Wikipedia:Arbitration Committee Elections December 2021/Candidates/Opabinia regalis/Questions

Individual questions

Add your questions below the line using the following markup:

#{{ACE Question
|Q=Your question
|A=}}

 Note: Per WP:ACERFC2020, there is a limit of two questions per editor for each candidate. You may also ask a reasonable number of follow-up questions relevant to questions you have already asked.

---

Questions from Mikehawk10

2. Standards of Evidence: What standards of proof should the arbitration committee use when deciding to enact a sanction on an individual, and should these standards vary depending on the type of sanction being considered?

   I'm not a lawyer, and I don't even play one on TV. But my layperson's understanding of "burden of proof" - at least as it's used in the US - is that it's deeply entwined with the adversarial nature of the legal system. There's well-defined sides making specific claims, and here's a way of assigning responsibilities for those claims. Although some Wikipedia arbitration cases do have a sort of de-facto "defendant", many don't. And as much as the process may fall short of the ideal, when it's at its best it's a sort of collaborative fact-finding exercise, and takes some inspiration from inquisitorial approaches.

   This reflects the biases of my own background, but what I've found more useful in looking at wiki-related evidence is the concept of falsifiability and what that looks like in practice. We may have some evidence in favor of a claim, but have we made a specific effort to look for and consider evidence against it? What's the best argument against a particular conclusion? What's the best alternative explanation for what you see, and can you find evidence against that alternative? I think deliberately stress-testing your beliefs in this way is a good habit. It has a similar effect to the "burden of proof" line of thought, suggesting that stronger evidence should be expected for stronger sanctions. But I think it's more likely to produce good results than having a group of mostly laypeople, some of whom may not even be in a country that uses a system like this, trying to apply an amateur understanding of a legal concept outside of its domain. Opabinia regalis (talk) 08:33, 16 November 2021 (UTC)

4. Standards for Evidence: When the arbitration committee is presented with off-wiki evidence (such as discord logs, screenshots of emails, or text messages), what is the mechanism that you believe the arbitration committee should implement to verify that the evidence presented is truthful, and how does your vision compare and contrast with current ArbCom processes in place?

   I don't think there's any single mechanism that would work here in all cases. I can't comment on arbcom's current practices, but when I was on the committee the joe job problem definitely came up a lot. Sometimes it's easy: if something happened on IRC, for example, many others (including arbs) might be able to verify from their own logs. In some cases it's appropriate to just ask the person allegedly responsible if it's authentic, if they really sent that email or if they can supply their own copies of logs. Sometimes there's strong technical evidence if you know where to look. My experience was that people sending fake evidence were usually not that sophisticated about it. A harder problem was authentic evidence of uncertain authorship - is that twitter/reddit/facebook/etc account really run by so-and-so? On Wikipedia we have extremely limited investigative powers outside of the WMF ecosystem, so this often has to rely on circumstantial evidence. This ends up being more or less a risk calculation: what's the risk of getting it wrong compared to the risk of failing to act, and who bears the brunt of that risk? Opabinia regalis (talk) 08:33, 16 November 2021 (UTC)

Thank you for your time. I look forward to your responses. — Mikehawk10 (talk) 05:54, 16 November 2021 (UTC)

Question from Ritchie333

2. In your statement, you mention taking a sabbatical. I think all Wikipedians should take breaks from editing every now and again to recharge their batteries and see what else is going on the world, so they don't get burned out. How does that reconcile with being on the Arbitration Committee, where the workload is sufficient (according to Barkeep49's notes) to require regular participation?

   Ha, I wish I took a proper sabbatical! I think if we get a second pandemic distracting people from Wikipedia, we'll probably all be trying to avoid being eaten by zombies or something and won't be worried about a website.

   I know you know this, but for the benefit of the tape: I think I'm one of the project's champion break-takers. I was inactive from mid 2007 to 2015, long enough that I had to re-run RfA. And then I was also inactive through 2020, though that was the same reason everyone's 2020 plans got screwed up. Breaks are good, especially long enough to get a little mental distance from whatever the drama of the day was when you left. Obviously you can't take an eight-year break on arbcom, but a two-year term is still a long time in a job with a pretty high historical burnout risk, and people should be taking breaks of days to weeks. You wouldn't want to do your real job for two years straight with no vacations or weekends or holidays either. That's why the committee size builds in a little slack, and one reason why dropping down to 13 didn't work out.

   I'll be honest, I'm not claiming that if I were elected I'd be the arb on the top of the activity list. But as far as I can tell the workload hasn't changed a huge amount since I was on the committee and I can do that level. I would probably take more official breaks - I very rarely listed myself as "inactive" before, because I figured that doing a little bit of work or doing it late was better than nothing. In retrospect I think that was the wrong strategy and was actually less efficient, because it stops you from really taking a mental rest and makes the activity bookkeeping harder. Opabinia regalis (talk) 08:01, 17 November 2021 (UTC)

Question from TheresNoTime

2. In your opinion, what is the Arbitration Committee's core purpose?

   I assume this isn't a "copy and paste from WP:ARBPOL" question ;) So to go with the pithy version, arbcom is there to soak up some of the suckitude so everybody else can get on with their core purpose of writing an encyclopedia. That's the point of having a last-gasp binding decision-maker to handle conduct disputes. I don't know that I'd call the other responsibilities outside of public cases "non-core", but it's undeniably a path-dependent historical result that they're all handled by the same group. Appeals, functionary appointments, auditing, handling private material, and so on are all important to the successful functioning of the project too. Opabinia regalis (talk) 08:01, 17 November 2021 (UTC)

Questions from Sdrqaz

2. When accepting cases regarding administrative conduct, an oft-used qualifier is that opening a case does not mean sanctions are inevitable. However, historically, that has not been the case. What are your thoughts on the Committee's approach to desysop cases?

   That's interesting, I hadn't seen that summary before! To get started I voted against at least two of the desysops on that list, so I may have been less... desysoppy? ... than the average arb. Of course sanctions aren't inevitable, but a case wouldn't be accepted if they weren't at least plausible. I also think there's a pretty high barrier to filing a case - it's not something people do for fun or because they don't have anything better to do for the next six weeks. That's even more true when it's directly against a colleague, as in many cases fundamentally about admin conduct as opposed to more general disputes that happen to involve admins.

   I'm surprised to see that the "review" remedy from the GiantSnowman case hasn't been tried more often since then, since (AFAIK) it seems to have worked out fairly well. In that instance the parties were actively engaged with the case, which certainly helped build confidence that the remedy was worth trying. In several of the more recent cases the primary party didn't participate, or did so only minimally, which I think is a trend that bears further examination. A "dispute resolution" mechanism that is so exhausting or demoralizing that people just refuse to take part in it isn't very healthy or satisfying. It also doesn't really give the community confidence in the effectiveness of our admin-accountability processes, and may even deter people from filing legitimate cases.

   Another thing I've seen mentioned - unfortunately, I forget where - is that a number of the more recent desysop remedies have involved admins who aren't American. Since obviously there's some baked-in systemic bias in Wikipedia's culture, it's worth giving some consideration to the origin of that pattern. The numbers are small enough that it could well be a fluke, but it's not hard to imagine that we're more likely to interpret someone's behavior as problematic if they're not from the locally dominant cultural background.

   As for what to do about that, I'm not sure. To mitigate the effect of spending a long time under a community microscope, an idea that comes to mind is a mini-case format for cases primarily about a single individual or self-contained event: seven days total, 500 words per person for evidence plus desired outcome, brief narrative decision written like the closing statement of a big RfC rather than voted on piecemeal. But I'm not sure if that would feel like getting it over with or like getting railroaded. Everyone says cases are too verbose and drag on too long, until they're in one and then they need extensions. Something I suggested in my last term but never followed up on is dropping the "proposed proposed decision" style of workshopping in favor of gathering responses to simple goal-directed questions (eg "What outcome to you hope to get from this case? What will you do differently after the case? What do you want others to do differently?"). More conservatively, keep a tight leash on cases, especially workshops, and make sure people who aren't really involved aren't taking up all the conversational space. Opabinia regalis (talk) 08:01, 17 November 2021 (UTC)

4. Of the decisions taken this year by the Committee, which one did you disagree with the most? Please note that may include choices not to take actions and simple inaction where you felt the Committee should have done so.

   I specifically avoided going down the drama rabbit hole this year, so I'm going to have a think on this one first. Stay tuned! Opabinia regalis (talk) 08:01, 17 November 2021 (UTC)

   OK, I looked into it and, dare I say it, it seems like this has been a relatively low-drama year. I'm not going to nitpick the details of the RexxS case, which seems to have been the most controversial case decision this year, but there's one aspect of it that I think was a significant error: removing the filer as a party to the case. That might be OK if it's a purely procedural filing, or if there's private information about a really unusual situation, but this looks like a regular old case about a regular old dispute. I have nothing against the filer, who I'm not very familiar with but who I've seen around saying sensible things - but if I'd been on the other side of that case and my opponent in a dispute was excused before any evidence was in, I'd have a really hard time trusting the rest of the case to be fair. I imagine this is one of those good-intentions problems - some arbs wanted to make sure the evidence covered admin conduct and not the filer's specific issue, and some thought that being a case party kind of sucks and they should let people off the hook when they can, and a few emails later nobody had thought about what that looks like from the other side. Opabinia regalis (talk) 07:17, 18 November 2021 (UTC)

Questions from George Ho

2. The WMF approved its Universal Code of Conduct (UCoC) policy. What is your feedback on the UCoC?

   I haven't been involved in discussions on this, and I know it's annoying for people to backseat-drive with criticism of something they didn't offer to help with. I do think the idea is good - codes of conduct are very much in fashion these days, and it's increasingly an expectation that projects and organizations will have one. If Wikipedia had been founded five years ago instead of 20, we'd already have one and Wikipedia:Code of conduct wouldn't weirdly redirect to Wikipedia:Etiquette. If I were writing a code of conduct from scratch, I would've started with the Contributor Covenant already widely used in open-source software and adapted it for our context.

   What I like about the UCoC as written:

   1) Starts by saying what you should do, rather than a long list of don'ts. It's too stuffy a document for WP:BEANS but it's the same principle. From what I understand this is considered best practice in writing codes of conduct.

   2) Mentions content manipulation as a form of misconduct. Good! Often misunderstood or swept under the rug as a "content dispute".

   3) Mentions sexual harassment explicitly. A lot of male-dominated projects do not treat this as a priority or assume that a general statement about harassment or abuse is sufficient.

   What I don't like:

   1) Seems unclear on whether it's meant to be a baseline set of minimal expectations, or an aspirational document laying out best practices to aim for. Is this supposed to be telling me how to skate by with a C- or how to get an A+ Good Citizenship Award? For example, does "mentorship and coaching" really belong as a universal expectation? Clearly not.

   2) Doesn't seem to have been written with enforcement in mind. I understand why they'd split the writing of the document from the enforcement mechanism, but I think the effect has been an accumulation of stuff that isn't practically enforceable. Even our own WP:AGF isn't labeled a "policy", because it's impossible to enforce what assumptions people are making, but there it is in the UCoC. Listing things like "repeated mockery" as if that isn't just going to be, well, repeatedly mocked. The entire "psychological manipulation" section, which identifies a genuinely harmful behavior and yet will almost exclusively be cited by trolls, ironically probably making it harder for actual victims to get their concerns taken seriously.

   3) The Wikimedia movement does not endorse "race" and "ethnicity" as meaningful distinctions among people - reads like someone saw the Stephen Colbert "colorblind" bit and thought it was serious.

   Also, I think the UCoC document is too long, but I decided not to call the kettle black :) Opabinia regalis (talk) 17:47, 17 November 2021 (UTC)

4. You have served as part of the ArbCom for a long while. Which ArbCom cases have affected you the most personally as a Wikipedian, even when you agree or disagree with the decisions made, and why?

   As far as my non-arb wikilife, I've been lucky enough to mostly avoid cases - my editing interests tend to be obscure areas that don't attract too much dispute. The only case I've participated in (though not as a party) was GMOs in 2015.

   I guess probably everyone who was around for the Fram case and various related events would point to that, which was a virtually unprecedented level of community upset, and a major challenge as an arb to keep up with events as they unfolded and try to navigate conflicting priorities to come to a resolution. Unfortunately this all occurred when I was in the middle of a big real-life project so I had less time to devote to it than I would've liked, but I think the arbs who took a more active public role in those events did a good job. Another one that stuck with me, in that it changed some of my thinking around sanctions structure, was the first Magioladitis case, which I drafted, and which was so ineffective we had another case a few months later. I started out thinking that finely crafted sanctions narrowly circumscribing the area of conflict would be fairer than broad ones. This case, and other early events in my first term, convinced me that we'll have more success with broader and more standardized sanctions that are less likely to create conflicts about boundary-testing. Opabinia regalis (talk) 17:47, 17 November 2021 (UTC)

Questions from Kudpung

2. If one were to compare the structures of Arbcom cases and ANI threads (apples and oranges to some), what would be your opinion on the piling on and participation of users who are clearly not involved and their eventual influence on the deliberations of the Committee, i.e. should the Committee be examining directly uninvolved participants' comments for veracity, relevance and substance and taking those comments into consideration?

   I think arbcom and ANI have this much in common: the most consistently useful comments come from people who are familiar with the subject and the dispute, but aren't directly involved in it. The least useful comments come from people who aren't familiar with the subject but need to stick their oar in anyway. Unfortunately the former are rare and the latter are a dime a dozen.

   I think this is a bigger problem at ANI where pile-ons can have a direct effect on the conclusion reached. I'd say ANI is long overdue for some structural rethinking, and is probably a better target for the reform-minded than arbcom due to its higher volume. Arbcom gets a lot of attention but ANI has a much bigger effect on most people's routine wiki work.

   Arbcom has less risk of being influenced by pile-ons and semi-relevant commentary because the panel of people making the decisions is predetermined, so people self-selecting in or out of making comments doesn't directly impact the result. My personal practice was to read pretty much everything, and quietly ignore whatever wasn't useful. I realized over time that this can cause some consternation - people don't have real-time feedback on what the arbs aren't taking into account, so they'd see something crazy posted and feel the need to respond, refute, rebut, etc. when a lot of times this wasn't necessary. I've come around to wanting more active moderation of commentary in cases to mitigate this effect. Opabinia regalis (talk) 17:47, 17 November 2021 (UTC)

4. If an accused declines to actively participate in the Arbcom case against them (Cf. RexxS) and/or retires from the project (or if an admin voluntary cedes their tools) during it, in your opinion would this be a clear admission of guilt and one that permits the Committee to pronounce by default almost the most severe sanction(s) available within its powers?

   Of course not - I think there's an element of Betteridge's law of headlines here :) I think it's more likely a combination of hurt feelings, burnout, a sense of futility, and a decision that the whole mess isn't worth the effort and unpleasantness involved. Obviously this is a volunteer project and people can stop volunteering whenever they want - see an earlier question about my enthusiasm for taking breaks. But I think it's really concerning that our top-level dispute resolution process is apparently so unpleasant (or time-consuming, or perceived to be unfair) that it's common to just decide not to participate in it. That's something I think we should really prioritize as a community in the next year.

   As for sanctions, well, a case is supposed to be about settling disputes, not determining guilt. Arbcom can only make decisions about how to settle the dispute based on the evidence it has, so it's maybe inevitable that not hearing from one of the major participants might end up biasing the result in favor of those who did submit evidence. It's always in someone's favor to give their view of events, and they're the only ones who can really speak to their own motivations and state of mind. But the fact that someone didn't participate shouldn't "count against them", so to speak. Opabinia regalis (talk) 17:47, 17 November 2021 (UTC)

Thank you for your answers, Opabinia regalis. Kudpung กุดผึ้ง (talk) 23:57, 17 November 2021 (UTC)

Betteridge's law of headlines is a good analogy. I have just one follow up question based on my 2 Qs and your excellent answer to Mikehawk10's Q1. This is absolutely not a request to reexamine any former Arbcom cases, or even to ask you how you would have voted on them - I never ask candidates how they would have voted on other cases:

2. It's sometimes been suggested that Arbcom simply totals up the consensus of the plaintiff and non involved commenters, rubber stamps it, and applies the punishment that the participants demanded, ignoring any 3rd party comments in the accused's defence. Indeed, as part of a vast, long, and otherwise exemplary voluntary service, the issue(s) under prosecution might not represent a consistent and damaging pattern of abuse of user privileges after all. Some past and present members of the Committee are however quite adamant that it is not their role to be investigative and verify the veracity of the claims or the true motivations of those clambouring for harsh justice. This could (and does appear to) have very negative consequences for an accused party who after his/her initial statement has not commented further, (Cf. e.g. RexxS) and misguidedly placed his/her confidence in the Committee to make an equitable judgement.

What's your stance on this hypothetical investigatory role of the Committee?

1. Well, I think you touched on two different kinds of investigations here. One is investigating the truth of claims presented as evidence in cases, which I think everyone agrees is part of the committee's role. The other is seeking out additional evidence not presented in a case. I think the latter instance of arbs developing their own evidence has been more controversial and I know there's been disputes over this in the past, where parties felt they didn't get a chance to properly respond to evidence that arbs' individual research efforts turned up. Obviously there's no sharp dividing line - for example, I think everyone on the committee quickly figures out that you can't take the description of an isolated diff at face value and should read the whole thread it was part of, and the links within the thread, even if none of it was explicitly presented as case evidence. And I think it's good for arbs to use the questions section to ask about things you happened to come across in the course of reviewing the evidence, or pieces of evidence you don't understand. But if an arb is doing a lot of their own research then it's probably more appropriate to recuse and present evidence as a participant. Opabinia regalis (talk) 07:17, 18 November 2021 (UTC)

Thank you again for your answer Opabinia regalis. Ironically, as far as I can make out, the general consensus among past and present Arbs is that they do not consider examination to be part of their role, even if some comments are ostensibly mendacious or purely vindictive. So thank you again. Kudpung กุดผึ้ง (talk) 09:42, 18 November 2021 (UTC)

Questions from Cwmhiraeth

2. As a former member of ArbCom, did you ever take part in a case in which you probably should have recused yourself?

   I don't think so. Because I participated in the GMO case, I've always recused from requests related to it, even though I don't really edit in that area these days. If I'd been around for the Medicine case I would've recused and will from any follow-ups. I don't recall any other examples of cases running up against my interests and activities. Mostly the opposite, topics and disputes I had no idea even existed. Opabinia regalis (talk) 07:17, 18 November 2021 (UTC)

Question from wbm1058

2. Should ArbCom-appointed checkusers be allowed to unilaterally tag a user as a likely sockpuppet of another user, point to the (perhaps years-old) sockpuppet investigation of the alleged sockpuppeteer, and then block the accused new user based on only behavioral evidence (i.e. with non-confirming technical evidence), without first conducting a community examination of the alleged behavioral connection and gaining a community consensus to block? Indeed, per this example, "Undid revision 1055911694 by xxxxx leave this for User:yyyy" the impression given by that edit summary is that the decision is only to be made by the accusing checkuser admin – who has yet to open a formal discussion on the matter. Are you OK with that?

   CU isn't an exact science. Ideally we'll get the paperwork cleaned up if there's a mistake, especially if there are identifiable people involved, but being mistaken for a different banned user socking is one of the occupational hazards of being a banned user socking. It's polite to wait for the original CU, but certainly not mandatory if new information comes up. I'm not a CU so can't look myself.

   But, isn't this Wikipediocracy drama? No thanks, I don't want to import that stuff here - isn't the flow of drama supposed to be the other way around? Opabinia regalis (talk) 08:57, 19 November 2021 (UTC)

Questions from A7V2

2. As a hypothetical, suppose two otherwise very positive content editors are simply unable to get along (with the associated disruptions to discussions, etc), and arbcom ultimately imposes a mutual IBAN on them. One of the editors subsequently finds the restriction too restrictive and stops editing. Was the sanction a net positive for Wikipedia?

   "Net positive" is an inherently empirical judgment that's impossible to make from an isolated hypothetical. If someone is persistently unable or unwilling to participate without behaving in ways the community finds disruptive, then yes, it's a benefit to the project if they find something else to do. Our most important resource is volunteer time and if it takes too much of it to deal with one person's behavior, they need to move on. Sometimes, if an otherwise productive editor is unwilling to work under a specific restriction, it's a sign that the restriction itself is problematic. If someone is really having a problem following a sanction it's worth a second look at the issue. But if you've done your due diligence and are convinced the sanction was needed, and the person still isn't willing to stick to those conditions, sounds like it's time for them to take a break.

   Interaction bans in particular are interesting because we do this "socially", but most sites have mute/ignore lists for this kind of interpersonal issue. Here you can mute someone's notifications and take their talk page off your watchlist (and should if you're ibanned, or even heading that way) but there's no technical enforcement of an iban as a sanction. Maybe a user script that warns you if someone you're avoiding has recently edited a page or posted in a thread would be useful. Opabinia regalis (talk) 22:21, 21 November 2021 (UTC)

4. In general, should editor retention be a factor when deciding on sanctions? Should the type of editor (in a broad sense) and type of contributions they make matter in this regard?

   Editor retention: yes and no. On the one hand, nobody's irreplaceable. (Good thing too, I'd charge a lot more than $0 per hour to be the single individual responsible for holding a top-ten website together!) On the other hand, committed long-term volunteers don't grow on trees; we know we have a hard time recruiting editors and getting them through the early learning curve. Failing to take retention into account seems not only dismissive of someone's contributions, but also kind of wasteful. On the third hand (what, you don't have three?) focusing too narrowly on retention of one editor risks overlooking the effects of their behavior on third parties, which are important to keep in mind but often the least visible in dispute-resolution processes. But in general, if there's two equally reasonable options for restrictions and the editor is more comfortable with one than the other, why not give it a try? Interaction bans and topic bans sometimes trade off in this way, for example. Or topic bans (socially enforced) compared to partial blocks (technically enforced).

   I'm not quite sure what you mean by "type of editor", but my instinct is to say no, this isn't relevant. If that means, say, someone who focuses on writing new articles compared to someone who focuses on technical work, that might play into what sanctions make most sense - maybe fine-grained topic bans are hard to enforce in technical areas not many admins understand well, for example. But it wouldn't affect my thinking about retention. Opabinia regalis (talk) 22:21, 21 November 2021 (UTC)

Thanks for your answers. For Q2 I was deliberately vague about "types of editors" though the examples you give are the kind of "types" I had in mind. A7V2 (talk) 01:02, 22 November 2021 (UTC)

Questions from Kolya Butternut

2. An SPI on Daner's Creek was opened this year in connection to an arb announcement, Statement regarding Flyer22 Frozen. An investigation into one of the potential socks was declined as stale. Do you think it's important to investigate old sockpuppets so that others may recognize the patterns of new sockpuppets of actively socking users?

   The CU data was stale at the time and hasn't magically gotten any fresher since then, so this was correctly declined and I don't know what further investigation you're expecting. And while I didn't follow that case or its aftermath closely at the time, even I know you've been asked not to keep revisiting this issue. Opabinia regalis (talk) 22:21, 21 November 2021 (UTC)

   (responding to below, but in the box): This is a strange choice for you to cite as an example for the general case, considering the background. In any event, I don't think it's possible to have a general answer other than "it depends". In the case of 30 edits from a year ago, I think it is time to move on. Opabinia regalis (talk) 06:17, 22 November 2021 (UTC)

Hi Opabinia regalis, I think there maybe some misunderstanding here. My question was meant to ask whether it is helpful to investigate behavioral evidence about old socks which may be potentially connected to an active master. My question isn't meant to be limited to the example case, but any active masters with old sock evidence. (I was actually told that if I think there are accounts with similarities to Daner's creek that I can post them at SPI.) Thank you. Kolya Butternut (talk) 23:10, 21 November 2021 (UTC)

Question from Littleolive oil

Removed please see talk page.

Questions from Moneytrees

2. The Antisemitism in Poland proposed decision. Do you have any thoughts about the numerous ARCAs filed after the case, or what could've been done differently?

   As you know, I co-drafted that case, and I'll be the first to admit it is not exactly a contender for top marks in drafting quality. This case overlapped with the very time- and resource-intensive Fram situation, and because of inactivity and resignations at the same time, there weren't many arbs available to work on the case. It would have been a challenge even under better conditions, because the dispute is in a content area none of us were experts in. A lot of the content involved non-English sources so claims about POV-pushing and source misrepresentation were very difficult to evaluate.

   This is also a case where I didn't have to eat my own dog food, because only two of those ARCAs were started during my term, and only the first one finished. So I'm reading them now for the first time. My first conclusion is that I hate the way we archive ARCAs and they make the talk page a huge mess, but never mind. Looks like there's three clusters of requests: VM/Icewhiz, the EC restriction, and the sourcing restriction. The editor-specific modifications seem like expected consequences given the events after the case: Icewhiz of course was banned, so there was no need for a two-way interaction ban, and subsequently VM successfully appealed his topic ban. The need for EC seems to have come up due to Icewhiz's socking habits after his ban, which I don't think we could reasonably have anticipated at the time of the original decision. The inconsistency between "article" and "content" raised in the August-September 2021 request was probably avoidable, and I think the resulting standardization in the "omnibus" motion was a good idea.

   The biggest challenge seems to be the sourcing restriction. IIRC this originated as a one-off discretionary sanction that then was adopted as a case remedy - much like the original 30/500 that evolved into EC. To some extent you expect some growing pains when doing this kind of "remedy innovation" (again, much like 30/500 took some time to get right, and probably some people still don't think it's there). I like the suggestion raised in some of the discussions there about a community guideline in the style of MEDRS for controversial historical content, and if that existed then it would obviate the need for this kind of restriction in an individual topic area. In any case it looks like in some ways the restriction outlived its usefulness and was rolled back a bit, and may well become obsolete if poor sourcing stops being a common behavior in the area. I think that if the case had been structured differently from the outset, with more of the common participants in the area listed as parties, it might have had a more traditional outcome with individual restrictions being proposed rather than a general topic-based one, though I'm not familiar enough with events since then to say if it would have worked out better. Opabinia regalis (talk) 06:17, 22 November 2021 (UTC)

Question from Epiphyllumlover

2. There is an active and routine off-wiki freelancing market where Wikipedia editors from non-English speaking countries sell RfC votes and talk page comments to paying editors on enwiki. I have watched this corrupt discussions on enwiki, but would feel guilty reporting it, since I know that the editors actually making the comments really need the money. In addition, I feel that editors from the non-English speaking countries have the potential to contribute more to Wikipedia-- but fall into selling votes because it is both lucrative and requires little understanding of wiki code. It would be a shame to drive them away, given the great potential which would be lost. Would you support a WMF-funded bounty program modeled after the Nordic model approach to prostitution, where the editors from non-English speaking countries could receive a financial bounty for turning in their employers to ArbCom for discipline, while at the same time also be offered access to an exclusive Wiki syntax training program so they can build skills to pursue greater things?

   I think this is pretty far afield of what arbcom actually does. Contrary to popular belief, arbcom is not the disciplinary committee, and I suspect their annual budget of $0 is not going to make an effective bounty system. If the WMF wants to do that, I suppose they could, though I don't think it's likely to be a success or find community support.

   As for the part that could be within arbcom's remit, I think that should start with providing some evidence of these alleged markets privately. That doesn't have to mean turning in specific people, but a claim like this is pretty extreme and needs substantiation before we invest effort in evaluating potential solutions. Opabinia regalis (talk) 07:31, 23 November 2021 (UTC)

Question from Hijiri88

2. What is your stance on two-way IBANs that are imposed because of one-way harassment? In the past, ArbCom has rejected one-way IBAN proposals in favour of two-way IBANs on purely technical grounds (that they are subject to being gamed, that they "don't work", etc.), but if such a one-way IBAN is implemented and it doesn't stem disruption (for example, if the hounded party is simultaneously placed under a TBAN that is not placed on the hounding party), and instead only leaves the harassed party subject to repeated remarks of "User X is subject to an IBAN with User Y -- he wouldn't be banned if he hadn't done something wrong" and unable to explain the context, would you be open to repealing it? (I am assuming that BANEX applies to these questions.)

   That really depends on the situation. I've noticed it's sometimes the case that both people in a two-way iban believe that they are blameless victims being unfairly sanctioned for their opponent's obvious misdeeds. And at the other end of the spectrum, sometimes an iban of any kind is just a stopgap and what we really should be doing is blocking the harasser. If it's truly asymmetrical, yet also not severe enough to just block the problem person, then I generally agree that the person being harassed shouldn't be sanctioned just for the sake of others' convenience. If the situation warrants, a middle ground might be changing the presentation - voluntarily agreeing to not interact with the sanctioned party - or adding a note in the iban itself explaining the reason, to avoid the problem you mention. Opabinia regalis (talk) 07:31, 23 November 2021 (UTC)

Question from Nosebagbear

2. I frequently handle regular indef appeals, of which a significant fraction is non-technical "Duck" sock blocks. In many of these cases, the blocking admin hasn't and won't share what is often a non-direct set of behavioural evidence to the accused/blocked. When I review it, most are clear-socks, but a few are judgement calls. In either case, impossible standards are being demanded in the appeal - prove a negative, without knowing the exact (complex) material which they need to rebut. This leads to several related queries: 1) is this lack of public sharing of info (to avoid aiding future socking) policy-backed? 2) is it morally justifiable? 3) How are non-expert users supposed to make a viable appeal under these circumstances? 4) Does this method not encourage outright lying, as it pushes people towards having to concede the socking, even if they didn't?

   I think that's four questions formatted as one :p

   To be honest, if I signed up for a website and got caught up in some kind of error that I couldn't convince the management to resolve, and I still wanted to use the site, I'd just wait a few days and start a new account and move on with my life. Which I'm sure is what lots of people do in this situation. It's a crummy first experience for new users, and I'm sure we lose some would-be new editors who don't take that step, but it's hardly some travesty of justice or stain on our moral fiber or whatever. It's just a website. (Same applies to everyone out there reading this and getting ready to be shocked! shocked I say! that an arbcom candidate could endorse sockpuppetry like this!) I admit I don't work on unblock requests much, so I may be wrong, but I don't think falsely confessing to sockpuppetry is something genuine new users are very likely to do. I think they're much more likely to either give up or reregister.

   As for what to do about these blocks, as a practical matter we don't want to give it away if we've figured out that some recurring sockmaster has a behavioral "tell". SPI is not Sockpuppet School. But if others reviewing a WP:DUCK block are making a genuine effort to see the connection and still only seeing a rabbit, then maybe it's not quite duck-ish enough for a block. A good next step there is to contact the blocking admin by email to have a more frank conversation than would be possible in public where the sockmaster could be reading. Blocks are supposed to be peer-reviewable by other admins, so making hard-to-explain "duck" blocks and not having/using email would IMO be questionable practice. Of course, if a particular case really does need to be tracked closely or for some reason relies on actually private information, the blocking admin should send their evidence to arbcom (or CUs/functionaries, as applicable), where there are institutional-memory resources that admins as a group don't have.

   I don't think appeals of duck blocks need to be more complicated than any other appeal. The user should just explain what they were doing - how they came to edit that article/choose that username/whatever. It's a judgment call whether the explanation sounds plausible. In a lot of cases, if they get unblocked but shouldn't have, they'll get themselves in trouble again in short order. To avoid this kind of issue I think many cases of low-level sock blocks would be better off as blocks for straightforward disruption. Opabinia regalis (talk) 05:02, 25 November 2021 (UTC)

Question from Banedon

2. You wrote above that "... I think it's really concerning that our top-level dispute resolution process is apparently so unpleasant (or time-consuming, or perceived to be unfair) that it's common to just decide not to participate in it. That's something I think we should really prioritize as a community in the next year." I think this issue is really concerning too. Given free rein, how would you fix it?

   Good question! There's only so far "free rein" for one person can go, I think, because a lot of this is about how people who aren't me would feel about things. As you know, deciding to run for arbcom is a slightly crazy decision that makes you unrepresentative of the general editing population ;) See for example the suggestion in this answer above for an accelerated/mini-case format for issues of limited scope. Cases are often criticized for dragging on too long, but would something like that feel to the editors involved like "well, at least we got it over with" or "no fair, they didn't even give me a full case!" I think the decision earlier this year to move the analysis section and make workshops optional is a step in the right direction.

   In terms of a stressful environment, I see a lot of resonance between arbcom cases and RfAs - both have had declining numbers for a long time, resulting in disproportionate amounts of attention paid to each one. The highest-drama examples were probably always stressful, but even the "routine" ones attract a lot of commentary and have an outsized perceived effect, resulting in the participants feeling like they're under a microscope being criticized by a lot of people for a long time. I think we could fruitfully poach some ideas from this year's series of RfA RfCs (WP:OMGWTFBBQ!) that might translate to the arbcom context. If nothing else, we could poach the format and run a similar (but hopefully shorter) RfC on arbcom cases. I have my own ideas about how I think the case pages should be restructured, for example, but I'd want to put those out for community review before pushing them too hard, because it really might be just me who thinks it would be a better environment with better records if case requests had their own subpages, and ARCAs were subpages of the case they belong to.

   Speaking of ARCA, I think that's another area that needs attention. Even as case numbers decline, a growing fraction of the project is subject to arbcom decisions from past cases. Most editors' "exposure" to arbcom now comes from DS, not disputes that make it to a case request, and AE/DS is a notorious thicket of confusing and inconsistent rules. I'd love to see the DS review started last year come to a conclusion.

   Probably the most salient changes I'd like to institute if it were all up to me are at ANI, even more than at arbcom. Some of the worst conversations on Wikipedia, in any forum, are the high-volume, rapid-fire back-and-forth comments that are impossible for the subjects of the thread to keep up with. We keep adding more and more rules to try and organize the chaos, but the more we write the less people read. I think we have a road with a lot of dangerous drivers, and we've tried the equivalent of bigger speed-limit signs and more speeding tickets, and it's time to try installing some speed bumps.

   This will get criticized for being a technical solution to a social problem, but I think building in some nudges toward good behavior into the environment in a generalized way so it doesn't feel like personal criticism would help a lot. When you write a post at ANI (or some areas of arbspace too), a little word counter appears in the corner and turns orange at 400 words in the rendered text and red at 500. You can still post longer, but there's a little hint to keep it short. (I know, irony alert!) If you post to ANI more than 10 times in 15 minutes, you're partially blocked for an hour. If you do it more than 15 times in 24 hours, you get blocked for a day. If you post more than 5 comments in one thread or post in more than 5 threads, you have to click through a little pop-up box saying "dispute resolution works best with diverse voices participating, and you've already posted N times". Generally use the "built environment" to give people cues on what they're supposed to do and how they're supposed to behave, and reduce the amount of friction and baseline irritation that even long-term editors experience in dispute-resolution venues.

   Nah, never mind, I only wrote all that so you wouldn't read to the bottom where I admit to my nefarious plot to become Dictator of Wikipedia ;) Opabinia regalis (talk) 08:42, 25 November 2021 (UTC)

4. Many years ago when voting on this decision, you wrote "Support, but I think we need one of these for Banedon as well. I see 15 notifications". A few of the other arbitrators agreed with you, yet it seems like nothing happened. Why? If it's because you opted not to push harder to have a separate finding of fact added, why did you make that decision?

   That case was in October 2016. And on this 639th day of March 2020, I don't exactly remember :) Looks like nobody proposed another finding, probably for the reason Callanecc suggested, that whatever you were doing at the time was not overall significant enough to need it. Some arbs have the philosophical view that there's no need for findings unless they have a corresponding remedy - I don't in general, but probably wouldn't bother with extra findings for relatively minor stuff like this. Opabinia regalis (talk) 05:11, 30 November 2021 (UTC)

Question from Atsme

2. What is your position about ArbCom finally following through with DS amendments and irreversible unilateral actions in the name of AE, all of which was put on the back burner in 2021?

   Luckily it's a wiki so nothing's irreversible :) When I started paying attention to arbcom a few years back, there were a lot of disputes around AE/DS enforcement, particularly the problem of one admin deciding to implement a block or other sanction after others had already declined to act. Possibly as a result of the high level of drama - which led to at least two cases - things seemed to settle down in this area during my term. Disputed sanctions or complaints about being overruled tended to get hashed out at ARCA, where I think the slow pace and annoying formatting of the page counterintuitively helped settle things down.

   I haven't been following the AE/DS area since I left the committee so I can't say whether this problem has come back; it does seem like there's a swinging cultural pendulum on this topic, as with a lot of other enforcement-type areas. The consultation lists the perceived "first-mover advantage" as one area for examination in the next phase of DS review. It seems like the planned schedule for the review has slipped (like so many things in arbcom that aren't immediately on fire) and I'd certainly like to see it through. If I were elected I'd want to look at what the current committee has done so far before proposing any major changes. If I were to pick a top priority for DS review/improvement, I'd start with "awareness", a system long known to be a mess. For some of the other issues around first-movers or enforcement actions, I've always wondered why we don't have a "two-admin sign-off" kind of system for some high-stakes actions. Whether this is an appropriate use case, I'm not quite sure, because multiple admins already tend to weigh in in places like AE or in tense ANI threads, but what I had in mind is more of a shared responsibility than just being part of a consensus. Opabinia regalis (talk) 08:46, 27 November 2021 (UTC)

4. Do you really have time for this?

   If anybody tells you they know exactly how much time they have over the next 730 days, they're blowing smoke. Especially if they tell you that in the middle of a pandemic. But yeah, I have the time to do now what I did previously. I guess it'd take longer to fill in all these redlinks. Opabinia regalis (talk) 08:46, 27 November 2021 (UTC)

Thank you for throwing your hat in the ring and for your responses to my questions.

Question from Robert McClenon

2. Some of the most troublesome disputes in Wikipedia are protracted content disputes that are complicated by conduct issues, such as incivility, civil POV pushing, stonewalling, and filibustering. These disputes often go to WP:ANI more than once. This is a two-part question, and maybe will be considered to be one question or two. First, how should ArbCom decide when it is necessary to accept a case that is a combination content-conduct dispute? Second, there are relatively few available mechanisms for dealing with such protracted cases short of conduct adjudication (WP:ANI, Arbitration Enforcement, or an arbitration case). Third Opinion is for straightforward cases with two editors. Dispute Resolution Noticeboard is normally for relatively simple disputes that will take two or three weeks to resolve. Do you have any ideas for how to try to resolve protracted content-conduct disputes to minimize their division of the community before arbitration is sought?

   Well, I think people do have a few more options than that - on the content side of things, you can run an RfC, go to specialized noticeboards, use relevant processes like RM, even recruit more input from a wikiproject. None of that is necessarily "dispute resolution" but our normal processes can have that function. I guess I'll answer the second question first, I have been a big fan of very structured, moderated RfCs since participating in one that I thought was going to be a complete dumpster fire and then... it actually worked well and pretty much solved the problem. These are a lot of work to set up, but I think we could help with a little more social infrastructure around it. For example, we have people who sort of individually volunteer themselves as closers of big RfCs and messy ANI threads, but we don't really have any organized process for recruiting closers of RfCs after they've run their course, much less volunteers to moderate while they're ongoing. Come to think of it, as I was writing this I realized it sounded familiar and I have suggested something like this before (see that meandering thread, and also this follow-up for my withering away of the state plan for arbcom's eventual slow obsolescence... unfortunately I think the end of that path is not a community without arbcom, but a community that has a new group called something else that does the same thing).

   As for your first question, I don't know that there's anything too different to be done in accepting these cases. They have to have a reasonable amount of unsuccessful prior dispute resolution, and sufficiently affected by the conduct aspects that there's reason to think arbcom could actually help. I will admit these are my least favorite cases - they affect content, so they matter in ways that back-office arguments don't, yet they're often about very specific topics that arbs have no relevant knowledge of. And often there's a dispute on-wiki because there's a dispute in the real world, so there's no chance of really "solving" the problem, just keeping it somewhat under control. Even though I don't enjoy them, I actually think we may be a bit too conservative in requesting or accepting cases like this. An arbcom case is 6 weeks or so - if the alternative to arbcom takes longer than that with just as much drama, was it worth the effort to avoid arbcom? Encouraging more cases of this type does run the risk of turning Wikipedia:Requests for Arbitration into Wikipedia:Requests for Discretionary Sanctions and Topic Bans, but I think there's room for streamlined handling of this kind of request with an obvious real-world component, like the way the COVID-19 and Horn of Africa sanctions were implemented. Opabinia regalis (talk) 05:11, 30 November 2021 (UTC)