're going to replace one problem with a new problem. ~ Rob^Talk 18:27, 24 July 2015 (UTC)

Meritless complaints at ANI seem to get dismissed pretty quickly, often without bothering the person complained about, and I'd expect the the BARC people to be able to see through meritless complaints easily enough and deal with them. Mr Potto (talk) 18:40, 24 July 2015 (UTC)

I agree. I would expect the same type of newbs who make ArbCom threats (who often seem to be an "attorney") would threaten or fire up frivolous complaints. But they would get shut down by the "gate" stage. Having said that, such a requirement would prevent them making foolsof themselves. All the best: Rich Farmbrough, 02:09, 27 July 2015 (UTC).

Aren't Arbcom part of "the community"? And how would this "Arbcom2" a.k.a. BARC be "more community"?

My understanding that ArbCom is part of the Wikipedia community. After all they're all volunteer editors, voted in by volunteer editors. Basically "the community" is sometimes used for shorthand for "non-ArbCom editors", but really it's a misnomer and not accurate, as ArbCom are all bona fide community members.

My question is how would this BARC be more a part of the community then ArbCom? I would argue it will be less so, as five of its members would have to be bureaucrats; in other words there is more restrictions on community membership than with ArbCom. ArbCom members are all administrators, but non-admin ArbCom members could be elected, except this doesn't seem to happen in practice. --Jules (Mrjulesd) 18:33, 24 July 2015 (UTC)

Aren't bureaucrats part of the community too? Mr Potto (talk) 18:41, 24 July 2015 (UTC)

Yes but a more restricted subset. With ArbCom anyone can get elected. With BARC very few are eligible for 5 of the posts. --Jules (Mrjulesd) 21:20, 24 July 2015 (UTC)

as Arbs are elected for fixed terms of office all Arbcom members have been elected relatively recently, much more recently than most of the Crats, some of whom became Crats in 2004/2005. So BARC isn't as community based a system as Arbcom, at least not in terms of who makes the decision, but it would be a community process in terms of transparency. As cases involving private evidence would be reserved for Arbcom all BARC cases would be viewable by the community, this of course raise two further glitches with the process, firstly what do we do when there is a dispute as to whether private information is relevant to a case? Secondly if Arbcom loses some of its transparent cases it will become harder for the community to judge arbs and know who they want to reelect. ϢereSpielChequers 15:54, 26 July 2015 (UTC)

So BARC isn't as community based a system as Arbcom...: agreed and my main point. ...but it would be a community process in terms of transparency.: well I think a better solution would be to try to increase the transparency of desysoping by ArbCom if this was felt to be deficient. An RfC along those lines may be more productive than all this BARC "community" desysoping business. what do we do when there is a dispute as to whether private information is relevant to a case? I think the plan is for these to be deferred to ArbCom. --Jules (Mrjulesd) 16:05, 26 July 2015 (UTC)

I think the BARC plan is for ArbCom to handle all cases where it is agreed that personal information needs to be considered. One of my concerns about having two rival desysoping processes is where it is unclear which route a case should take, one way would be to give anyone involved in the case the right to have it handled by Arbcom because they intend to supply some information that includes personal off wiki info. But the drawback of that is that it is close to giving participants a right to choose Arbcom over BARC. That could lend itself to gaming and controversy, especially if the accused comes back from a weekend off to find themselves three days in to an 11 day BARC case where they consider that some off wiki info is relevant and want the case switched to Arbcom. This becomes a bigger problem if BARC gains the power to desysop for reasons that Arbcom doesn't consider. ϢereSpielChequers 16:33, 26 July 2015 (UTC)

Well that would be pretty ironic, if that was the case. If admins brought before BARC claimed private information each time, to get all the cases deferred from BARC to ArbCom. It would make the whole BARC process pretty moot and self-defeating. It also shows how half-baked this proposal is if they haven't thought of that.

I think that having two processes for desysops could lead to many of these confused and contradictory scenarios. Far better, in my view, to have a single process, and to have RfCs for reform if necessary. Who should you go to for a complaint against an admin? ArbCom? BARC? Two processes for one action seems pointless and overtly bureaucratic. Just have ArbCom for desysoping, or just have BARC, but to have both I feel should be avoided. That this will solve the problems with RfA seems unlikely to say the least. --Jules (Mrjulesd) 17:06, 26 July 2015 (UTC)

Comments from the proposers make clear that they believe this will improve RfA by attenuating the "appointment for life" aspect. That argument requires the hypothesis that BARC will desysop people where Arbcom currently would not. (Or, that the threat of being BARC'd will improve admin behavior - but you need to have some 'successful' BARC cases for that to be realistic.) So the use case here is situations:

• that are not emergencies - those go to Arbcom
• that are not highly complex disputes - those go to Arbcom
• that do not involve private evidence - those go to Arbcom
• that are not about long-running patterns of behavior - there's not enough time in a week to compile and examine the evidence, so those go to Arbcom
• that would produce a desysop where Arbcom would either decline the case or do something else.

Do those situations actually exist? Opabinia regalis (talk) 17:19, 26 July 2015 (UTC)

Firstly all this "appointment for life" business is in my view ridiculous. As well as inactivity, I've a substantial number of desysops by ArbCom, probably more desysops than successful RfAs. If they think more desysops is the answer for admin retention I really don't get it. RfAs need reform, not more committees.

Secondly, no I don't think these cases really exist, not if a thorough job is to be done. --Jules (Mrjulesd) 17:42, 26 July 2015 (UTC)

The issue/problem here is that no one seems willing to "endure" ArbCom just to "report" WP:ADMINCOND issues. That is absolutely a "hole" in the system. People will only go to ArbCom for "abuse of the tools" issues, or shady COI stuff. The unfortunate result of this is that is has effectively given some Admins the "license" to be JERKs to other editors and Admins. I definitely would like to see something done about this. The problem on my end is that I don't think this BARC proposal is the way to handle it (as it is overly bureaucratic(heh) and redundant to ArbCom). I'd rather see an ANI-specific proposal crafted, so at least there would be some process to deal with "bad conduct" (as opposed to "abuse of the tools") Admins. But, as I said – I don't think this BARC proposal is the best solution for that. --IJBall (contribs • talk) 17:48, 26 July 2015 (UTC)

OK if an admin is being a jerk towards you, or any other editor for that matter, you can bring it to ANI. You can also get a consensus for whether you have a case at ANI. ANI can't desysop, but if there was consensus for your views at ANI it would give them far more credence if you were to take the case to ArbCom.

Basically you can make complaints against admins at ANI already. We don't need a second ANI. --Jules (Mrjulesd) 18:01, 26 July 2015 (UTC)

As I have already said, if the problem is getting an issue to the door of the committee in the first place, then it would be more productive to try and solve that particular aspect of the problem. One of the possible solutions, one which is both simple and requires no changes to policy, would be to create something like Wikipedia:Request an RfA nomination. A bunch of experienced editors or admins, if editors are afraid of possible retaliation, volunteer to file cases on behalf of others who, through inexperience, fear or whatever, can't or won't report an administrator to arbcom. The original report could even be done in private, they'd quickly review the allegations and, if they think a case is appropriate, they could file an arbcom request. There are also more complicated alternatives, which, however, would involve changing policy (including my favourite: a community-driven reverse RfA) My point is that, while there are problems with the current system, this process does not tackle them. Salvio ^{Let's talk about it!} 18:11, 26 July 2015 (UTC)

I'm willing to accept the theoretical possibility that there could be some cases that don't go to Arbcom because people don't understand our systems, or find it too much work to collect the evidence for a credible case. I'm not convinced that this proposal would solve either possible problem, I can't remember a community desysop proposal that would have done so. But we could tweak the existing Arbcom process to address that theoretical gap. If we had a group of volunteers who were willing to look into complaints, advise people whether to take their claim forward and help them collect and organise the evidence, that should address that concern more effectively than any of these community deadminship proposals could do. ϢereSpielChequers 18:11, 26 July 2015 (UTC)

Great minds think alike... Salvio ^{Let's talk about it!} 18:12, 26 July 2015 (UTC)

This is a version of Wiki-Advocates? All the best: Rich Farmbrough, 02:16, 27 July 2015 (UTC).

Perhaps a reincarnation of Wiki-Advocates, of course if we trialled it for a year and didn't get anything useful we could be a little more confident that the argument that there are big bad admins out there who nobody dares take to ARBCOM is total hogwash. In which case it could be reinterred waiting the next reincarnation of Wiki-Advocates. ϢereSpielChequers 09:30, 28 July 2015 (UTC)

@WSC Well I think the story must be there are rogue admins out there that deserve to be desysoped, but either (a) people don't want to go to Arbcom to deal with them for unspecified reasons, and/or (b) the haven't been dealt with by Arbcom in a satisfactory manner. But I would really like to know which cases people in mind. Can anyone bring up a case where BARC would have been preferable to Arbcom, or a case where desysoping by Arbcom is not possible? What is really needed is some example situations were this BARC business would have been preferable. --Jules (Mrjulesd) 10:33, 28 July 2015 (UTC)

Is it tested that AN/I can't de-sysop for cause any-more? I certainly recall requests due to security that were honoured. Maybe a 'crat can comment on what they would do if there was a community consensus for de-sysop at AN/I or Village Pump? All the best: Rich Farmbrough, 02:16, 27 July 2015 (UTC).

Comparison to Arbcom

Here is a table comparing the procedures and committee membership of this proposal to those already established for ArbCom. Feel free to edit if I've missed anything.

		Lightweight Community De-adminship	Arbitration Committee
Cases	Who brings a case?	Any user (but complaints about multiple incidents must be filed by at least two users)	Any user, but there is typically consideration of "standing"
	How are cases accepted?	Vote by members of the committee	Vote by members of the committee
	How long does it take?	Fixed schedule: 3 days from request to acceptance 7 days from acceptance to closure 1 day from closure to final decision	Variable schedule set for each case; typically runs 1-2 months Flexible to accommodate participants May be subject to delays, often due to constraints on arbitrators' time and/or the quantity of evidence received
	What happens when committee members are unavailable?	Unknown	Individual arbitrators noted as inactive at the beginning of each case
	On what basis are admins to be desysoped?	Unknown	Abuse of tools Repeated misuse of tools Conduct unbecoming
Membership	Who is on the committee?	Five bureaucrats self-selected for each case Five static community members	Fifteen arbitrators, less those inactive or recused on a given case
	How is the committee selected?	Bureaucrats: self-selected for each case Bureaucrats themselves selected by 7-day RfB with 85-90% community support Community members: annually appointed by bureaucrats after indeterminate "community consultation"	Annually elected to overlapping 1-2 year terms Elections run for 1 month, in two 2-week periods for nominations and voting Experience requirements for voters Votes cast with SecurePoll system Support percentages ~60-80% for successful candidates
	Who is eligible to join?	Bureaucrats: any current bureaucrat Experienced admins are eligible to run for bureaucratship Community members: unspecified constraints on consideration for appointment	Any community member in good standing who: meets specified experience criteria is willing to self-identify to the Foundation has disclosed any alternate accounts In practice, competitive candidates are experienced admins
	How are problematic members removed?	Bureaucrats: No current de-bureaucratship policy Indeterminate process for removing a given user from the self-selected bureaucrat group Community members: Indeterminate	Resolution by 2/3 of active arbitrators

I am unable to see, comparing these two proposals, how the present one can be described or advertised as more "community" based than ArbCom. Opabinia regalis (talk) 22:13, 24 July 2015 (UTC)

The reality is this isn't a community desysoping process but a bureaucrat desysoping process. Of course it is marketed as a community process to fool people into accepting it. While it is a much less democratic and community driven process than ArbCom. But 10/10 for marketing. --Jules (Mrjulesd) 22:23, 24 July 2015 (UTC)

Yes, and I think that's my problem with it. The fact is, a true "community"-based desysopsing process is going to have to be some process that comes out of ANI/is related to ANI. I don't see how the current proposal gets a true "community desysopsing" – this proposal seems quite the opposite (and seems fully redundant with ArbCom to boot). (Note: I still haven't voted "oppose" yet, though I am currently leaning in that direction for sure...) --IJBall (contribs • talk) 22:53, 24 July 2015 (UTC)

• Seconded. In theory this seems like a godsend to the flow of RfA candidates and voter satisfaction, but in practice, it's just giving lynch mobs and mad congeries a way to pursue their vigilantism against admins that they don't like. Esquivalience ^t 01:45, 25 July 2015 (UTC)

Accepted your offer to edit the table, and made a couple of minor changes. First, any user can bring an arbcom case, and I'm not aware that we have ever discussed their "standing" when considering case requests. Certainly not in the last six months, anyway. Second, some cases are certainly delayed by the availability of some Arbcom members; others are delayed by the volume of evidence (for example the recent Lightbreather case which had substantial private evidence from several parties, engagement with the WMF and several lengthy private debates before it could be resolved). I suspect this would delay BARC outcomes too, the timetable notwithstanding. If certainly should delay them - If would be disappointing if evidence was rushed through just to get the deadline done. And agree with you on which one is more "community based," though BARC still does propose a reasonable degree of community involvement. -- Euryalus (talk) 12:48, 25 July 2015 (UTC)

Thanks, I misremembered - other users have discussed "standing" (at least once in the AE case), but not arbs.

As for delays, I'd assume that this "lightweight" process specifically for admin conduct wouldn't be looking at the volume of private evidence or having the extensive private discussions that might come up in an arbcom case. I updated the table again to be clearer about my reason for including time constraints on the part of the members. The current proposal does not discuss what happens if members are unavailable or what a quorum might be. Opabinia regalis (talk) 16:35, 25 July 2015 (UTC)

Looking at the comparison table I'm left with the thought "How in hell can this be described as 'lightweight'"? Answer; it can't. It's just another bureaucratic (and by that I mean bureaucracy, not the bureaucrats) nightmare we do not need. --Hammersoft (talk) 11:28, 6 August 2015 (UTC)

Bureaucrat activity levels

Some information on current bureaucrat activity levels. Let's agree that there's not really enough specifically bureaucrat work to go around, and just look at general participation levels. Specifically, since this is an admin-review process, it would be useful to know how engaged the crats are in routine admin tasks, as well as how frequently they edit overall.

Result: only half the current bureaucrats have taken 10 or more admin actions in the past year, or made 10 or more edits in the last month. Yet the "community driven desysop" proposal requires five of them to caucus at a time within a week and a half? Opabinia regalis (talk) 05:37, 25 July 2015 (UTC)

Bureaucrats by activity level

Bureaucrat	Admin Actions (1yr)	Edits (30d)	Edits (1yr)
28bytes	218	4	632
Acalamari	537	473	4669
Addshore	31	11	142
Andrevan	2	71	454
Avraham	15	136	690
Bcorr	10	0	274
Bibliomaniac15	12	0	82
Brion VIBBER	0	1	16
Cecropia	0	1	28
Cimon Avaro	0	1	7
Cprompt	0	1	3
Deskana	2	6	86
Dweller	138	360	2600
EVula	3	11	124
Ilyanep	0	0	1
Infrogmation	16	19	150
Jrosenzweig	0	0	0
Kingturtle	9	10	202
Maxim	71	9	311
MBisanz	413	205	3053
Nihonjoe	268	122	3099
Pakaran	1	3	21
Raul654	0	0	12
Secretlondon	0	1	93
Stan Shebs	3	0	5
UninvitedCompany	0	4	15
Useight*	1	17	26
Warofdreams	1	259	465
Wizardman	1036	53	16117
WJBscribe	49	93	531
Worm That Turned	191	107	1664
X!	0	2	53
Xeno	217	564	5610

*Edit count for Useight's Public Sock

This is interesting. Looks like we would need to elect more bureaucrats before BARC can operate properly, something which I imagine would be a knock on effect of this proposal anyway. Sam Walton (talk) 09:54, 25 July 2015 (UTC)

My edit count is inflated due to renames and the automated page moves that accompany them. I'd be interested to see a table that discounted those edits. –xeno^talk 11:30, 25 July 2015 (UTC)

I can probably add renames later when I'm not on a tablet, unless someone beats me to it. Opabinia regalis (talk) 17:02, 25 July 2015 (UTC)

Let Opabinia regalis be fully aware that this current proposal has nothing to do with bureaucrat activity issues. It's about affording a new layer of protection for admins against the baring of the teeth of the peanut gallery and anti-admin brigade while at the same time offering an easier access for intimidated users to an equitable process. I can't really imagine any reasonable admin rejecting such a system, and the side effect is that a more interesting mandate and more to do might attract more candidates for RfB. Discussions about bureaucrat activity are currently taking place in other venues including suggestions by the bureaucrats that they should deprecate themselves. Fortunately they cannot do that unaided by a strong community consensus, but if they were to succeed in disbanding the Bureacrats as a group, I would be truly sorry to see some of them go. Kudpung กุดผึ้ง (talk) 14:09, 25 July 2015 (UTC)

How exactly is this proposal a "new layer of protection for admins against the baring of the teeth of the peanut gallery and anti-admin brigade"? It looks more like a second kick at the can to me.

I really do think this whole thing would be a lot more plausible if you did not choose crats for this task, it seems like an arbitrary choice. It seems it would have made more sense to create a new group than the redefine an existing group. The low number of crats, their relative inactivity, and the fact that they have not expressed an interest in this task make their choice problematic. Chillum 14:38, 25 July 2015 (UTC)

Kudpung, you've proposed a process that depends on gathering five bureaucrats on demand for a time-intensive eleven-day-long task. I'm at a loss how you can think that current bureaucrat activity levels are not relevant to how practical a process this is. (FWIW, I opposed the bureaucrat activity RfC, but essentially on grounds of 'if it ain't broke don't fix it'. And I doubt anyone arguing for deprecation of bureaucrat the user group is looking to see any of the individual bureaucrats go anywhere.)

I can't really imagine any reasonable admin rejecting such a system - this is the problem! An idea you can't imagine anyone disagreeing with is one that's still half-baked. The point of having a community discussion about it is to give people a chance to kick the tires and see if they hold up. Ahem, pardon the mixed metaphors. Opabinia regalis (talk) 17:02, 25 July 2015 (UTC)

I think it depends on what you mean by "rejecting". There's a lot that I would prefer to be done differently, indeed most supporters have their qualms about one aspect or another. But clearly the "support" is very strong for the idea overall.

Describing this as "time intensive" assumes that each proposal will somehow be complex. It should not be. If consensus is not clear, then it is no-consensus. If consensus is clear then there is similarly no problem. Discretion only applies to a grey area of a grey area. And in this case the proposal goes into too much detail for my taste, giving BARC votes and ArbCom a tie-break capability - I would just let 'crats decide if there is consensus to de-sysop - which as I understand it is the current, but unused, position.

All the best: Rich Farmbrough, 02:29, 27 July 2015 (UTC).

An odd fact, which I'm not sure which way it cuts but I find interesting, is that three of the most actively editing 'crats as listed on that table are themselves former arbitrators, and a fourth is an AUSC member. Newyorkbrad (talk) 22:38, 25 July 2015 (UTC)

recusal unavailability and quorum

I'm assuming that both the Crats and the other members of this committee will recuse themselves where appropriate in the same way that arbs do. But the very tight timescale means that a lot of people will be unavailable for particular cases and will have to decline them because of the timing. That could lead to some very small groups of decision makers, so I would hope that a quorum should be applied, I would suggest a quorum of five, two of whom must be Crats. Any case that doesn't achieve quorum could either have case times extended to allow for more participation or be referred to Arbcom. ϢereSpielChequers 15:25, 26 July 2015 (UTC)

I rather like Kudpung's H. H. Asquith approach to assuring quoracy. All the best: Rich Farmbrough, 02:32, 27 July 2015 (UTC).

Appointing additional bureaucrats would help, but to take on a commitment like this is not a small thing, with such a short timescale any member would have to be fairly confident they had a chunk of wiki time available pretty much every day for the next eleven days. That may be more possible for those with mobile broadband than it would have been a few years ago, but it seems to me a big ask for a volunteer, especially the five non crat members who would presumably be expected to each take several cases per year. This proposal just seems a lot less practical than Arbcom. ϢereSpielChequers 23:58, 27 July 2015 (UTC)

NYB emphasized here that the final deliberations are currently scheduled to run for one day. Not only do you need to know in advance that you have frequent online availability for the next eleven days, but you need to know eleven days in advance that you'll be available essentially around the clock on the final day in order to fully participate in the deliberation phase.

This timeline is not actually possible for volunteers with real-life commitments. One cannot perform the functions of this committee without being fully available all day on a daily basis, and willing to pull an all-nighter as necessary to deliberate with people in other time zones, on a schedule dictated by incoming requests. Restricting a powerful user group to only those able to make such an implausibly large time commitment would be a serious systemic bias issue. Opabinia regalis (talk) 01:18, 28 July 2015 (UTC)

It's a proposal, I have no problem with the actual length being up for debate in the next phase, but I'd like there to be firm deadlines. Worm^TT(talk) 07:50, 28 July 2015 (UTC)

It is a deeply flawed proposal whose originators are not amending it to respond to reasonable concerns. This is not some minor detail that could be addressed in a subsequent phase, the whole premise of the proposal is to create a speeded up alternative to the Arbcom process, if the speeding up is dependant on a schedule that is this unrealistic then the proposers need to come up with a realistic proposal before it reaches the next stage, either that or agree to reframe the proposal as using the same timeline as Arbcom. ϢereSpielChequers 09:44, 28 July 2015 (UTC)

I've said before and I'll say again, I ensured there was enough flesh on the bones of this proposal that the community could see what we were aiming for and what the shape of it would look like. I'm happy for parts to change, we operate by consensus. I do agree that the time line could do with debate and re-evaluation - and in the next phase I'll be putting in some empirical analysis to address the comments regarding activity and timescales, but it is a discussion for the next phase. As numerous other people have said, you cannot expect a fully fledged proposal for such a major area to enjoy full consensus of the community - that's why we're doing it step by step. Worm^TT(talk) 09:58, 28 July 2015 (UTC)

I'd like to add that "amending it to respond to reasonable concerns" is absolutely not a good idea; users who have already voted would then have supported or opposed a different set of ideas to the ones presented at the end of the RfC. Best to leave everything intact and make amendments in the next phase. Sam Walton (talk) 10:01, 28 July 2015 (UTC)

If amendments are controversial it is always possible to contact those who previously commented. But I do expect a fully fledged proposal to need full consensus of the community before being implemented, especially if it for such a major area. I intend to continue as an editor even if this is implemented, but I can't in all conscience nominate people for adminship while this is being taken seriously and I doubt I'm the only one who wouldn't continue as an admin if this was implemented as proposed. ϢereSpielChequers 10:38, 28 July 2015 (UTC)

I don't understand why you wouldn't continue as an admin but ok. On a related note, final RfC on the final version of this process, after procedures have been settled, would probably be a good idea though. Sam Walton (talk) 10:46, 28 July 2015 (UTC)

I'm a fairly active wikipedian, but I have a life which sometimes means I need to be away from Wikipedia for three or more days at a time, sometimes I go a week or more without doing more than pop by and fix a few typos. If this process is implemented unamended then that makes me more vulnerable than I care to be to various trolls out there, and it would be sensible for me to stand down as an admin. I'm not sure what you mean by "after the procedures are settled" as the bits I object to are in the current proposal - the procedures are presumably for things like the majority needed, what happens if there is a tie, etc. If this proposal is significantly amended at a later stage then my opinion of it might change. Of course there's then the issue of how it works out in practice. If BARC turns out like the Oppose section of some RFAs, with any flawed rationales likely to be picked apart by other editors, then I might even pick up the tools again, but only once I was confident that if I was off wiki when someone tried to BARC me there would be editors to argue my case well enough that my absence didn't matter. But the current proposal would limit that to the talkpage so if the target admin was not around an organised off wiki lynch mob would be difficult to stop at BARC. ϢereSpielChequers 18:42, 28 July 2015 (UTC)

Enough flesh on the bones? As far as I can see, this proposal consists of:

• A group of users who haven't expressed interest in the job, weren't elected for this task, and don't have the activity levels to carry it out;
• A group of users whose selection process is left entirely undefined, except for the proviso that they will be chosen by bureaucrats and not by the "community" so prominently featured in the title;
• A timeline that is physically impossible to meet by anyone whose life obligations involve tasks other than Wikipedia;
• An overall procedure whose level of operating "bureaucracy" is indistinguishable from Arbcom;
• A use case that doesn't appear to actually exist, and for which no examples, real or hypothetical, have been provided;
• Absolutely nothing about how this committee will accept or decide cases or what standards they would use.

I'm not trying to throw rocks here, but I just don't get it. There's no flesh. There's not even any bones. This is a cotton-candy proposal.

Worm says above in the next phase I'll be putting in some empirical analysis... OK, it's great that you plan to do some empirical analysis, but this is the part you do first. Do the analysis in an unbiased way (for example, present a proposed data collection scheme for review before collecting any data), decide if the analysis supports a need for any proposal at all, and then design a proposal based on the results of the analysis. Opabinia regalis (talk) 16:37, 28 July 2015 (UTC)

And worst of all: no real advantages over Arbcom, while having several deficiencies compared to Arbcom. An increase in bureaucracy (no pun intended), while less community focused and less accountable for sure. Supposedly faster, but having no good reasons for believing this will happen. --Jules (Mrjulesd) 19:04, 28 July 2015 (UTC)

Do the bureaucrats want this extra responsibility? And do they have the numbers to cope?

This process heavily depends upon bureaucrat participation. After all five are necessary for the process to work. And there aren't many of them, with 33 listed above. And most are not very active, with only 12 making at least an average of an edit a day (over the last year). And few are joining: I've yet to see a RfB take place, probably because of the very high standards expected. My question is: are the the bureaucrats ready to take on this extra work? And will this be the case in the future, with so few to choose from?

Maybe what is needed is a straw poll on bureaucrat viewpoints on this proposal, as their support will be deciding factor on whether this is workable. Obvious WormTT is favourable, but I've just noticed the bureaucrat WJBscribe opposed. --Jules (Mrjulesd) 11:30, 28 July 2015 (UTC)

As to the numbers to cope, the two recent crat chats (1 and 2) show that when something is going on for them/us, the crats can turn up. Worm^TT(talk) 11:59, 28 July 2015 (UTC)

Indeed, I think I count 14 bureaucrats participating in one or both discussions. That said, I noted on the talkpage that the participation of those who have been largely inactive as bureaucrats for several years caused some consternation. WJBscribe (talk) 16:21, 28 July 2015 (UTC)

That's supposing that 'crats will feel the same way about RfA and BARC, which may or may not be the case. Also it's a far more involved process to desysop, or at least it should be. Assessing consensus is less complex than assessing evidence. --Jules (Mrjulesd) 12:08, 28 July 2015 (UTC)

It is ironic that an RfC that is designed to address the fact that there are too few new administrators -- because qualified candidates believe with good reason that they don't stand a chance of passing -- has an issue with too few bureaucrats -- because qualified candidates believe with good reason that they don't stand a chance of passing. --Guy Macon (talk) 12:55, 28 July 2015 (UTC)

I was doing the math earlier and chortled to myself when I realised if this was an RfA it would go to a crat-chat and would probably (going on their past discussions) end up as no consensus.... Only in death does duty end (talk) 13:19, 28 July 2015 (UTC)

OK now we've got a second bureaucrat opposed to this, Avraham. And they're the ones supposed to take on this extra responsibility? Without bureaucrat support this proposal is worthless. --Jules (Mrjulesd) 15:31, 28 July 2015 (UTC)

Please let me clarify. I am opposed to the suggestion for the reasons I list, which are fundamental to its intent. However, should the suggestion become implemented, I will, of course, fulfil the duties to which the community has entrusted me to the best of my abilities. -- Avi (talk) 16:06, 28 July 2015 (UTC)

Avi is correct to draw a distinction between opposing the proposal and unwillingness to participate in the process if it is supported by community consensus. However, I think I fall into both camps. I am uneasy with this proposal, which isn't what I signed up for. If passed, it would fundamentally change the bureaucrat role. I also worry that it would undermine the community confidence that is needed for bureaucrats to perform their current functions - these BARC discussions are likely to polarising. Issues from BARCs may spill into RfAs in several ways:

1. Those who had tools removed by BARC seeking them back to the community; and
2. Candidates being asked to comment on conduct that has been adjudicated by BARC.

Such spillover risks eroding bureaucrat's neutrality (or at least our perceived neutrality) when it comes to determining the outcome of RfAs. I may be too pessimistic and I appreciate the hard work of the proposers, who I know have the best of intentions. But the potential implications of this policy are such that I don't think I could participate in BARC. WJBscribe (talk) 16:34, 28 July 2015 (UTC)

The points you raise, Will, are, as per usual, valid and pertinent. One's personal integrity cannot and should not be compromised, and if any of us feel we cannot perform these duties simultaneously with our others, then we will each be forced to choose which ones to perform and from which to recuse. At this point, I think it is possible to approach every case neutrally, and I believe that I would be able to review the community consensus about someone's RfX after a BARC removal with the same equanimity and impartiality we do with an RfX after an ArbCom desysop—each case being determined on its own merits. I would hope that if we continue acting as we have in the past, that the community would respect that and not begin to impugn our integrity, but only time will tell that. Let me just state that the very fact that these issues concern you is exactly why I think you would be able to perform admirably in these situations, should they come to pass (which we both have opined we would prefer that they do not). It is this integrity which needs to be the cornerstone of such a process, were it ever to be accepted by Wikipedia. -- Avi (talk) 16:57, 28 July 2015 (UTC)

(edit conflict; responding to WJBscribe) While that's a thoughtful position, of the issues raised with this proposal, I find this concern to be one of the less troublesome. For example, if five bureaucrats participate in a BARC decision desysopping User:X, those five 'crats might not be the right people to close X's subsequent re-RfA, but there would still be plenty of other 'crats who could do so.

I'm more troubled by the increasing likelihood that there won't be five reasonably active bureaucrats willing and available to serve in this role. Although not intended by the proposers, if the proposal is adopted and the survey for 'crat-member-volunteers comes up short, I foresee that we'll wind up electing a group of new bureaucrats specifically for the purpose of staffing this Committee. Newyorkbrad (talk) 17:01, 28 July 2015 (UTC)

Thanks for your views WJBscribe and Avraham. I think that for this proposal to succeed it needs to have strong support from bureaucrats as a group, and I have only seen lukewarm support so far. Wikipedia is a volunteer project, and thrusting extra responsibilities on groups, without them generally being wanted, seems unfair and unwise. I'm convinced more feedback is needed by bureaucrats before proceeding, to try to gather consensus on whether they want, as a whole, these new responsibilities. Any thought on how this might be gained? --Jules (Mrjulesd) 17:45, 28 July 2015 (UTC)

If this comes to pass, then I believe Brad is correct, we will need a slew of new bureaucrats, with a preference for those who have experience in these matters (namely, former arbcom members) and we will have just turned the bureaucrats into ArbCom-lite, which is not a desideratum in my opinion. -- Avi (talk) 18:11, 28 July 2015 (UTC)

Clarification on this RfC's outline

@Worm That Turned and Kudpung: just to make sure that I've understood the purpose of this RfC, this is the first of three RfCs on the subject, with the first to assess whether the community is open to a discussion of the proposal, the second allowing modifications and fine-tuning, and the third deciding whether or not to implement the proposal as policy? G S Palmer (talk • contribs) 12:32, 28 July 2015 (UTC)

The plan as I saw it was to see how the community felt about this sort of proposal, then if accepted, find consensus on the contentious points. If the final proposal is significantly different from the proposed, I'd certainly expect another phase - though perhaps not as a straight Yes/no RfC, but as a couple of alternatives, or perhaps a yes/no for a trial period. Firstly, I'd like to get through this phase and the next though. Worm^TT(talk) 13:30, 28 July 2015 (UTC)

I think that the problem that myself and a number of others have is that this proposal is too "bare-bones" for us to be able to throw ourselves whole-heartedly behind it, and we would rather wait until the details have been discussed and finalized before implementing it as policy. If this is going straight to approval, then it may be somewhat of a deal-breaker. I for one can see a number of significant changes that would benefit the process, and would rather withhold my support until I see if any of them work their way into the final draft. I do support the basic concept, but I think it needs improvement. G S Palmer (talk • contribs) 13:55, 28 July 2015 (UTC)

It seems a significant number of Support votes are under the impression that their vote is a vote towards a fuller discussion, not a vote towards implementing this plan with fine-tuning to be achieved by consensus in a later RfC before the actual implementation occurs. I certainly thought this was a sort of straw poll at first, and moved to Oppose when I realized that this is not the case. This should be taken into account when this RfC is closed. ~ Rob^Talk 17:54, 28 July 2015 (UTC)

Just to note

... that I'm aware of this discussion (and have been for some time), am watching with interest, but don't plan to participate before a decision is taken. I've always been happy to follow the community's consensus, including when the Crat role has been narrowed, rather than widened. --Dweller (talk) 21:12, 28 July 2015 (UTC)

Another proposal for Community desysop

I am not much into the wiki-politics and administrative side of Wikipedia, and never will be. I am also very certain this is entirely the wrong place to post this, so please accept my apologies. If the idea is not wholly abortive (and it may be, I don't know), others can move it wherever they want.

I have thought about this "community desysop" question, and wonder why it can't be done using elements somewhat similar to the Arb election process. I suggest this:

1. This process runs parallel to the existing Arbitration method of desysopping, and does not wholly replace it.
2. A bot can generate a sortable table of inactive admins and their length of inactivity, updated periodically. That doesn't mean inactive in using the admin tools; it means inactive as editors.
3. At an established time every 6 months, every established editor has the right (not the obligation) to anonymously nominate a maximum of 8 sysops for community desysop. [The system would retain the username and noms, but these would not be displayed publicly].
4. However, the nomination form is tailor-made with two key features: first it has radio buttons that only allow 2 reasons: Inactivity and Problematic behavior. Second, no more than 4 sysops can be nommed in either of the two areas. So each editor has the chance to nom only 4 sysops for being problematic. There is no discussion and no evidence --nothing -- in this phase. The form also includes a link to the aforementioned table of inactive admins. The whole nom process can take less than 10 minutes for any given established editor.
5. After a specified period (3 weeks?), the noms are closed.
6. Bureaucrats will silently discard "Inactive" noms that do not meet some clear guideline of edits per time period, plus those who have temporary amnesty (see below).
7. A simple list of nominations for desysop and the reason given are published. This list is without total number of noms per any admin, and without the names of those who nommed any given admin, just a bare list of noms and reasons. I know this will be embarrassing for some, as editors who have never had any problems with anyone will occasionally get nommed. This problem will work itself out (see below).
8. The list will then be sorted into two categories (again, Inactive and Problematic). Then each category will go through the same process as an Arb election, but the result will be desysopping. The time period for !voting and the consensus level for desysopping should be the same as that for electing an Arb. [This is where "discussion and evidence" enter the process.] Bureaucrats determine consensus as per usual, etc.
9. Every sysop nommed in any given 6 month period gets an automatic amnesty from being nommed again in the immediately subsequent period. They do not, however, have amnesty from the existing Arbitrated path (which is why the two paths run in tandem rather than one replacing the other).

To my mind, inactive editors will be weeded out very quickly. They can of course reapply via RfA if they return to editing. Silly noms of non-problematic admins will never pass the threshold... Finally, somehow or other, those sysops who are placed into the !voting stage (esp. if for non-serious reasons) should be encouraged to reply very sparingly, if at all, to any comments made in the evidence/discussion, at least initially. Hasty reactions are the catalysts of drama and interpersonal conflict; silence is a fire extinguisher. Let others defend you. • Lingzhi ♦ (talk) 02:31, 29 July 2015 (UTC)

I am sorry to be harsh here, but this proposal is just hopeless. First of all, at a time when there is increasing concern about (no pun intended) wiki-bureaucracy and rules-creep, the complexity of the process you describe is undue. Beyond that, your suggestion is effectively that any administrator would have to go through a reconfirmation/desysopping discussion and vote, perhaps each and every year, at the request of any one individual editor, who might not have any legitimate reason and who in fact would be without a vehicle for articulating any reason more than two words long. Um, absolutely no thank you. Newyorkbrad (talk) 04:32, 29 July 2015 (UTC)

My idea was that once a year is not too much, and silly noms can be ignored by the admin. This suggestion is 100% community-driven, creates a vent for this expressed desire, but makes it difficult to desysop. It's "Put up or shut up." But your concerns have merit. Tks. • Lingzhi ♦ (talk) 04:36, 29 July 2015 (UTC)

Another problem here, if another is needed, is that pretty much every reasonably active administrator has at least one editor who dislikes him or her or feels aggrieved by some comment or action of his or hers. If this process became popular, we could wind up with a situation where most or all of the several hundred active admins were up for desysopping discussions at once. Newyorkbrad (talk) 04:43, 29 July 2015 (UTC)

Not a problem for two reasons: six month amnesty makes noms staggered, and I repeat, admins who are in no danger of desysop can just ignore, ignore, ignore. Plus those who *are* in danger get a 6 month period of peace if the desysop effort fails. Again, "Put up or shut up." • Lingzhi ♦ (talk) 04:45, 29 July 2015 (UTC)

In addition to NYB's comments, there is the issue of using the Arb election method rather than a threshold of support method. The Arbcom election process differs from RFA in that you are electing a certain number of people, that is reasonable when building a committee, unreasonable when disciplining miscreants. With the election method you have a certain number to be found guilty/desysopped, sometimes the election may come after a major scandal and the election then becomes a search for the five ringleaders, and many who you'd want to desysop get off scot free sometimes nobody has done anything wrong and it then becomes and unpopularity contest with the least popular lynched. Then there is the issue of timing, we are volunteers, it isn't right to impose an intense period of activity on volunteers without them having the option to decide when they are available. ϢereSpielChequers 10:14, 29 July 2015 (UTC)

Thinking out loud

I had an interesting thought on a possible process to remove troublesome administrators. What I'm going for is a process akin to the procedure to remove featured articles. I think this is a balanced approach, and would like to get some comment on it:

1. Discussion (like featured article review): an initial discussion occurs at, let's call it, "administrator conduct review," (ACR) where the community gets to discuss the administrator's conduct. ACRs would be open to all members of the community, and discussions would go on there for at least a week. Then, three or four ACR coordinators decide whether there is something substantially wrong going on, or whether the user who opened the ACR was simply angry or aggrieved. If it is the former case, then it moves on to the petition stage. If it is the latter case, then the issue is declared resolved (Some points for discussion: Allow the aggrieved user to go to ArbCom afterwards, or it's over? What about double jeopardy and possible cases of sua sponte on the part of an ACR coordinator and how that would be resolved? Ideally, any ACR coodinator would have the trust of the community, and if they fucked up badly or turned out to be an undesirable individual, they can be removed by consensus. Additionally, on what grounds does an ACR coodinator have the ability to summarily close a case?)
2. Petition (like featured article removal candidate): The next step is, let's call it "administrator removal review." (ARR) Invovled editors are notified. Proper noticeboards (AN, AN/I, etc.) are notified. Then, the community is allowed, for a week at least, to vote for or against removing the administrator's bit. After an allotted time has passed, bureaucrats do what we elected them to do: observe and act from consensus. If there is consensus to remove, they remove. If there is not, they do not. Perhaps they maintain the same discretionary range at RfA (70%–80% is it?). The bureaucrats are allowed to open a crat chat, just like at RfA.
3. When to refer to ArbCom: if there are private matters involved (off-wiki harassment, etc.), then it's best we send that to ArbCom for reasons I hope would be obvious. Otherwise, at a supposed ARR, should the community be able to, say, admonish an administrator, or ban them from the wiki, as ArbCom does? If it does, it would cut out the need to go to ARR for a desysop but some other place for a ban, thus further simplifying the process.

In my opinion, this procedure is simple, lightweight and no more bureaucratic than Wikipedia's extant procedure to remove featured articles (which seems to work quite well, so perhaps we have a test case right there). Thoughts? Cheers, --ceradon (talk • contribs) 05:49, 29 July 2015 (UTC)

To a man with a hammer every problem is a nail, one of the advantages of Arbcom is that it has a whole toolkit of responses to situations, it can admonish, ban, desysop, topic ban or indeed exonerate individuals. If you create an arbcom lite and only give it a limited range of options then one of the downsides of that solution is that you will sometimes get a worse decision than Arbcom would have made. So if you really think Wikipedia needs the extra complexity of an alternative to Arbcom, and you want that alternative to be at least as good a system as Arbcom, then it needs as varied a set of sanctions as Arbcom has. ϢereSpielChequers 10:01, 29 July 2015 (UTC)

• Yeah, this is something that concerned me as well: Some admin misbehaviour needs stronger remedies than a mere desysop, compare the Wifione debacle. BARC would either need to remand such issues or be authorized to install other restrictions as well. Jo-Jo Eumerus (talk, contributions) 10:26, 29 July 2015 (UTC)

• Worth also considering what happens when an allegation is made which on investigation demonstrates editor misconduct by the alleger or by other third parties rather than the accused (witness the recent Occultzone Arbcom case as a recent example). Does BARC also have authority to impose sanctions on non-admins if the BARC case warrants it, or is its role a simple "yes/no" on admin misconduct with all other issues remanded back to ANI? Presumably these are questions for the stage two RFC, so views welcome there when it gets off the ground. -- Euryalus (talk) 03:38, 7 August 2015 (UTC)

Closers

I'd like to be a closer in this one. As always, I encourage anyone who's uncomfortable with that for any reason to say so, and if we get one or two more closers (which would be a good thing), I hope they'll be willing to ask for the same kind of feedback. Closers should play the smallest possible role that gets the job done, so I encourage the participants to find a way to address various questions that have been brought up, including the canvassing issue, so that the closers don't have to. (And by the time it gets to the closers, it would be too late to do anything about potential canvassing anyway.) RfCs like this one have to run at least 30 days, and it's my hope that we'll be able to finish this up without any unnecessary delays, so my current plan is to start talking when we enter the last week of the 30-day period, but I'll defer to the other closers on that (if we get them). - Dank (push to talk) 11:56, 3 August 2015 (UTC)

I am wondering what standard of consensus will be used? For example should a new policy for de-sysoping people be expected to have at least the level of support that one needs to pass RfA? I counted yesterday and it was around 68%, it may have changed since then. Numbers aside I think people on both sides have made strong points. Chillum 13:57, 3 August 2015 (UTC)

Based on the current 'crat chat, the support percentage would need to be higher than 74%. Risker (talk) 12:33, 5 August 2015 (UTC)

I disagree, Risker. It is not a pure percentage measure, but the nature of the supports and oppositions which are important as well. If all we followed was rote percentages, we could use a script. There needs to be a range in which discretion can be used. -- Avi (talk) 14:21, 5 August 2015 (UTC)

Possibly if there is enough shouting and insults of other editors led by a former ArbCom member, the crats will blow in the wind and come down your way. I'm not inclined to expand crat functions when we saw yesterday that bad faith in a bad cause can be successful. Once.--Wehwalt (talk) 07:48, 6 August 2015 (UTC)

If there is going to be multiple closers, it would be ideal if that group was composed of both administrators and non-administrators. –xeno^talk 14:10, 3 August 2015 (UTC)

Absolutely. And a panel of 3 is a very good idea. Dennis Brown - 2¢ 14:11, 3 August 2015 (UTC)

Chillum: That's another question that I'm likely to defer on ... thanks for offering an opinion on this, and I hope other voters will, too ... but this much I can say from previous experience: if the vote were anything close to 90% support or 50% support, and I tried to push back against the vote count, that would be seen as an attempt at nullification of the voters by an over-eager closer. Conversely, for a vote count near 70%, to claim that the vote count itself decides the issue would be seen as dodging my responsibilities as a closer. Xeno and Dennis: as I mentioned today at WP:AN (search for Dank, it's my only comment), I used to put up a notice at WP:AN and elsewhere asking for extra closers in RfCs like this one. More closers would be good, but I'm not going to solicit them any more, in part because I'm trying to avoid too much involvement, and in part because people sometimes think it's something they want to do at first and then lose enthusiasm when it comes time to do the work. Bottom line: I'd like more closers, and if it happens, it happens. I don't want a closer who (probably through no fault of their own) becomes an issue in their own right. - Dank (push to talk) 14:19, 3 August 2015 (UTC)

On the topic of the support percentage, I think it's worth considering that if the oppose voters are primarily opposing particular details that that could be noted as something to carry through for further discussion in the next phase. We already have consensus for a desysopping process so I think it's worth treating this RfC phase as one to be concluded with support or opposition of the general idea bearing in mind that this isn't meant to be the only and only RfC on the proposal. Sam Walton (talk) 14:31, 3 August 2015 (UTC)

I have a real problem with that idea, Samwalton9. A very significant percentage of the support votes *obviously* are supporting a "community desysop process in principle" while having issues with this particular version of it. You have written as though the only job for the closers is to decide whether there is consensus for this proposal to be the community desysop process. I disagree. In fact, I do not think the previous "consensus" is valid any longer; it's been years and the project has changed considerably since then. Risker (talk) 15:19, 3 August 2015 (UTC)

Really? It's only been 3 years. Given that the RfC includes a statement that the community has agreed a desysop process is needed I've read the votes in the context of this being the particular one which should or shouldn't be used. I can't imagine that I'm the only one so this should really be clarified before anything is closed. Sam Walton (talk) 17:47, 3 August 2015 (UTC)

"Only three years" is a very long time in the wiki-world, and there are almost twice as many participants here as participated in the prior discussion. Many "support" votes actually don't support the proposal being made, although they support the general idea of community-based desysop. The proposal is so long and involved, and the "official" question so difficult to ferret out that it leaves a very bad taste in my mouth. Risker (talk) 20:09, 3 August 2015 (UTC)

I'd be willing to join in as a closer. But I agree that a three-closer panel would be most beneficial. So, now we would need a non-admin (if you're okay with me assisting in the closure.) --ceradon (talk • edits) 14:25, 3 August 2015 (UTC)

As a potential closer, I can't be involved in picking the closers, so I can't say anything either way. - Dank (push to talk) 14:31, 3 August 2015 (UTC)

I think a Committee of Five (or more) is required, and they should open a "closers chat" (analogous to a crat chat in case of contentious RfAs, the closers propose and debate the closure on the project page, commenters stay on the pertaining talk page. Kraxler (talk) 14:33, 3 August 2015 (UTC)

Well ... I can't pick the closers but I've got an opinion on that. The only two times that come to mind when we had more than 3 closers were for two different Pending Changes RfCs ... from memory, one took more than a month to close and the other took more than four months, in both cases because not everyone was available at the same times. If you want more than 3 closers, then the community should get to work on solving the problems that arise with more than 3 closers. On your other point ... yes, yes, and yes, I welcome any format that provokes an active role in closing by as many people as want to comment. - Dank (push to talk) 14:39, 3 August 2015 (UTC)

What I said here was just quoted on the RfC page to support a different point ... that's a valid opinion, but it's not my opinion, and I didn't mean to imply here any criticism of what's being proposed in the RfC. I don't believe that committees of greater than three are always slow. It's just the haphazard way that we assemble closers for RfCs that has given non-optimal results with more than 3 closers so far, which is probably why we've only done it in two cases for CENT-worthy RfCs that I can recall. But ... whatever you guys want will be fine. - Dank (push to talk) 16:44, 3 August 2015 (UTC)

I refactored the comment in question, as I see how it could be read wrongly and the point I intended was a general matter of practicality, not related to specific potential closers or comments posted in this thread. Opabinia regalis (talk) 17:33, 3 August 2015 (UTC)

(edit conflict) Five is perhaps too many cooks, and might just needlessly prolong the process of closing this. I do like the idea of "closers' chat" though. --ceradon (talk • edits) 14:42, 3 August 2015 (UTC)

Only a little request, please set a decent timeline for the close. Cavarrone 17:07, 3 August 2015 (UTC)

The default duration is 39 days, so it's still got quite a while to run. Doug Weller (talk) 17:10, 3 August 2015 (UTC)

Yes, and I am fine with respecting such a duration. I was referring to the process of assessing (and writing) the close after the standard duration expired, per what Dank said here. Cavarrone 17:33, 3 August 2015 (UTC)

Hey Dank, how's it going? : )

Just thought I'd drop by and note I wouldn't be helping close this one (which of course should be obvious since I commented here, but still... : )

It might not be a bad idea to drop a note on WP:BN to see if anyone there might be interested in closing. - jc37 23:04, 3 August 2015 (UTC)

Good to see you, I'm sure there will be another one along soon that could use your talents. - Dank (push to talk) 01:23, 4 August 2015 (UTC)

Many of us (bureaucrats) have opined one way or the other here, FWIW. -- Avi (talk) 23:22, 3 August 2015 (UTC)

Given that this so deeply involved the 'crats I think it makes a lot of sense for one of them to be involved in the closure(If any active ones have not already commented). Chillum 23:27, 3 August 2015 (UTC)

• I think that it should be left open for an absolute minimum of two weeks, so this is a bit premature. Stifle (talk) 08:49, 4 August 2015 (UTC)

If there are to be multiple closers, I strongly request that any discussion between them about the result should happen onwiki. WJBscribe (talk) 08:59, 4 August 2015 (UTC)

• I'm comfortable with that. - Dank (push to talk) 11:48, 4 August 2015 (UTC)
• No problem at all. --ceradon (talk • edits) 11:49, 4 August 2015 (UTC)

• Given the blatent canvassing at the start of this discussion I'm astonished anyone thinks this isn't too tainted to reach a conclusion... Spartaz ^Humbug! 11:50, 4 August 2015 (UTC)
  • The voters will have to find an amicable way to resolve that issue. By the time the closers get involved, it's too late to do anything about it. - Dank (push to talk) 11:57, 4 August 2015 (UTC)
    • On the contrary, this is something the closers should take into account. Chillum 14:22, 4 August 2015 (UTC)
      • I agree -- this isn't anything "the voters" can easily "resolve". Whether the concerns about canvassing are warranted and how much (if at all) this has influenced the poll is an important question, but the 150 participants of the RfC are not the people who need to do the research to find out about this. It is, however, the first question a closer has to answer: is this RfC valid and has it been run in a regular way? —Kusma (t·c) 14:41, 4 August 2015 (UTC)
        • I realize that, for someone who supposedly won't start talking until the last week of the RfC, I'm doing a lot of talking, but there are just a few items where it's going to be way too late to say anything if I wait. This is one of them. I don't disagree that all of us, including the closers, need to be vigilant about non-representative votes, and whether important votes are tainted by the participation of people who "shouldn't be here", according to whatever the voters think the standards are for that. But I can't lend my support to any process that invites everyone to make major commitments (of time, and of themselves, and even a little bit of their reputations) only to tell them at the end, "Sorry, we're throwing this out on a technicality, you can all go home now". Closers have not been elected, and don't have access to some special font of wisdom. Voters are equal participants and have equal responsibility to make the RfC process work and to fix things that are broken, while the RfC is ongoing if possible, out of a concern for the value of other people's time if for no better reason. Ceradon, any thoughts? - Dank (push to talk) 15:01, 4 August 2015 (UTC)
          • Many people have significant problems with this RfC. From the unresolved canvassing question to the problem that many of the "supporters" seem to oppose parts of the proposal as written, and many "opposers" support some kind of community desysop process (but not this one). That this RfC is a bit of a mess is not a technicality -- it is the reason it is going to be quite difficult to close fairly and properly. If you are not up to the task, perhaps you should leave the job for somebody else who isn't already accused of being involved. —Kusma (t·c) 05:42, 5 August 2015 (UTC)
• Dank, with all respect; I believe you should not be involved in closing this RfC. You have a prior working relationship with Kudpung, he being one of the top ten contributors to your talk page, and given Kudpung's reaching out to you to be involved in the RfC (User_talk:Dank#Community_desysoping) even before it opened, and given your being the second signee to the RfA reform task force [1]. You are clearly involved, and are coming from a biased (if unintentional) perspective. We need people who have not been involved in this at all. --Hammersoft (talk) 17:46, 4 August 2015 (UTC)
  • A little help here? Can anyone remember where I've ever expressed any opinion, positive or negative, on any desysopping process? I can't, and I can't remember that I've even had a position on that, I've tended to shy away from the whole issue. If you search for "Dank" on that page you're linking to, you won't find my signature (I only signed on at first, when it was a very general "let's make RfA better" effort). Lots of people reach out to me for lots of things; I don't agree that makes me biased. I don't see any names among the supporters that I could be said to have some kind of special relationship with. - Dank (push to talk) 18:40, 4 August 2015 (UTC)

• We've already had problems with canvassing taint on this RfC. We do not need more problems in that arena. There is no pressing need for you to be a closer on this. You encouraged us to voice concern if we were uncomfortable with you being a closer. I am doing so. You are obviously close to the subject and are involved. We don't have to have you as a closer on this. --Hammersoft (talk) 18:52, 4 August 2015 (UTC)
  • Yes I did, and I'm glad you spoke up since you have concerns. So, here's how this works: if you or others come up with something more than "Kudpung reached out to you, and he's a very bad person, so that taints you", then I will happily step aside. I'm really not seeing the argument, but I welcome feedback from anyone on this. Show me a diff where I'm taking a position on any desysopping procedure. I know that I've always shied away, I know that I've never had a position, but it's possible that many years ago I contributed in some way to some kind of conversation on the subject, I'll be happy to have a look. - Dank (push to talk) 19:13, 4 August 2015 (UTC)

• I never said Kudpung was a very bad person. I reiterate what I said above; we do not have to have you as a closer. You are clearly involved, and there are plenty of other people who are not involved who can handle this task. --Hammersoft (talk) 19:30, 4 August 2015 (UTC)
• For more evidence of the inappropriateness of you being the closer of this RfC; The proposer of this RfC said you are "the principle architect of revived efforts to do something about RfA" [2]. This is buttressed by the fact that your presence is all over Wikipedia:Requests for adminship/2013 RfC, being one of the primary authors behind that effort. I applaud your efforts and in no way seek to demean the enormous amount of work you put into those processes. However, it clearly shows you are directly and keenly involved in RfA reform. We do not need someone with such a strong involvement (nor indeed ANY involvement) closing this RfC. --Hammersoft (talk) 19:58, 4 August 2015 (UTC)
  • Um ... are you saying you're expecting RfA reform to come out of this RfC? Is that what people are expecting here? I thought this was about a desysopping procedure for current admins. Have I ever been involved in that subject in any way, or encouraged Kudpung or anyone else on this subject? Is there any evidence of any inappropriate relationship with anyone involved here? - Dank (push to talk) 20:57, 4 August 2015 (UTC)

• The direct tie between this proposal and RfA is stipulated in the very first bullet point under "Benefits" on this very project page. See Wikipedia:Administrators/RfC_for_BARC_-_a_community_desysopping_process#Introduction. Further, one of the joint proposers of the RfC stated "there is a significant portion of the community who believes it should be difficult to pass RfA because it is difficult to remove the sysop right, that was a strong finding from RFA2011" [3]. This proposal and RfA reform are inextricably linked. I'll say, yet again, there is no reason why you must be a closer on this RfC. You asked for people to voice their dissent, and now when we do you're taking us to task for doing so. I find this surprising. Dank, I respect you, but you need to step aside from being a closer on this process. --Hammersoft (talk) 21:10, 4 August 2015 (UTC)

• If you dont recognise that being approached by someone who is accused of canvassing this RFC in connection with this very RFC makes you unsuited to close this RFC then its no surprise that you are already disciunting the blatant canvassing. Whatever the result anyone disagreeing will always question the close. This is why you shouldnt close this. Spartaz ^Humbug! 18:59, 4 August 2015 (UTC)

  So far, there was no consensus that it was canvassing.--Ymblanter (talk) 20:18, 4 August 2015 (UTC)

Could I get some help here? I respect Hammer's positions and always have, but I don't believe that my occasional involvement with RfA reform (in which I've tried to remain as neutral as possible anyway) has anything to do with desysopping procedures, and I'd be surprised if people were expecting to get RfA reform out of this RfC. Any thoughts? Does my past RfA reform work disqualify me here? If you want me to stay on, then it's looking like I better stay quiet until the RfC is over. - Dank (push to talk) 21:23, 4 August 2015 (UTC)

Since most voters in RfCs don't keep a watchful eye on the talk page, I'll give this another 48 hours before I officially bow out ... but with 3 opposed and none supporting, from past experience, it looks likely that I'll have to bow out. I have always pushed back against the things that are broken with the RfC system in general, and one of those things is the willingness of the community to waste everyone's time on things that can't possibly work, for one reason or another. (If canvassing occurred to such an extent that it makes the vote count irrelevant, then that would of course be fatal to any RfC ... I haven't followed the diffs yet to come to my own conclusion, because it's not the proper time for a closer to be doing that, the voters should be doing that, out of respect for each others' time, in this and every other RfC where the issue comes up. Or, someone might want to propose some kind of committee to help with certain "fatal" problems that come up early in some RfCs, if you'd rather shift the burden away from the voters.) The fact that I've tried for years to make RfCs work better has always meant that it was just a matter of time before I came to be seen as too much of an activist to be appropriate as a closer, and maybe that time has come, I don't know. I do know that, in the many years I've been trying to help with various problems regarding adminship, I've been completely unsuccessful, so I should probably concentrate on things I'm better at, at this point. - Dank (push to talk) 11:59, 5 August 2015 (UTC)

Personally I'd support having you as a closer Dank, I've seen you close a number of times and every time you give a thoughtful and thorough rationale. I certainly don't see the bias. Worm^TT(talk) 12:09, 5 August 2015 (UTC)

Dank, I think a lot of your work, but I think that it's probably a good idea to pass up the "opportunity" to close this one; you've closed just about every "contentious" RfC we've had in the past several years, and it's probably time to let for others step to the fore here. You have a long history of work in the RFA reform area, and this RfC explicitly states that the proposal is an element of the larger RFA reform process. It's okay for you to participate in some of these discussions instead of closing them; I think your voice here could be more valuable as a participant than as a closer. Risker (talk) 12:41, 5 August 2015 (UTC)

• Dank, there is no reason whatsoever to extend any thoughts people have of you closing this RfC onto the excellent work you have done in RfCs in general. I know for my part my comments are strictly directed only at the thought of you closing this RfC. Your involvement in RfCs in general is not something I have considered except where it has to do with adminship. My reasons for my opinion that you should not be involved in the close of this RfC is summed up in your own words "in the many years I've been trying to help with various problems regarding adminship". It need not be any more complicated than that. --Hammersoft (talk) 12:50, 5 August 2015 (UTC)

Okay. I don't feel hounded here, I really appreciate the support, I agree with the point that I might be seen as involved, and Risker makes a very good point about my involvement in so many RfCs. That's enough for me, there's no reason to drag this out. I'm out. Just two things: 1. As a copyeditor, I suspect some are going to misread "it's probably time to let others step to the fore here" to mean "it's probably time to let others step to the fore", as if I've been shoving them aside all along; I think the record will show that for many years I and others have been begging people to close the more contentious RfCs without much luck, and I've deferred to other closers when they did sign on wherever possible. 2. I get that others think there's a connection to RfA reform here, but I don't think this RfC will affect RfA one way or the other, and that's why I thought I was neutral enough to close here. After the RfC is over, I'll tell you why I thought that. Thanks again. - Dank (push to talk) 13:12, 5 August 2015 (UTC)

Thank you, Dank; I've made a bit of a copy edit to indicate that I'd like to see others take on this role, rather than suggest that you're doing anything to prevent them from doing so. I'll look forward to the post-RfC discussion about your opinions - although in fairness I think it would be useful to the community to have someone of your knowledge and experience voicing those opinions *during* the RfC rather than as a postscript. Risker (talk) 13:52, 5 August 2015 (UTC)

I'm sorry, I really can't comment. - Dank (push to talk) 14:46, 5 August 2015 (UTC)

I think the question of closers should be left until we are nearer the close to allow for the participation of those who have not yet expressed their views on the proposal.--Wehwalt (talk) 14:38, 6 August 2015 (UTC)

Alternative RfC

I've drafted a proposal for a different process for de-adminship, based off of the process at the German Wikipedia. It relies more heavily on the community, does not form any new committee, and utilizes our existing policy/infrastructure at WP:RfA. The proposal can currently be found at User:BU Rob13/RfC for Administrator Re-election. If you are interested in working on the proposal, please let me know. My intention is to hopefully put it out there and request comments somewhat soon, but I will not do that at all unless at least a few experienced editors support the idea of putting out this RfC. I'm a relatively new user (2 months), and do not have the experience to feel comfortable releasing it by myself. If anyone is interested in the proposal or has any ideas on how to improve it, please feel free to edit the page or discuss it on that proposal's talk page. I'm hoping this is an option that addresses many of the concerns of the opposes in this RfC.

Beyond requesting help on formulating this, I have a procedural question. My thoughts on how a second proposal should be handled is that it should go through a first round RfC similar to this, where the question is "Do you support this proposal when compared to the current situation?" That question mirrors the one asked in this RfC, where support of this proposal is somewhat conflated with opposition of the current system or support for reform. If both proposals were to reach consensus using that question, then a second RfC could be used to decide which of the two should go forward to an RfC where the details are decided. Does this seem wise? I think it's the best way to handle two RfCs when the proposers of this one have clearly indicated they do not wish another proposal to be inserted here. ~ Rob^Talk 19:51, 5 August 2015 (UTC)

RfC timetable?

My apologies if I've missed this somewhere on these increasingly lengthy pages. For those of us who are still thinking through our positions on the proposal—or have a general sense of our positions but are figuring out how best to express them—is there a specific closing date scheduled for this RfC? If not, can we at least have the assurance of a couple of days' heads-up before it is closed? Thanks, Newyorkbrad (talk) 19:56, 5 August 2015 (UTC)

It was proposed at this talk page that the RfC gets closed after 30 days, but so far there are no closers, so I guess it will stay open at least couple of weeks, and possible until the end of August.--Ymblanter (talk) 05:48, 6 August 2015 (UTC)