Talk:Requests for comment/Adding global abuse filter rights to global sysops

@Billinghurst: It looks like this didn't go anywhere in the past 2 years. I propose we close this RfC and call it a day. Effeietsanders (talk) 02:57, 23 January 2018 (UTC)

Talk to the stewards. I proposed a month back then, and only the stewards can do anything.  — billinghurst sDrewth 04:29, 23 January 2018 (UTC)

@Billinghurst: You mentioned that you wanted someone to make a recommendation based on the RfC. Sounds fair, but if you don't specify the person (and where you would want that person to publish their personal recommendation), that'll never happen. I'd say, pick your favorite neutral steward, and ask nicely :) What about User:MarcoAurelio? Please note this would always be a personal recommendation based on input, someone else will then have to take it to the next level. And a recommendation might also be that based on this input, there's no recommendation. Effeietsanders (talk) 07:45, 23 January 2018 (UTC)

Closed as unsuccessful. I concurr that the discussion went nowhere. —MarcoAurelio (talk) 12:34, 23 January 2018 (UTC)

@Effeietsanders: Sure. It was a proposal for stewards so they could get some assistance with global abuse. [Personally, I can already see it, and see value, and the result of any change has no impact upon me.] The community/stewards can do what they wish to with this. Excuse my lack of enthusiasm.

My current view is that the broader community is ambivalent as long as they have someone doing it for them, and the little and medium wikis generally don't have the skill set. It is predominantly something like black magic.

The whole aspect of global abuse filters and their management is, umm, interesting.

Global abuse filters came about to manage large amount of bot spam, and some of the xwiki vandals. The filters were tightly controlled in the initial roll-out so that they could be put in place with least resistance, and manage concern about filter writers, ie. least scare factor for a quick rollout, so just give the right to stewards. Since that time, nothing has changed. There is zero requirement for them to be steward only, and as long as the operating rule of no blocks on global filters are applied, it could be more broadly managed. Instead we have the current apparent paralysis. Meta administrators have access to spam blacklist (hits logged for admins only) and title blacklist (hits are not visibly logged anywhere) which are truly globally active, and more powerful, yet spam filters are more tightly controlled, better logged, and can do less damage.

<exasperation>Go figure. [For me personally, it is just easier for me to retroactively blacklist problematic domains, than it is to proactively design and test global abuse filters, and then try to get them turned on for testing and action.]

It would appear that stewards seem to want to hang on to the power of global abuse, though don't seem to be doing much to update them or promote global abuse. Or collectively they don't have broad enough knowledge to manage them, so it falls to a small number of stewards to do that, effectively a small monopoly based on individual preference to act or not. It is not currently about the community's needs in an actively dynamic space. I could point to the job description for Stewards which has no mention of abuse filters, let alone global abuse filters.

I have emailed stewards about it, and left them messages, to little or no feedback. If people are trusted to recommend updates to filters, which are implemented unenhanced on multiple occasions, and that is it. Meh! I am pretty much disheartened by it, so I do a little bit, and have predominantly moved on, and will do other things around the place, and people can have their spambots to clean up. It used to be about collective contributions, and now it seems that there is a level of applied authority and control by a specific collective. It isn't like the old days. If we need these things controlled by the stewards alone, then let it be so. If we do not, then allow the community to do it on a graduated and trusted scale, and my proposal covered some of that. The community gets what its discussion and its consensus allows.</exasperation>  — billinghurst sDrewth 08:37, 23 January 2018 (UTC)

The steward role has evolved over time, with every new technical feature rolled out at the global level. Originally, as you know, it was just userrights-interwiki on Meta. Once global accounts were rolled out, that became part of the package. Same thing for global renaming, global abusefilters, and OAuth approvals. You're right that it doesn't need to be this way, and for global renaming it isn't. I think the main stumbling block from devolving more functions out of the stewards group (I'm thinking global blocking/locking and the global abusefilter here in particular) is simply the effort that such a move would take. The same arguments always come up: it infringes on local project autonomy, and the new group won't be vetted as well as stewards are. In my opinion, both arguments are absolute nonsense. Global AbuseFilters, like the global spam and title blacklists, should be completely global and edit-able by Meta admins. Global accounts should be lock-able by global sysops. There should be no reason for people to get full access to all wiki functions just to have access to these things. – Ajraddatz (talk) 09:16, 23 January 2018 (UTC)