User talk:Casliber/Fivecrat

Broader application

We often have a need for a committee of this sort, however they are currently formed and dismantled as the need arises. Wikipedia:Administrators' noticeboard/Incidents/I have blocked Betacommand was an example that quickly comes to mind. More examples would be good, to see if there is a pattern. John Vandenberg ^(chat) 08:58, 27 November 2008 (UTC)

Concerns

Two situations where this process would be employed are currently mooted. Case (a) is for straightforward cases of admin misconduct, on the referral of the ArbCom. If the ArbCom has to review the case anyway – sufficiently to determine that the case is 'straightforward' enough not to require their full attention – why couldn't they just handle it rapidly themselves? The ArbCom does not appear to be inundated with admin misconduct cases at this time, nor have they shown an inability to respond in a timely manner to straightforward cases of admin misconduct. (In egregious cases of misconduct, they have not hesitated to use their power to request emergency desysopping prior to a case even being filed.)

Where ArbCom does not act rapidly to desysop is in cases where the degree of misconduct is ambiguous, where there exist potentially extenuating circumstances, where there are long-running disputes between multiple users, where the evidence is extensive or complex, where the putative misconduct takes place within the framework of a much larger dispute, and/or where community standards and expectations are unclear. In none of those cases would referral to a fivecrat committee be appropriate. (See also Wikipedia:De-adminship proposal checklist.)

Part (b) of the proposal addresses Arbitrator misconduct. Have there been cases in the past where a sitting Arbitrator has engaged in misconduct that – for another editor – would be deserving of Arbitration? What, if anything, was done?

It should be noted that the perception that Arbs are invulnerable to censure is a mistaken one. It is at least possible that Jimbo retains the power (de facto, if not necessarily de jure) to fire an Arb outright — particularly if the community backs that decision. It is definitely true that the Board of the Wikimedia Foundation retains the final say in how the English Wikipedia operates. They could, if the need arose, pass a resolution to unseat an Arb.

I am also concerned by the privacy implications of opening Arbitration records to 'crats. 'Crats are not currently required to disclose their identities to the Foundation, nor did the community evaluate them with respect to their abilities to judge what amount to Arbitration cases. Access to private information (which may include real names and addresses of other editors, Checkuser data, and Oversight records) is too far beyond the scope of their current responsibilities. We would be asking 'crats to (in effect) act as Arbs once in a blue moon, without prior experience, in situations likely to be of the utmost delicacy and sensitivity, exposed to the most intense and unpleasant scrutiny.

On a philosophical note, I'm troubled by the postscript to this proposal. 'Cratship isn't, and shouldn't, serve as a "recognition of merit and honour". That's what barnstars are for. As well, I fear that the dramatically expanded responsibilities proposed here will not increase the number of 'cratship candidates. Rather, it will turn each RfB into a miniature ArbCom election. Few – if any – candidates will receive a consensus. Finally, RfB outcomes are currently decided by the existing pool of bureaucrats. Would we be happy if the ArbCom were allowed to choose its own members? Would we be setting ourselves up for judicial pissing contests, where the ArbCom decided to start stripping 'crats of their powers for abuse of process? TenOfAllTrades(talk) 09:29, 27 November 2008 (UTC)

Wow. Many very good points, especially the last one.

To make case (a) workable, I think it needs to turn on its head. Arbcom could review the decision made by the panel of five 'crats if a community member felt that the decision was unfair. This would mean that cases like the desysop of uncommunicative sysops would be routine (i.e. User:Can't sleep which went to a motion), and it would also give the community an opportunity to work with the 'crats to steadily develop a workable set of "clear desysop criteria". John Vandenberg ^(chat) 09:54, 27 November 2008 (UTC)

Ten, you make some good points and enough to radically change or ditch the idea/proposal as is, although how about flipping it? Cheers, Casliber (talk · contribs) 10:02, 27 November 2008 (UTC)

I think that the handing the 'crats the 'clear-cut' cases run into the same objection that I raised with the original proposal — that is, the ArbCom is already capable of dealing with those simple cases rapidly and effectively, so the savings of time and effort would be minimal. Meanwhile, giving the 'crats jurisdiction over more complicated cases – or even further expanding the scope of their mandate to develop 'desysop criteria' – seems to be asking for trouble. They haven't been vetted by the community to fill what amounts to the role of Arbitrator, nor have we empowered them to make policy.

Letting the 'crats have the first kick at a case with the option of a subsequent appeal to ArbCom means, I suspect, that the controversial cases will all get bumped to ArbCom eventually anyway, nullifying any time savings provided by a streamlined fivecrat process. While handing the ArbCom a pre-investigated case may somewhat simplify and speed the the evidence-gathering phase of their process, they may find their hands tied to an extent when presented with a pre-established decision. Overturning the fivecrats call is likely to upset a lot of people.

Possibly worse, the ArbCom might not choose to step in, leaving us with mob-rule-driven desysoppings. Despite occasional claims to the contrary, the current role 'crats fill in the RfA is almost entirely that of vote-counting robots. Witness the recent flurry of criticism on WP:BN when Nichalp dared to identify votes in an RfA that he (as a 'crat) thought didn't provide sufficient supporting rationale and would be discounted.

This proposal would ultimately be very similar to one rejected last month at Wikipedia:Removing administrator rights/Proposal. I won't (further) rehash the arguments that were presented there, but the talk page is instructive. That proposal is what ultimately led me to draft the De-adminship proposal checklist, which I most humbly suggest would also be a good reference. TenOfAllTrades(talk) 15:46, 27 November 2008 (UTC)