Re: Fram
Posted: Sun Jun 16, 2019 3:36 pm
Risker scaring the children......
On what basis do you come to the conclusion that this reaction (or something akin to it) was not expected or predicted? I've been working since Day One on the assumption that it was pretty much entirely predicted to shake out the way that it has. Even the timing is not surprising, occurring within a week of a long-scheduled board meeting. So far, none of this has surprised me. Maybe it's because I spend a lot more time operating at the "global" level. Risker (talk) 15:28, 15 June 2019 (UTC)
Change is, by design, disruptive. They may not have predicted this exact response to the disruption, but I am certain they anticipated that the action would cause disruption. There have been a lot of indications that changes in the management of user behaviour were coming. Keep in mind that T&S and the WMF aren't just dealing with English Wikipedia, and that our user behaviour problems, while minor compared to some other small projects, have a disproportionate impact on the perception of the global umbrella of projects. The fact that it is now out in the open that these changes have been in progress for over a year, and that various iterations of similar penalties have already been imposed on other projects, tells us that this is part of a larger plan. I do have the advantage of being personally acquainted with at least half of the people involved in T&S or in the chain of authorization for OFFICE actions, and none of them are fools; every one of the ones I know would have fully anticipated that Enwiki would go "nuts" when they made their first OFFICE local block here. The WMF - and the Board, just about every member of whom is actively involved in the work being done on the 2030 strategy development - has been moving toward a more global approach to just about everything for a long time, and here on this project we've generally been turning a blind eye to it and acting as though we're too important to mess with. We don't exist in a vacuum, much as some may want to think we do. Risker (talk) 16:11, 15 June 2019 (UTC)
I guess the question is whether or not disruption (which I'm sure they'd agree this is) is the same thing as destruction (which is a lot more questionable). In a lot of ways, our project is becoming increasingly self-destructive. How many people on this project know that the majority of sub-Saharan African Wikipedians edit either this project or French Wikipedia? How welcoming are we to them? Do we seek them out, treat them with respect, understand that they're probably better arbiters of what constitutes a reliable source about Kenya or Lesotho than those of us sitting in the Northern Hemisphere? Are we dealing effectively with edits coming out of the Indian subcontinent, another major global area contributing to our new editor cadre? We aren't talking about that stuff on this project, and I'll lay odds that most people are completely unaware of where the potential for new growth is coming from, and the support systems and mechanisms that oldies like us had when we first started editing just don't exist anymore.
As a community, we've embraced globalization throughout the WMF when we've thought it to our advantage. We were happy with the introduction of SUL, a lot of people were genuinely excited with the introduction of global preferences and global user pages, we were pretty much thrilled to bits with the introduction of the "paid editing" clauses to the TOU, and we were proud to be the pilot projects of what became the global legal fees assistance program. The roots of the Trust & Safety program are right here on English Wikipedia, and many of the activities we are seeing now on a global level were first developed to address issues on this project; when you look at the list of OFFICE banned users, more than 2/3 of them primarily edited this project.
This is wandering pretty far off-topic here, but I suppose my key point is that we're not doing a great job ourselves of resolving the low- and medium-level user behaviour problems, despite knowing for years that they've been adversely affecting new editorship, and there's good reason to believe they've affected editor retention, as well. These are really hard problems to solve - and they're problems in just about every type of online community. I don't think this was the best way to address things, but to be honest I suspect we would have wound up with almost the same discussion if the T&S team had come here and said "hey you've got some user behaviour problems that are adversely affecting your project, and we suggest doing XYZ" than if they just did XYZ. We wouldn't have wound up with the desysop/crat issues, but I'm pretty sure it would have been just as contentious otherwise. Risker (talk) 17:33, 15 June 2019 (UTC)
The Terms of Use are not opt-in. The OFFICE global ban program is not opt-in and in fact was created and expanded because of pressure by English Wikipedia. SUL was never opt-in; it was all-or-nothing. Global preferences is a preference program that affects anyone who, while logged in, goes to another Wikimedia project; it has defaults that are largely benign but it's not really opt-in. The use of the LFAP is optional for editors to whom it applies, but it applies regardless of what project(s) the editor contributes to. I don't really think any of the programs I've pointed out are really optional. Risker (talk) 17:58, 15 June 2019 (UTC)
I'll note there's a difference between "you can create an alternate to the paid editing TOU but unless your community supports that, this TOU applies" and "you can opt in to this paid editing TOU". [Complete aside: I was the person who piloted the "alternate disclosure policy" for Mediawikiwiki and tech projects.] Coming back to the core issue that seems to be at the base of the WMF/T&S action here, we have long known that there are very serious difficulties in addressing behaviour issues amongst the group of editors who've been labeled "unblockable"; this group includes most administrators, and a lot of long-term prolific editors. In the latter case, they have in fact often been blocked, but the blocks don't tend to stick. There are probably only about 20-30 administrators who could successfully block an unblockable, and even then they'd be risking their bits to do so. I am certain that just about every administrator who carries out blocks has refrained from blocking one of those individuals at least in part because they know (even though the block has been more than earned) they'll spend days defending the block, and will likely burn up a good chunk of whatever social capital and sweat equity they have in doing so. We know as a community that this is a problem, we've known it for years, and we've avoided addressing it. Arbcom isn't the answer - it's not designed to address this sort of stuff, and we really haven't given it either the authority or responsibility of doing so. There have been complaints going back almost as long as I have been on this project (almost 15 years now) that our community can't or won't deal with this issue. So I'll go a bit further out on a limb than I have to this point: I don't think that T&S is the right answer here, either; it's not the process or the course of action I would have recommended (had I been consulted), nor the one that I think will achieve the best results. On the other hand, I don't think I or anyone else in this community has come up with any other, better courses of action.
I think perhaps the issue here is that there isn't a consensus on how to interpret the core action here. Some see it as a wake-up call that our user behaviour problems are more serious than we have admitted, and we need to make "human resources" type changes in our project. Some see it as a flat-out usurpation of community independence. Some see it as a much-needed step because the project has, in fact, failed to enforce its existing policies and/or has not developed processes or systems to address problems we know we have, and attempts to resolve these problems have not only been unsuccessful but have been openly and actively blocked. My suspicion is that elements of each of these perceptions are correct; that this one action has a lot more aspects to it than simply one issue. Risker (talk) 19:41, 15 June 2019 (UTC)
I've lost count of the number of times it's been discussed; I could probably come up with a list of 50 prior discussions, small and large, if I wanted to spend all weekend at it. At most, we've come up with genuinely tangential applications (e.g., not using external websites to attack people) but really haven't hit the core "low-intensity chronic aggression" issues. And now this recent discussion has lost its way, too. So...how do we get the discussion to stay on track, to come up with actionable positions and plans that address general comportment rather than fringe cases, and then bring it into force? That's the most important challenge we face. Risker (talk) 20:25, 15 June 2019 (UTC)