Wikipedia:Arbitration Committee Elections December 2017/Candidates/Opabinia regalis

Opabinia regalis

I'm still hoping my attempts to bribe other people into running will work, but in the meantime I thought I'd put my money where my mouth is. (Or my mouth where my money is, in this case.) I was elected to a two-year term in WP:ACE2015 (statement, questions) and am throwing my hat in the ring for re-election.

Two years in, I've changed my mind about some things, but my basic qualifications remain the same, and still consist almost entirely of being a level-headed, deliberative, analytical person. Luckily I've learned during my term that being good at changing your mind is an essential skill for an arbitrator. Two-year terms are long enough to eat some of your own dog food, and hey, sometimes it doesn't taste good :) Astute observers will note that my edit count has gone down lately - I changed jobs this past summer and real work has been eating into my hobby time - which was one of my biggest hesitations in deciding to run again, but I do think that I've been reasonably effective in the job even with less time on hand, and I anticipate a less demanding schedule after the new year. I think the argument I made in my 2015 statement, that as a scientist I'd bring a thoughtful attitude about evaluating evidence, has worked out well on the whole. I'm still on the lenient side in terms of sanctions and still capable of talking to people like a human despite my lack of a spine, though I think I've slowly moved toward a firmer approach with repeated disruption.

When I ran last time, I emphasized my unusual background as a source of new perspective - I registered in 2006, had a successful RfA six months later, faded out of the project about six months after that, returned in 2015 after such a long absence that I needed a new RfA, and then ran for arbcom the same year. Now I'm just part of the project's furniture and will have to hope some new young guns get elected to serve that outsider role :) I think the committee benefits from a good mix of experienced arbs and fresh faces, and I've found it enormously helpful to serve along with some longer-memoried arbs, so I hope I can still be useful on the other side of this particular fence.

Mandatory stuff: Obviously I meet the requirements and have already signed the NDA. Alternate accounts are User:Opabinia externa (used on my phone and while traveling), User:Opabinia robotus (old account intended for a bot that I never got around to developing), and User:🐱 (new since last cycle, created to test emoji usernames after someone asked if it worked). Opabinia regalis (talk) 18:55, 18 November 2017 (UTC)

• Opabinia regalis (talk · contribs · count · logs · target logs · block log · lu · rfas · rfb · arb · rfc · lta · checkuser · socks · rights · blocks · protects · deletions · moves)

Individual questions

Add your questions below the line using the following markup:

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

---

Questions from Collect

2. Should the existence of a "case" imply that the committee should inevitably impose "sanctions"?

   Ask a recycled question, get a recycled answer ;) I doubt there's much signal value in this question; nobody says "yes, once you're in a case you're doomed". Opabinia regalis (talk) 09:02, 19 November 2017 (UTC)

4. If an administrator has openly stated a strong aversion to an editor's article edits on that editor's talk page, is that sufficient to indicate that the administrator is no longer impartial concerning that editor?

   Sufficient? No. Not without knowing what the admin said and in what context. Opabinia regalis (talk) 09:02, 19 November 2017 (UTC)

6. a. In cases where the person involved in a case is actually out of the country during that case, ought there be a delay to give that editor sufficient time to address "new evidence"?
   b. Where multiple editors present evidence against such a person, ought space and time for rebuttal be given?
   c. Where evidence is added at the last minute, should the clock be stopped to allow actual time to rebut the last-minute evidence?
   d. Under what circumstance, if any, should arbitrators be allowed to present evidence in the proposed decision which was not previously presented by anyone else?

   It doesn't really matter how you number them; you are one person asking each candidate seven questions, many of which are recycled from previous elections. I'd encourage you to reconsider whether it's a good use of everyone's time to read and write all that text.

   But, since I never say no to a perfectly good soapbox, I'll use these as a general jumping-off point for thinking about case progression. We get a lot more feedback that cases are too long than that they are too short. If a participant has some kind of schedule conflict then I'm happy to make reasonable accommodations - I'm hardly one to talk about real life getting in the way - but we also owe it to the community to get the issue sorted out at a reasonable pace. A few times we've just suspended cases with one major party who was unavailable, and that works out for certain types of cases. But at some point, if there are major and ongoing real-life concerns, that's more important than participating in a boring back-office Wikipedia procedure anyway.

   I do think it's a mistake to worry too much about who posted evidence in what order, when it was posted relative to the deadline, etc., and likewise a mistake to invest a lot of effort in "rebuttal". Cases can get adversarial sometimes, but the evidence phase is meant as fact-finding rather than as marshaling up the best arguments for your "side". The best way to deal with the evidence phase is to make sure your own submission is clear and well-organized rather than to look for opportunities to cross-examine the opposing side. It's actually pretty difficult to pull the wool over a majority of the committee's eyes all at once, and if anyone suspects that's what happened, they might at least consider the possibility that it's their own view that's a little obscured. Opabinia regalis (talk) 09:02, 19 November 2017 (UTC)

No question from Gerda Arendt

2. No question, unless you want to say if you agree with yourself ;) --Gerda Arendt (talk) 22:16, 18 November 2017 (UTC)

   I may have changed my mind on a few things over the last two years, but I still agree with myself from less than a month ago ;) Opabinia regalis (talk) 09:02, 19 November 2017 (UTC)

Question from Carrite

2. So, what did you learn about ArbCom over the last two years? If you were Queen of the Universe, what would you change about the ArbCom process?

   I learned that the mailing list is way less interesting than everyone thinks ;)

   Some people will read this and think it's obvious, and others will read this and won't believe me, but: 99.99% of the time (only because nothing is ever 100% certain), if it looks like arbcom is engaging in some kind of conspiracy, or is pursuing an agenda, or is out to get someone, or is circling the wagons, or is carrying water for the WMF, or whatever else we're supposedly off doing in secret, the real explanation is much simpler. What's "really" going on behind the scenes is a bunch of people with disparate schedules muddling their way along doing the best they can with a difficult situation. Everything I thought was a conspiracy while watching from the outside in 2015, wasn't. Everything others thought was a conspiracy while I was on the committee, wasn't. (OK, I'm talking about the last few years; I'm not bored enough to read the archives back that far so the early arbs could be lizard people for all I know.) Yelling "Star Chamber! Forced disappearances! Show trials! Fascism!" is never the right interpretation of events no matter how tempting it is to post that stuff on WT:ACN after a controversial or opaque decision. Our faults have mainly been that we're slow and indecisive, not that we can successfully pull off conspiracies.

   I say that part first because that was the most jarring part of the transition, I think - having people suddenly start reacting with suspicion that whatever I was doing had some ulterior motive or was designed to hide something. I suspect that this dynamic underlies the observation that lots of people promise "reforms" of various kinds in their election platforms and then don't do much reforming. It turns out that an abstraction like "transparency" was never really the biggest problem, and promising to increase it doesn't translate well into concrete, practical changes. (Of course I'm aware that a possible criticism of this claim is that I've been coopted by The Man.)

   Now, if I were really Queen of the Universe, I wouldn't be screwing around with arbcom's internals, because I'd be drinking cocktails on my own private beach while waiting for my spaceship to arrive so I can survey my domain. Or at least I'd give myself a couple of extra hours a day. I'm going to save the second part of this question for tomorrow; it's my bedtime :) OK, if I'm not Queen of the Universe but only Queen of Wikipedia, can I start by issuing a royal decree requiring editors to take good advice? Some of the most frustrating situations that have come up since I've been on the committee have involved good-faith editors who just. will. not. listen. I know I'm not the first and won't be the last arb who ran for the job thinking "well, I always want to say 'fuck you, not doing that' when someone tries to tell me what to do, so I'll be good at dealing with it when other people react that way". And I'm neither the first nor the last to discover it doesn't work that way and once you have some kind of perceived "power", even if you're not using or invoking it, people with that pattern of behavior react to you that way. (Now that I think about it, a Royal Decree is unlikely to fix this problem :)

   On a more practical level, I would love some bespoke software for internal arb communications. I said in 2015 that people should smack me if my term ends and we aren't using a CRM yet, and here we are still using GNU Mailman like it's 1999. Even though the overall email volume is much lower than its peak, there are still issues that get missed or fall through the cracks on the mailing list. (For the benefit of everyone reading this other than Carrite, see here - we are at about 20-25 messages per day on arbcom-l, well down from the 2009 traffic peak.) I'm as guilty of this as anyone, but it drives me crazy that "low-profile" editors whose usernames I don't recognize are more likely to get lost in the shuffle. I'd really like an issue tracker with easily searchable full message archives that is just as convenient as email, requires no technical skills to use, and will surface for each arb the most urgent current issues/threads awaiting their input. Past committees have explored better software and failed; it sounds like a straightforward request but is surprisingly difficult. (We actually do have an alternative to try, but that's been in suspended animation for a while, which is also partly my fault for being slow and distractible, which is precisely the problem I'd like to see resolved through software....). Opabinia regalis (talk) 18:01, 19 November 2017 (UTC)

Questions from Nick

2. What do you believe are the biggest errors of judgement that the Arbitration Committee have displayed previously as a collective body. In your answer, please give examples from both your time on the committee, and from before your time serving on the committee. Discuss how you would have approach the issues with the benefit of hindsight, how things might have been done differently and if re-elected, how you would try to limit these errors.

   Good questions. Considering that one of the things that prompted me to start editing again in 2015 was the Gamergate case, which was widely reported in the press at the time, I actually have very little critical to say about that case. (Granted, it's massive and I've never dug through it in depth.) When I've looked back through older cases - either because a current issue came up at ARCA or because I was curious to see how previous committees had handled a situation similar to a current one - I notice a couple of common antipatterns. One is that the committee is sort of unavoidably terrible at dealing with cases that involve serious off-site or real-life harassment. Arbcom has very little investigative power outside of the English Wikipedia, and no outside institutional recognition with which to attempt to mitigate the effects of harassment on editors, other than calling on the WMF to help out. What tends to happen, though, is that the harassed editor is stressed and frustrated, and makes this known in their on-wiki behavior, and the grinding-on of an arbcom case makes it worse, and then arbcom says "hm, they aren't behaving very well!" This is unpleasant, even in the case where the editor really was causing problems or behaving poorly before the harassment started. You saw this pattern in the Gamaliel case, you saw it in the Lightbreather case, and there are older cases with similar histories. I wish I had some better ideas on how to mitigate the on-wiki effects of this pattern, even though arbcom as such can't do much about the off-wiki effects. I'd be tempted to just suspend the case if another instance came up with uncontrolled outside interference, but that leaves a case hanging over someone's head, creates the wrong incentives for would-be harassers, and doesn't help the community solve whatever problem the case was originally about.

   The second antipattern is common to most of the cases about "civility", usually involving editors who are productive in mainspace and work well with people with whom they have an established history, but are volatile and don't take criticism well. I thought in 2015 that both the AE1 and AE2 cases looked an awful lot like arbcom was playing the role of Eric Cartman - you know, "respect my authoritah!" - and was letting itself get dragged into distracting power struggles. The thing that really stood out in 2015 was desysopping Yngvadottir over AE "reversal" issues - I understand the logic that led to that decision, but it was so wasteful and looked from the sidelines like a power play. (But see Carrite's question above; pretty much everything I thought was a power play turns out to be kind of the opposite, just normal non-powermongering people muddling along making decisions that sound OK in the moment but aren't very strategic.) I honestly thought I'd be better at this, but I found myself sliding down exactly the same mental grooves when confronted with editors who just would not take feedback on board. I think the big take-home message here is to be very careful about sanctions and rules so you avoid boundary-testing and brinksmanship - I remember how very tempted I was to go do totally innocuous things that technically violated that mass injunction from the beginning of AE1, so I can only imagine the temptation is worse when it's a sanction directly applied to you and not a dragnet you happened to get caught up in.

   One more thing (groan...), I'll join the others in criticizing the non-party remedy in the Hardy case (as someone who was entirely on board at first). No better example of how easy it is to see things from within the arbcom bubble and forget the experience of watching arbcom work from the outside. The most annoying outcome of that unforced error is that it undercuts the point that the behavioral dynamics in the ANI thread were counterproductive, regardless of who exactly the behavers were. Now that's known as "the case with the non-party remedy" and not, as it should have been, "the case where ANI made a lot of smoke and then it took six weeks to determine there wasn't any fire". Opabinia regalis (talk) 04:51, 20 November 2017 (UTC)

4. What do you believe are the biggest errors of judgement that you have displayed yourself whilst acting as an arbitrator and as a functionary. Discuss how you would have approach the issues with the benefit of hindsight, how things might have been done differently and if re-elected, how you would try to limit these errors. I ask this primarily to evaluate your suitability to serve as an arbitrator, but also to provide some useful information/answers for any first time arbitrators who may be elected.

   I took a long time to get to this one because frankly I'm dumb a lot ;) Even after two years, every time someone asks me a question about the finer details of discretionary sanctions I feel like the project's best example of the Peter Principle.

   I'm probably suffering from recency bias and forgetting some of the dumb things I did early on, but one example of not being very smart is these threads about the Guido den Broeder unban and subsequent reban; most of that discussion occurred while I was wrapping up at my old job and starting a new one, and I really should have just gone inactive rather than trying to muddle along and being pointlessly grouchy and oppositional about an issue where nobody actually disagreed with the eventual outcome. I touched on this above, but another example of not thinking things through (and without even the excuse of real-life distractions) was the "non-party remedy" in the Hardy case. I know there's been quite a few instances where someone called me out, either on-wiki or by email, when I said something that clearly reflected the reality-distortion field from within the arbcom bubble, and I hope people continue to do that if I'm reelected.

   I mentioned in my statement that I've come around to stiffer sanctions (at least, stiffer than my instincts typically suggest) for cases of repeated disruptive behavior. Two years is long enough to eat your own dog food, and you can actually get some feedback on the longer-term outcomes of the decisions you made or argued for. I've noticed that there have been several cases where I advocated either a lenient sanction or a very narrowly scoped one, and where I got my way, and where either the committee or the community or both eventually had to add more sanctions because the original ones were ineffective. Probably the best example is the first Magioladitis case, which I drafted, and which obviously didn't solve the problem considering we had a whole second case a few months later. I do think there's value in asking the community to clarify some policy or procedure if ambiguity has contributed to the conflict. But in terms of individual remedies, I'm now more in favor of drawing the boundaries of topic bans more broadly rather than trying to surgically excise someone from the exact areas they've caused problems in - that tends to encourage boundary-testing rather than disengagement. One thing I'm kind of surprised about is how little systematic reevaluation we as a community do of the success or failure of different types of dispute-resolution efforts. I'd be really interested in seeing more of that, because we shouldn't have to keep trying dog-food recipes we already know taste bad. I made a very limited effort at reviewing arbitration remedies that recommended community action, but never really followed up; I'm kind of hoping to get around to more of that whether or not I serve another term.

   Also, I cheated and peeked at Callanecc's page and if he was wrong about the advanced-permissions thing, then so was I. I was all for it until someone in the community comments phase pointed out that our proposed procedure and the stewards' procedure for emergency issues should probably be the same. Opabinia regalis (talk) 04:10, 21 November 2017 (UTC)

Questions from InsaneHacker

2. The arbitration policy states that arbitrators are expected to recuse themselves from proceedings if they have a conflict of interest. What do you consider conflicts of interest? Are there any situations that may be considered COI "grey areas" that you have an opinion on?

   Any time an arb is already involved in a situation or feels unable to make neutral judgments about it, they should recuse, no question. I participated in the 2015 GMO case and have edited in the area (though not recently), so I've always recused as an arb from anything arising from that case. I've acquired a few more real-world COIs since last year - I work for a for-profit company now, not an academic institution - and so would also recuse in the unlikely event that my employer or clients had something to do with a case. FWIW I know there have been disputes about this in the past, but I can't recall any real conflicts or uncertainty arising during my term about anyone recusing or not. Opabinia regalis (talk) 18:31, 19 November 2017 (UTC)

4. There is currently no requirement that ArbCom members have to be administrators, but historically every arbitrator has also been an admin. Do you see any value in having non-administrators on the commitee, why/why not? (I'm especially interested in answers from non-admin nominees)

   To be honest I don't see all that much overlap in skillsets between the arb and admin roles, so I'd be happy to see good non-admins elected. Despite what the archives of WT:RFA would have you believe, most admin work is easy, and most of the socially difficult work most similar to the arbcom workload is typically done by admins but doesn't actually require the tools. I actually considered dropping the bit after the 2015 election as a way of pushing this issue, but decided to get a feel for the work before pulling any tricks like that, and it turns out that I just don't think it would be very practical to do the job without the admin tools. Minimally, you need to be able to block and unblock and delete, and if you can do those you might as well have the whole package. So from a practical perspective I think this is only viable if the non-admins get the bit upon election. Opabinia regalis (talk) 18:31, 19 November 2017 (UTC)

   Addendum: I thought I remembered this coming up before. This 2015 RfC says There is a strong consensus against granting the administrator right automatically to non-administrators upon appointment to the Arbitration Committee. I haven't read the RfC itself, so I don't know whose opinion I'm making fun of here or whether I'm missing a point that would change my mind, but based on my own experience I think that position is straight up cuckoo crackers. Opabinia regalis (talk) 20:00, 19 November 2017 (UTC)

6. Do you think that administrators or users in good standing who generally contribute to the project in a constructive manner should be given more leeway when it comes to sanctions against them (also referred to as the Super Mario problem)? If not, do you think this happens currently, and if so, what can be done to prevent it?

   I think your description is much more general than the "Super Mario" effect, which IIRC originally meant: "behavior that gets an administrator desysopped would get a non-admin banned". As I said about a similar question in 2015 (#3 there), the narrowly defined Super Mario effect is empirically checkable. I haven't actually done the analysis I proposed there, but my guess hasn't changed; I suspect this effect is more perceived than real. The broad version you describe is certainly real, but I don't think it fits the "Super Mario" paradigm, and I consider it a feature, not a bug. Committed, productive volunteers have earned the benefit of the doubt - even when they're being disruptive, you can make a good guess that they mean to help. This doesn't apply only to highly visible people - if you've been around for a year and have 5,000 edits and wrote a GA, then we have good evidence that your work is valuable to the project. If you've been around for a week and have 499 edits and most of them involve complaining about "discrimination" because the Palestine-Israel articles are EC-protected, then we don't have that evidence. It's just a matter of having an informative prior. Opabinia regalis (talk) 19:23, 19 November 2017 (UTC)

Questions from Kingsindian

2. One of the first acts of your ArbCom tenure was banning The Devil's Advocate (TDA) for off-wiki activities, which ArbCom said amounted to harassment. According to TDA's account, TDA was banned without informing them that they were being investigated for harassment, and they were not given any chance to respond to the charges. (Indeed, according to their account, they did not get any response from ArbCom at all prior to their banning, except a short note of acknowledgement of their email.) As far as I am aware, this part of TDA's account is correct.

   At the time, I (among many other people) raised the concern that this action by ArbCom violates WP:ARBPOL provision In exceptional circumstances, typically where significant privacy, harassment or legal issues are involved, the Committee may hold a hearing in private. The parties will be notified of the private hearing and be given a reasonable opportunity to respond to what is said about them before a decision is made. (ignore the case of Cla68 for now, and just focus on TDA). See also this RfC on the Village Pump about private hearings in cases of alleged harassment. Can you explain how ARBPOL was not violated, and if it was, why the committee chose to do so? Kingsindian ♝ ♚ 04:30, 20 November 2017 (UTC)

   I can't say I think of every individual act of arbcom as a new "hearing". Especially since TDA came within a hair of being banned in the Gamergate case, was serving out a long block under the provisions of that case, and was banned explicitly under the remedy from that case: should future misconduct occur in any topic area, he may be banned from the English Wikipedia by motion of the Arbitration Committee. That being said, with the benefit of experience I would try to do better with coordinating communications. This had a preventative aspect in that TDA's then-active block for unrelated misbehavior was due to expire soon; if a similar situation recurred I'd extend the block, email him, and then proceed with a ban if the committee was unsatisfied with the response. Opabinia regalis (talk) 04:30, 21 November 2017 (UTC)

4. Thanks for the response. I have a couple of comments/questions on the answer (which I found very unsatisfactory):
   1. I find the notion strange that an ArbCom motion indef banning a user is not a "hearing". ARBPOL is not a legal document and IANAL, but the aim of the ARBPOL provision, as I understand it, is to ensure that some semblance of due process be followed. Since the case by its very nature cannot be public, ARBPOL makes a provision for a private "hearing" (call it whatever you want), but mandates that the parties be given a chance to see the charges and respond to whatever they are being accused of. It is this issue about which most people were ticked off (in the RfC I linked above, for example), and not some marginal issue of insufficient communication.
   2. The preventive aspect of the ban: Given that the activities for which TDA was indef banned were entirely off-wiki, how is the ban preventative? Did TDA have a history of outing users on-wiki? (No. Indeed, they haven't even outed the user in question off-wiki in the couple of years since then.) Finally, I read over the discussion on the ArbCom noticeboard at the time to refresh my memory (see this and this (I did a Ctrl-F for "prevent"), and not one Arb explicitly made this "TDA's ban was preventative" argument at the time -- though some other people did, with about the same persuasiveness as above. Kingsindian ♝ ♚ 05:32, 21 November 2017 (UTC)

   Sorry you're unsatisfied, but I think this cud has been chewed well past the point where there's any nutrition left in it. Opabinia regalis (talk) 08:40, 22 November 2017 (UTC)

Question from Banedon

2. You write in your answer to InsaneHacker #3 that the Super Mario effect is more perceived than real, but you also write that committed, productive volunteers (should have) earned the benefit of doubt. Does this mean that you think Arbcom should discriminate more in favour of committed, productive volunteers when it comes to sanctions, since if Arbcom did do this, the Super Mario effect would be more real than perceived (and that is desirable)?

   As I said to IH, I think this formulation stretches the "Super Mario" analogy well past the breaking point. When Super Mario hits an enemy, he turns into regular Mario (gets desysopped). When regular Mario hits an enemy, he dies (gets banned). That's it. There's no room in the logic of video games for squishy subjective stuff like "the benefit of the doubt". Also, I don't like the word "discriminate" in this context; it has too much of an association with behavior based on prejudice and bias, when what I'm talking about is exactly the opposite - that is, judging people based on their individual history.

   If we didn't want anyone making judgments about other editors based on their past history with the project, we wouldn't all have handy contributions lists easily available, we wouldn't worry about IPs' lack of a persistent identity making it difficult to interact with them, and we wouldn't talk about or write rules about things like "accountability". We could have gone the 4chan route and let everyone edit as "anonymous" if we had no interest in persistent identity or relationships, but that would be an impractical way to write an encyclopedia. I think this is important because, unlike a lot of online communities to which we're sometimes compared, we're organized around developing a product, not just playing a game or chatting on a forum. We have obligations to our readers, not just to each other, and that makes the quality of someone's contributions important. Of course I'm not talking about anything so transactional as "you get three free personal attacks for every FA you write" (what would be the exchange rate on that, anyway?). I'm talking about making full use of the information we have available in order to best serve the project's goals. Opabinia regalis (talk) 05:02, 21 November 2017 (UTC)

Question from Iridescent

2. (Note that this is prompted by, and a follow-on question to, your answers to InsaneHacker above.) A couple of weeks after you originally joined Arbcom, you appeared to agree—or at least, didn't appear to disagree—when I said I've always thought arbcom clerking should be deprecated—I've never cared for the notion that the arbs are special snowflakes who can't be expected to descend from their ivory tower to maintain the dozen-or-so pages in their purview, nor for the idea of arbcom having its own private police force to stop the great unwashed saying things the Great Council of Elders might find objectionable. (Plus, as you've probably discovered by now, the position attracts more than its fair share of oddballs whom I certainly wouldn't want issuing statements or performing actions in my name.). In the light of two years experience on the committee, do you (a) feel that arbitrators have any kind of special status in comparison to people working in other areas of Wikipedia, and (b) if (as both yourself and NYB assure me) the committee workload has dropped so drastically in recent years, do you feel the existing "we do the thinking and the servants do the dirty work" setup is justifiable in light of the fact that no similar formal hierarchy exists in any other part of Wikipedia, or do you feel that if the Committee orders an action to be taken they should be willing to take that action in their own names?

   Are lightning rods and punching bags "special statuses"? ;) No, of course there should be no "status" outside the arb pages, except whatever status as experienced editors/wise elders/village laughingstocks arbs may acquire on their own merits. In practice, I notice every so often that someone acts as if something entirely uninteresting I said on some random page about some totally unimportant subject is significant because an arb said it, and it's always surprising to realize people are thinking of me that way. I think my actual interpersonal persuasiveness has declined over the last couple of years - when I got the bit back I couldn't warn someone of something without it being interpreted as a threat, because I could block them (even if I had no intention of it); then when I joined the committee it became impossible to make a prediction without someone assuming I intended to use Secret Arb Powers to make it come true. Being part of The Establishment is weird.

   As for the clerks, I'm of two minds there. I don't like the "hierarchy" aspect of the system, but I don't think it's quite as unique as you suggest; SPI has clerks too, for similar reasons. I wouldn't want to sound unappreciative of help people have volunteered to provide, or to nitpick a job I'd be frankly terrible at. I do appreciate that someone might fix my bad formatting and remember to update the templates and do the notification posts and whatnot, but realistically, yes, the workload is not so overwhelming that I couldn't do that stuff myself. The fact that there are people who specialize in doing these jobs lets the procedures get increasingly complex, and the fact that I don't have to do them means I don't care how complex they are, and that's not a great set of incentives.

   For the part of the job that's "dirty work", and not just tedious, I gather that the argument in favor is that clerk actions provide a useful separation between the arbs deciding a case and the clerks merely performing maintenance as it progresses. (I'm thinking of removing personal attacks from case pages and the like.) Unfortunately I don't think anyone buys that argument. I have no problem with knocking down that strawman and having clerks accompany their actions with "Opabinia told me to do this", but I worry there wouldn't be enough water to dump on case-page brushfires if we dispensed with the system entirely. The email/case/ARCA workload mostly operates at a slower pace than the demand for "dirty work" clerking - speaking for myself, I can keep up with the former much better than the latter. Opabinia regalis (talk) 07:44, 21 November 2017 (UTC)

Questions from DGG

2. Officially, Arbcom is supposed to deal primarily with conduct disputes, not content, and to interpret and apply policy, not make it. But it has always seemed to me that most conduct disputes have their origin in disagreements over content, and that Arb Com has in fact been most successful when actually dealing with content concerns, as in the pseudoscience and nationalism related cases, even though it may have to word it indirectly. And it has also always seemed to me that the necessary interpretation of policy can in effect amount to making policy, as with the cases involving BLP. What do you think? DGG ( talk ) 04:15, 21 November 2017 (UTC)

   I actually had the impression that the proportion of cases that originate from content disputes has declined recently, but I can't say I've done the statistics. A lot of hot-button content areas have by this point reached arbcom and been placed under DS as a remedy (so now the bickering is displaced to AE and ARCA, which at least is better than doing it in mainspace... :) In the past arbcom has had a pretty bad reputation for content-oriented cases - I know this is going back a while, but the climate change cases are the example I had in mind - and I do think it's difficult to pass judgment on behavior during a content dispute when you're not all that knowledgeable about the content at issue. Problems like POV-pushing, misrepresenting sources, and persistent unwillingness/inability to use reliable sources are "conduct" issues, but can be difficult to identify, and if a dispute gets to arbcom it's near-guaranteed that both sides of the dispute will be accusing the other of these behaviors. As a participant in the GMO case, I thought this problem made itself evident. But on the other side of the fence, I felt a little self-conscious about reviewing the evidence and voting in the War of the Pacific case - what I know about that topic could be written on a postage stamp with room to spare.

   You're right that arbcom has sometimes developed remedies that were effectively policymaking - the original BLP "special enforcement" is an example, as is the "resigned under a cloud" formulation about adminship. Most of these results were accepted by the community in the end because they codified common sense. I think there are many fewer policy gaps now and the "modern" arbcom is unlikely to need to do this, or to get away with it. Opabinia regalis (talk) 04:55, 22 November 2017 (UTC)

4. There have been very few actual arb com cases in the last few years, which might indicate that the community is doing better with its problems, and that the basic rules are becoming well understood. It seems to me that most of the business at arb com has been dealing with ban appeals, which is done on the mailing list, and often involves considerations of privacy. I'm not sure we do very well at this. Based on your own experience and perception of arb com, what do you think about this? DGG ( talk ) 04:15, 21 November 2017 (UTC))

   I wouldn't say "most" of our business is appeals, but they certainly do arrive at a higher rate than cases/ARCAs/etc. I don't think we're very efficient at it, to be honest - the informal system of "whoever is interested in an issue re-circulates it on the mailing list until completion" doesn't work well with appeals. I did a fair amount of donkey work on tracking emails in 2016 but that was one of the things I've pulled back on since I've been busier IRL. I don't find that appeals take much time, really, but they take a lot of mental space. Reducing the cognitive overhead of tracking outstanding requests (appeals or otherwise) is part of why I want to magic-wand a working CRM into existence. As far as I can tell this has never been something that arbcom as an institution has been very good at, and when appeals have been handled efficiently it's usually been because an arb or two specifically took responsibility for the job. Opabinia regalis (talk) 04:55, 22 November 2017 (UTC)

6. When I joined arb com 3 years ago, most arbs thought that the terms of use were not necessarily enforceable policy at the English Wikipedia, and that arb com has no role in its enforcement. I strongly disagreed at the time--I think they are inherently policy to the extent they are applicable, and arb com has the same jurisdiction as for other behavioral policy. (Of course, we may want or need to interpret it further--and certainly can extend it.) To some degree, I think it possible that the prevailing opinion may have been changing a little towards the position I hold. Do you see it similarly from your experience at arb com, and where do you stand? DGG ( talk ) 04:15, 21 November 2017 (UTC))

   The WMF Terms of Use document has a whole bunch of stuff in it, but for some reason "enforce the TOU" has become some kind of code word for "investigate potential undisclosed paid editors". It's probably an artifact of my inactivity at the time the paid-editing terms were added to the TOU, but I've never understood why "is the TOU policy" is such a political issue; it seems to me analogous to the fact that arbcom is in charge of local CU/OS permissions but there's also an ombudsman commission responsible specifically for global policy. We have a local policy, WP:PAID, and if the WMF wants to have a separate structure specifically for enforcement of the TOU document, they should do that.

   Where I think the WMF could really contribute here, though, isn't so much in enforcement as in tool-building. There's a lot of friction around this issue because disclosure is fiddly, templates are hard to use, many paid editors are inexperienced with MediaWiki, and all of this creates loopholes unscrupulous people can and do drive barges through. I'd like to see tools developed along these lines, to streamline the high-friction processes and make TOU compliance easier:

   • All paid edits must be segregated. No mixing paid and volunteer edits under the same account. All volunteers must disclose their paid account if they have one. All paid editors must do all of their work under one account.
   • "Paid editor" is a user right. It's mutually incompatible with all existing rights other than auto/extendedconfirmed (or better yet, it has a separate "paidconfirmed" flag that prevents new-article creation). If someone is weaseling about disclosure, you say "Oh, gosh! There must be a mistake, you don't have the paid editor right! I'll give it to you right away!"
   • If you have the "paid editor" user right, you're prompted when you save an edit to fill in relevant disclosure details (maybe it'll retain the last set you used, to reduce repeat typing). This is baked-in to the page history (as a tag, maybe?). It's filterable and queryable - you can look up all edits made on behalf of the Doe Corporation, or all edits from paid accounts to the Doe article, etc.
   • There's a paid-pending-changes queue that volunteers can review and accept edits from, so paid edits will be reviewed before hitting mainspace but attribution is preserved.

   
   Now, I doubt they'll actually do any of that, probably with the claim that investing technical development in paid editing would legitimize it. But practical changes to reduce the pain points in the existing process would be a lot more useful than arguing about the bureaucratic status of the TOU. Opabinia regalis (talk) 07:11, 22 November 2017 (UTC)

8. As I see it, most arbs are of the opinion that the requirement that editors avoid outing applied equally to good faith and bad faith editors. I however think that it ought to be interpreted to apply with much less rigor to those who appear to be editing in bad faith or deliberately against the terms of use. (I recognize the difficulty in deciding initially who is editing in bad faith) Where do you stand? DGG ( talk ) 04:15, 21 November 2017 (UTC))

   I've said this before, but the key feature of outing is its irreversibility. Suppression takes time, and can't make people un-see what they saw, and can't stop people from making copies, and can't remove information from mirrors and dumps. We know from experience that exposing someone's personal information on Wikipedia is both used as a form of harassment, and exposes the victim to further harassment by others. Who would risk causing irreversible harm to another person based on their speculation that that person is acting in bad faith? That seems like an even more bad-faith act. What if that speculation turns out to be wrong? I'm wrong at least a dozen times a day, about all manner of dumb stuff, and I think I'm somewhere in the fat middle of the wrongness distribution. I'm pretty sure most Wikipedians are right there with me. (Maybe there's a few in the right tail of that distribution, but how would I know? After all, I'm wrong a lot!) In fact, "sleuthing" as a means of catching other editors in bad acts has been rejected precisely because it has led to spectacularly wrong conclusions in the past. Incidentally, the TOU specifically prohibits violating the privacy of other users (even if you suspect them of violating the TOU!) so this idea should be a non-starter for anyone arguing that the TOU are binding local policy in any case. Opabinia regalis (talk) 07:11, 22 November 2017 (UTC)

Questions from Fram

2. You voted for the unban of User:Guido den Broeder and were vere evasive whe asked for an explanation and arguments for this (see Wikipedia talk:Arbitration Committee/Noticeboard/Archive 35#Guido den Broeder). Why did you support the unban of a repeatedly banned sockmaster with a myriad of editing problems over many years and some very wacky ideas? Why was the voting on this unban not made public (who voted for or against, and who recused or didn't participate)? Would you handle things differently in a similar situation in the future?

   No, I wasn't "evasive", I was annoyed with you for demanding immediate attention to your non-urgent complaint. But I touched on this above in responding to Nick; I was busy IRL at the time and annoyed with Wikipedia in general for being a timesink, so you have my apology for sour attitude. On the substance I don't think I have anything in particular to add to what I said at the time - he'd stated an interest in editing innocuous areas unrelated to the topics he ended up editing. I don't think we'd been all that consistent about how to format unbans and where to put the notifications and whatnot; it just never really crystallized after the residual BASC backlog was dealt with in early 2016. We've explicitly asked for community input on a couple of cases since then (OccultZone, currently Δ), which works well when someone was banned for behavior that occurred on-wiki. That would work for recurrences of Guido's situation, but that was in a bit of a procedural black hole thanks to the age of the ban, so that set of circumstances is unlikely to occur again anyway. Opabinia regalis (talk) 08:26, 22 November 2017 (UTC)

Question from SilkTork

2. Hi. Thanks for stepping forward. I am asking this same question to all candidates. What can the committee do that the rest of the community cannot? SilkTork (talk) 15:03, 22 November 2017 (UTC)

   Shhh! Don't tell anyone about the WMF's arbcom-only liquor reimbursement program! ;)

   OK, I use this turn of phrase as much as anybody, but in this context I'm going to push back a little on the idea that "the committee" and "the community" are two distinct groups. I hope we aren't being voted off the island by standing in the election! :) The major distinction here, I think, is the difference between decisions made by a small, pre-selected group of editors and decisions made by open-ended, self-selected participation. The committee is the mechanism by which the community does certain things, because those things are best handled by small groups of trusted users. The current set of tasks in that category is mostly dealing with private information, managing advanced permissions, and dealing with administrator misconduct. For obvious reasons you need a relatively small group of trusted users for the first job. For the second, removing advanced permissions needs a small group for the same reasons as the first. (You can go to the ombuds if unsatisfied with a local solution, but that's basically the same structure - a small group of elected individuals.) Granting advanced permissions by open community elections has been tried and ultimately judged impractical.

   The third job, dealing with potential administrator misconduct, is a bit of a social fiction - in non-emergency situations, bureaucrats and stewards are willing to desysop someone if arbcom says they should, and not otherwise. It's perfectly possible (if highly unlikely) for everyone with the technical ability to implement a desysopping to collectively refuse to do so, if arbcom has gone crazy enough, or to go ahead and desysop without us (never been tried AFAIK, but who knows, it might stick!). It's also possible for "the community" to vote someone off the island by banning them, circumventing the need for desysopping, though I wouldn't recommend that approach and would be very disappointed to see it attempted. I haven't yet seen a proposal for "community" desysopping (that is, via open-ended self-selected participation) that wouldn't be more drama than an arbcom case. Cases may feel slow and bureaucratic while you're caught up in one, but they're intentionally structured to be investigative, deliberative, and not prone to rash decisions. I've said before that sometimes I think the value of a case is less in the decision itself (which may indeed reach the same conclusions the noticeboards did a month earlier) than in the structured format that separates evidence from proposed actions based on that evidence. Noticeboard threads, RfCs, and other mechanisms that rely on self-selection tend to have a momentum-gathering effect that can make it very difficult to focus on the core issues or to objectively evaluate the evidence being discussed. Opabinia regalis (talk) 18:48, 23 November 2017 (UTC)

Question from Biblioworm

2. On this page, I have drafted some detailed proposals (already written as formal motions) which would reform ArbCom's policies and procedures. As an arbitrator, would you propose and/or vote for these motions? If you only support some of the proposals, please name the ones that you support and the ones that you do not support. If you do not support a particular proposal, please elaborate as to what, if anything, would make the proposal acceptable to you.

   I'm sorry to be so negative, but I find most of these impractical. I get the impression they were developed by deciding first that we need "reform" and then looking around and picking some stuff that looked important, as opposed to deciding first that specific issue X is causing problem Y and then looking for a solution that fixes it without causing worse problem Z.
   

   1. Dumping the private-information responsibility on someone else doesn't fix any of the issues associated with its management and just adds complexity. Specifically dumping on the WMF is totally unworkable; not only do they not have the resources, they're far less accountable to the community on an ongoing basis than arbcom is. You don't get to elect WMF staff. (Doing this might also compromise their common-carrier status, but I'm not an expert on that so I won't speculate.) Direct election of functionaries has been tried (see here) but it didn't work well. I've been around for two rounds of functionary appointments and co-coordinated the 2016 cycle and there haven't been many complaints about the current process.
   2. I'm all for streamlining procedures, but the ones I picked out in 2015 as the top targets didn't turn out to be the bottleneck and probably looked like a weird selection to the sitting arbs at the time, so I'm not surprised that someone looking from the outside picks the wrong things.
      • The ombudsman commission has no mechanism to enforce local policy, so that one would be a nonstarter even if it weren't for the fact that auditing is a negligible part of the typical workload at this point.
      • Figuring out case scope can be a bottleneck, but having to produce a carefully worded statement about it would be slower, not faster.
      • Workshops already often start concurrently with evidence. (I admit I judge people who do this, especially if they jump right in listing who should be banninated without taking the trouble to read anyone else's evidence. I grant that reading the evidence page has probably never in the history of arbcom changed a case party's mind on which of their opponents in the dispute they wanted to ban, but still, we could keep up appearances here! :)
      • Having only parties give evidence is a bit pointless unless we really do go through with the idea that the party list is "users due for a sanction", not "users involved in the issue". I know I'm long-winded, but the suggested word limits in the bottom two sections are impractically small and jarring considering they appear on a 1300-word page that contains phrases like "hereby approves the removal of point (4) and of point (5)(i) of said section" ;)
   3. There is zero current demand for this much bureaucracy at AE. Was this originally developed in 2015? There were some notable controversies about AE sanctions and reversals thereof, but what's worked so far to moderate that problem isn't writing down a lot of detailed rules, but just taking things slower. The last couple of times there was an issue involving overturning AE sanctions, it just got hashed out at ARCA, no escalation and no desysoppings needed.
   4. I don't get how this is different from showing up on an arb's talk page and asking "hey, what do you think about this idea for a motion?" If you can't get anyone to propose it, you're not going to get anyone to vote for it.

   
   I realize that a sitting arb looking through a list of reform proposals and saying "No, those won't do" can sound like entrenchment, wagon-circling, defensiveness, regulatory capture, the consequence of being co-opted by the state, etc etc etc. To be honest I'm still not sure if some of the things I've changed my mind on since starting my term represent the wisdom of experience or the loss of perspective. But if I were in your shoes I'd either a) focus on identifying specific, actionable arb-related problems, even if they seem small, or b) focus on other areas of dispute resolution that are a bigger part of editors' regular experience (ugh, ANI). Opabinia regalis (talk) 05:06, 24 November 2017 (UTC)

Question from Smallbones

2. I’m asking all candidates this question and will use the answers to make a voter guide. Please state whether you will enforce the Terms of Use section on ‘’’paid editing’’’. Should all undeclared paid editors be blocked (after one warning)? Are administrators allowed to accept payment for using their tools for a non-Wiki employer? Can admins do any paid editing and still maintain the neutrality needed to do their work? (Note that only one admin AFAIK has declared as a paid editor since the ToU change). Do you consider the work done at WP:COIN to be useful, or is it just another “drama board”? Smallbones_(smalltalk) 00:06, 25 November 2017 (UTC)

   No comment on anything related to pending cases - and thus anything related to the intersection of user rights and paid editing - for obvious reasons. A lot of this has overlap with the questions DGG asked further up on the page; see there for an outline of what I think may be a better way to manage paid editors by streamlining disclosure and adding technical hurdles in the way of what may be tempting but inappropriate behavior. The other thing I've suggested before as an approach to the paid-editing problem (as well as a lot of other "bad content" problems, like "dated recentism" and poorly maintained/monitored BLPs) is to raise the notability standards for the classes of articles most susceptible to these issues. I'd probably start with articles about currently active businesses in that case. It's tempting to jump to the conclusion that if a problem is widespread, then what's needed is more "enforcement", but as a practical matter I think we'd get more bang for the buck from structural changes that reduce the need for specific acts of enforcement by making the problem behavior itself harder. Wikipedia is in a uniquely vulnerable position - highly visible material with deliberately low barriers to entry - and we need to be using our resources effectively to maintain the integrity of the articles we're putting in front of our readers. Sometimes I worry that we are burning up volunteer effort on trying to determine the exact motivations of the authors of articles there was never any real, organic audience for, when we could just be rejecting them on notability grounds alone.

   I don't actively follow COIN - or any noticeboards, really, because I've preferred not to get involved in other forms of dispute resolution since being elected. I like the concept of noticeboards focused on specific issues, where they can be addressed by experienced people who specialize in dealing with that issue. I have noticed that specialists in responding to one form of problematic editing (ahem, including arbs!) sometimes tend to overestimate their chosen problem's scale - after all, it's everywhere, they see it all the time! Opabinia regalis (talk) 09:58, 26 November 2017 (UTC)

Questions from Nuro Dragonfly

2. Hello, my question is simple; how will you correct the arbitrary removal of musical/band articles by specific types of 'editors' who claim a lack of 'notoriety' due to not being able to find some link to another website as somehow being the only standard WikiPedia excepts? I personally barely, if at all in my original works, will cite a website, with some exceptions. The arbitrary attitudes of these types of 'editors' is the reason that the Wiki has a serious lack of editors, who have the time and energy to correctly and with good faith write articles, to fill those missing ones, are falling by the way side. To be specific the individual attitudes of Admin Editors who have very little care for the efforts of others, regardless of some attempt at a non-biased and neutral Wiki adherence. I consider the complete body of works by musicians and bands to be the goal, not some mistaken interpretation on 'notoriety' on a specific album/song, and therefore it to be omitted. How will you deal with this matter in the Wiki Admin sphere post haste?Nürö ^{G'DÄŸ MÄTË} 01:57, 6 December 2017 (UTC)

   Making decisions about content is specifically not arbcom's job :) It sounds like you might want to review our policies on sourcing and verifiability and on inclusion criteria for music-related topics to better understand how to direct your efforts. Opabinia regalis (talk) 05:48, 6 December 2017 (UTC)

Question from Berean Hunter

2. Viewpoint 1: Policy should be interpreted as it is written and enforced as such. If the goals are not being met then the policy should be reviewed and perhaps changed but in the meantime this is the status quo. Viewpoint 2: Policy should be interpreted for its intent over the wording. Where conflict arises between wording and intent, either do not enforce or possibly customize enforcement to try to achieve the intent per IAR. How would you describe your own viewpoint relative to the two opposing views above?
    — Berean Hunter (talk) 02:56, 6 December 2017 (UTC)

   Both? Neither? It depends on the context? A little from column A, a little from column B? "Whenever someone poses a question involving two mutually incompatible extreme views, it is usually the case that they are intentionally setting up a false dichotomy as a discussion prompt"?

   Obviously, there are policies and there are policies. You don't have much room for IAR when it comes to policies with legal implications - there's no "eh, IAR, it was just a little bit of libel" - and likewise with those that affect people's real-world safety. Policies relating to content (and thus affecting our actual product, and thus affecting our readers) are generally more important than policies relating to our internal back-office processes and organization. Some of our most important community practices are not "policies" at all (though every so often someone takes it into their head that WP:AGF should be a policy, because surely changing which template is at the top of that page will significantly improve our ability to regulate what thoughts are in people's minds when they edit... ;)

   As an arb I've written plenty of text whose meaning I thought was perfectly clear and yet it didn't survive contact with the real world because of ambiguities I hadn't noticed or circumstances I hadn't thought of. And if that happens to text 15ish people have looked at, all knowing it would get a lot of community attention, then it's hardly surprising that the text of our dozens of policy pages might have similar problems. Ultimately it's a matter of degree - if people are regularly citing IAR over a particular policy or application, then it's probably due for a new RfC to reconsider the issue. On a personal level I'm more of an IAR type, but in the context of arbitration I think policy ambiguities discovered in the course of cases need to be sorted out by community processes, because problems that reach arbcom are sort of by definition not suitable for one-off IAR actions. Opabinia regalis (talk) 06:56, 6 December 2017 (UTC)

Question from Stormy clouds

2. Only seems fair to pose the question here, as I asked it the other way. You have had, per Banedon, run-ins with TRM in the past. Would you, as a sitting arb, be able to maintain civility and co-operation with him if he were elected? Do you feel that his presence, or the presence of other people who you have had negative encounters with in your time as an arbitrator, on the ARBCOM panel, would disrupt affairs to an irrational and unacceptable degree? I understand that I am indirectly asking you to comment on a competitor, but ultimately this election boils down to two opposing ideologies: maintenance of the status quo at ARBCOM, or a complete overhaul of the system. You, in my eyes, embody the former, while TRM embodies the latter. Are these stances reconcilable in your view? Regards, and apologies for the punctuality of my question, Stormy clouds (talk) 09:51, 6 December 2017 (UTC)

   Sorry for the slow response; it's been a busy week. My first reaction to this was: wait, what? I'm the status quo? Like I said in some previous response above, going from relative outsider to part of The Establishment is weird. I don't recall having any run-ins with TRM outside the context of being on the committee when issues involving him came up. I don't think our other editing interests have overlapped much. Even though arbitrators are elected individually, they work as a team, and I have no hesitation about doing that with TRM or anyone else.

   To be honest, the idea of a user sanctioned by arbcom on year then joining the committee the next appeals to the same part of my brain that has an instinctive NOPE response to being told what to do. It's a cute stick-it-to-the-man move, but... well, the history is that people running on various promises of "reform" tend to do less overhauling of the system and more reading the list/archives/arb wiki/etc and thinking "oh, that's what they were trying to do in that one case". There's no weird conspiracies lurking of course, if I were part of the conspiracy I would say that, wouldn't I?, there's no one out to get anybody in particular, there's just a lot of boring emails to wade through. Frankly the umpteenth "I haven't socked in like almost a week, can I be unbanned now?" email is a lot less interesting than thinking about how much better everything would be if you could just overhaul things however you want. Opabinia regalis (talk) 08:50, 8 December 2017 (UTC)