User talk:Barkeep49/Bureaucrats

Functionaries

Considering Wugapodes' point about the different selection criteria for crats, arbs, and CUOS, how about changing the backup plan from making "Checkusers and Oversighters" responsible for crat work, to making "functionaries" responsible for it? That's a wider and slighter more diverse group since it includes former arbs, who I think are more likely to have overlapping skillsets with the current crats than the average CU/OS. It also makes it possible to add the current crats to the new group for continuity, without having to grant any advanced permissions. And those crats that are already CU or OS wouldn't have to jump through the arbitrary hoop of a minimum number of checks/suppressions a year to keep participating in crat chats.

This could also edge "functionaries" towards being a more of a cohesive user group and less a prompt for the question "why do we have this mailing list again?". That is, they're experienced users trusted without private information, who will pick up and put down CU, OS, and crat permissions as needed. – Joe (talk) 08:46, 4 May 2022 (UTC)

• I think this is interesting, but I'm still suspicious of the assumed problem. If I can be melodramatic for a second, my favorite play is Julius Caesar. In 1.2, Cassius asks Brutus to consider the rising power of Caesar and their own role in the maintenance of the republic:

Men at some time are masters of their fates:

The fault, dear Brutus, is not in our stars,

But in ourselves, that we are underlings.

Brutus and Caesar: what should be in that 'Caesar'?

Why should that name be sounded more than yours?

Write them together, yours is as fair a name;

Sound them, it doth become the mouth as well;

Weigh them, it is as heavy; conjure with 'em,

Brutus will start a spirit as soon as Caesar.

Now, in the names of all the gods at once,

Upon what meat doth this our Caesar feed,

That he is grown so great? Age, thou art shamed!

Consider this allegorical to our situation. The role of crat, perhaps, is like Caesar: exalted beyond the bounds of the republican order. Deified and seen as beyond what mortals could attain. Joe and Bureaucrat: what should be in that "Bureaucrat"? Write them together, yours is as fair a name. Conjure with them, Joe will start a spirit as soon as Bureaucrat. Upon what meat doth these our Bureaucrats feed that they are grown so great? Age, thou art shamed! Perhaps the fault, dear Joe, is not in our stars, but in ourselves?

To pull back from the brink of theatre-kid edginess, what if we are the problem? We understand that the problems with RfA and adminship require more editors to run for adminship; why is that not the case with RfB? The concerns is that crats are too old and out of touch with community norms...and the very people who would remedy that with an RfB are instead thinking about how to abolish the group. Why is "run an RfB" so easily dismissed? None of this is to say abolishing or reforming cratship is a bad idea; if you had asked me three days ago I would have agreed. My point is that for all the costs and benefits of various reform proposals, maybe the least costly and most beneficial is the one we've already ruled out from the beginning? — Wug·a·po·des​ 23:47, 4 May 2022 (UTC)

@Wugapodes I supposed I'd be interested to see what the 'crats are getting wrong due to being so "out of touch" - was there any community consensus of that being a problem documented? — xaosflux ^Talk 23:55, 4 May 2022 (UTC)

@Xaosflux I think it's more a political problem than a material one. I've said (somewhere) that I don't think there's any actual issue in terms of judgment, but editors seem genuinely concerned that contentious RfAs are decided by a combination of (1) editors who are not frequently here (2) editors who uniformly created accounts a decade ago (3) editors who (with one exception) passed RfA a decade ago. In considering who is participating in RfAs and who is deciding them, the profiles don't line up. Whether this actually affects judgment doesn't, to me, seem like the issue at play. It's that the lack of representation has created a question of political legitimacy that has sparked so many reform proposals in the last few days. — Wug·a·po·des​ 00:15, 5 May 2022 (UTC)

Maybe Barkeep sees it differently but, as I understood it, this isn't coming from any criticism of the crats, just from the observation that they don't have much to do. Probably that's why people don't bother with RfBs any more. I've suggested this because I think the same can be said of other branches of the cabal: except for the CUs, and all the time arbs waste on routine CU block appeals, the functionaries are also underemployed. We're maintaining all these separate processes and user groups for a small set of tasks that are basically all in the same vein (handling nonpublic information and managing advanced permissions) which yeah, isn't a huge deal and probably isn't doing any damage, but the inefficiency niggles me. So maybe pooling them together would revitalise things a bit? – Joe (talk) 09:07, 5 May 2022 (UTC)

One committee with several...numerous even! subcommittees then  :) "In my Father's House, there are many mansions", as the fella said. SN54129 09:16, 5 May 2022 (UTC)

Yes, and I'd said as much a few days ago. Where my thinking is changing is around the assumption that groups with rare usecases should be reconfigured. My thinking around this is still in flux, so I may disagree with myself in a few days, but I think it's worth digging into why we're uncomfortable with maintaining these processes and user groups. To throw out a counter-narrative: granular permissions for discrete tasks make it easier for editors to tailor their permissions to their editorial and administrative interests. It's the classic argument for unbundling: not everyone wants to run an RfA in order to use suppress-redirect, so we split it off and evaluate editors on their ability to do that one task and do it well. The alternative is that plenty of people who could do a specific job well don't get to do it because the permission is wrapped up in lots of other unrelated considerations---I don't really care if a page mover knows how to edit templates, and it would be bad if we yanked page mover from those without TPE. For Crat, CU, OS, and Arb the workload is specialized because for each one we want to make sure that the people doing their respective tasks are able to do it well. What practical benefit is there in adding to the decision criteria for joining any of those groups? Will RfBs become easier or more frequent if we also need to consider whether we trust the candidate with access to non-public personal data? Will the CU corps be expanded or improved if we need to consider their ability to contribute intelligently to an eventual crat chat? There's overlap in all of these groups, but that overlap exists because at each point we only needed to ask a singular question: is this person able to do this specific job and do it well? That task might be rare or narrow, but that isn't necessarily a bad thing as long as it is done well when the task needs done.

Now, maybe that feeds into why people don't want to run RfBs, but to the extent that crat tasks are rare, I would argue it's a symptom of the lack of RfAs. Crat tasks are causally linked to RfA, and I don't think it's a coincidence that the bulk of our crats come from periods where we historically had scores of RfAs each year. That's not to say fixing a symptom is necessarily bad, but we should consider the consequences in light of other community reforms. We're working to fix RfA, recruit more candidates, and generally increase the throughput of that process. If those reform efforts succeed, what are the consequences should we also implement this restructuring for crats? Right now there aren't many RfAs so there isn't much to do, but if RfA reform succeeds then presumably there will be a fair bit to do. If we also succeed in increasing crat workload or push it onto other already busy groups, we risk overworking the crats and winding up with similar problems like those seen in Money's crat chat. This inefficiency could, ironically, undermine RfA reform by adding an additional pain point. This is all hypothetical, but it's worth considering at least: in what situations might our plans backfire? Will we be prepared if the lull in demand is only temporary?

So, for me, all that leads me to question why we're treating the lack of RfBs so differently from the lack of RfAs. The simplest solution, to me, seems to be running more RfBs. While there may be various reasons people are hesitant, we should tell them to run anyway. If we view RfB as a way to build capacity for the success of RfA reform, or if we view it as a potential avenue for implementing or supporting new kinds of reform, then we have compelling arguments to convince new candidates and the community of their need for the wrench. Overall, I'm not against structural solutions, but I worry we haven't seriously explored social solutions let alone exhausted them. — Wug·a·po·des​ 20:05, 6 May 2022 (UTC)

FWIW, running parallel to my work here, I've also been trying to find RfB candidates (and think I probably have found 1) so I'm not opposed to "just find people to run" idea but also I'm not sure that's actually a justification for keeping a role whose current workload is so radically different than when it was conceived. Best, Barkeep49 (talk) 21:31, 6 May 2022 (UTC)

Transparency

As a crat who has neither C/U or O/S rights I admit I have a vested interest here. All three groups are indeed trusted, and we crats are currently underused. But by necessity C/U and O/S operate in secret and with a very active back channel. Crats work in public and don't have a back channel. If you want that to continue there may be some merit in leaving the current crat system in place. ϢereSpielChequers 09:37, 4 May 2022 (UTC)

Copying my response from BN: I suppose my biggest concern is that CUOS's are not managed by the community. CUOS's are also not accountable to the community, they serve at the pleasure of arbcom (barring a situation where WMF steps in to remove one). The CUOS selections are about the least transparent process we have, using a secret review process. CUOS's can be arbitrarily added or removed for any reason anytime the smallest quorom of arbs can be bothered to show up and vote on a motion by a 50%+1 margin. I'm far more in favor of the community taking back the CUOS management process from arbcom, leaving arbcom to be more focused on dispute resolution and less on investigations. (UTC) — xaosflux ^Talk 10:06, 4 May 2022 (UTC)

Given the nature of the work they do, I don't see how they could be directly answerable to the community. Case in point, how could I make a meaningful asssessment of the activities of either C/U or O/S users without having the tools to see what they are doing? Better to have an elected group of trusted editors who can have the tools needed to watch them. ϢereSpielChequers 20:58, 21 May 2022 (UTC)

Besides them checking on eachother, there is also the Ombuds commission charged with investigating abuses of CU/OS. — xaosflux ^Talk 23:22, 21 May 2022 (UTC)

Yes, and that's appropriate for those particular tools. But it is very different to the openness that crats operate under. ϢereSpielChequers 11:18, 22 May 2022 (UTC)