Wikipedia:Arbitration Committee/Anti-harassment RfC

Background

edit

On 10 June 2019, an English Wikipedia administrator was banned by the Wikimedia Foundation (WMF) from editing the English Wikipedia for a period of 1 year. A large community discussion followed, which included various statements by stakeholders, including the Trust & Safety department of the WMF, the Arbitration Committee, the Board of Trustees, and the CEO of the WMF.

Using materials provided to them by WMF, the Arbitration Committee opened a case to investigate the matter on 24 July 2019. In its final decision posted on 21 September 2019, the Arbitration Committee vacated the ban set by the WMF, took over the decision to remove the editor's administrator tools, and established that a Request for Comment would be opened to "focus on how harassment and private complaints should be handled in the future."

Overview

edit

In the open letter from the Arbitration Committee to the Wikimedia Foundation Board of Trustees, we asked that the WMF commit to leaving behavioral complaints pertaining solely to the English Wikipedia to established local processes. We also acknowledged that ArbCom has struggled to handle civility and harassment complaints in a way that adequately balances privacy against transparency, and due process to the accused against victim protection, and pledged to solicit comment from the community and the WMF to develop clear procedures for dealing with confidential allegations of harassment, based on the existing provision for private hearings in the arbitration policy.

In this RfC we are seeking feedback from the community to inform potential changes to ArbCom policy and procedure, which will be developed with input from the Wikimedia Foundation. Any amendments to arbitration policy will go through the standard ratification and amendment process before being implemented.

Participating in this RfC

edit

In contributing to this RfC, please use common sense in referring to specific past examples of harassment without the consent of the victims in those examples, as commenting here may attract new attention to an old, now-stable situation. Please do not use this space to make new accusations of harassment. This is a brainstorming session, so please contribute ideas freely; endorsing an idea or pointing out potential concerns is fine, but please refrain from bolded votes.

Although we have chosen to have this discussion in the form of an RfC and not a case, arbitrators and the ArbCom clerks will be maintaining this page as we would a case. We will strictly enforce here the behavioral expectations at an ArbCom case, and remove or move discussion that is determined to be off-topic. If you object to a clerk action on this page, do not undo it; rather, contact the Arbitration Committee or clerk team by email to request reconsideration. Disruption at this page may result in you being banned from participation.

To keep the length of this RfC manageable, we ask that you not add new discussion sections directly. If you believe there is an important area of discussion that has not been covered, please request it be added by the arbitrators by posting your proposed text on the RfC talk page.

Request for Comment

edit
The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.
In this RFC the community was asked to weigh in on 8 topics of concern regarding Wikipedia editors ("editors"), the Arbitration Committee (ArbCom), Trust & Safety (T&S), and the Wikimedia Foundation (WMF). There were common themes presented across some of the questions, so if a related question contains similar themes that will be indicated in parentheses (e.g. "Q1"). Please note that while there may be proposals listed that arose during this discussion, any significant/policy changes to ArbCom must go through the standard processes as described in the Overview.

One of the overarching themes of responses to the questions was that ArbCom will always be under some form of scrutiny or displeasure from certain areas of the community. However, since they were elected to be trusted members of the community, they should do their best knowing that a majority of users supported their term when they were elected (Q1). However, that does not mean they should be entirely absent from ArbCom proceedings (Q6) or jump too quickly to conclusions when it comes to the presumption of innocence (Q5).

Q1, on the matter of private evidence impacting sanctions
ArbCom, by its very nature, will occasionally have cases that involve private evidence - be it email correspondence or links to off-wiki websites - that cannot be publicly displayed in the public-facing case evidence. This private evidence is of most concern when it is the sole (or majority) reason for a case being opened and/or sanctions being filed; multiple examples were given where the results of a case were given without one being formally opened on-wiki, or where supposedly "private" information was actually present in diffs on-wiki the entire time.
While many agreed that private evidence should stay private, there were a few main suggestions regarding how ArbCom should deal with private information:
  • ArbCom should disclose if/when private information is being used to inform the case
  • ArbCom should "categorise" any private evidence so interested parties would know the provenance of said information
  • ArbCom should open a public case report, even if the evidence is 100% private, so that editors are aware that a discussion is taking place
  • ArbCom should only use private information when absolutely necessary - if sanctions and/or findings of fact can be based on public/on-wiki evidence, then that should be prioritised (Q2)

Q2, on fear of retaliation
To summarize multiple editors' opinions in this section, "there is no easy solution" to the issue of retaliation as a result of harassment and subsequent case filing. That being said, many of the editors agreed that if the information is public then the case should be handled publicly and not behind closed doors (Q1). Additionally, admins should be more willing to do what is necessarily “lower down” in places like ANI, and bump cases to ArbCom after these interventions are shown to be ineffective (Q7). While there was a suggestion for some form of intermediate location for cases to be handled between ANI and ArbCom, there was no significant agreement on what that should look like; among the ideas were bringing back RFC/U, having some form of formal mediation process between the users (Q8), or having the functionaries act as some form of private investigators vetting private information before it reaches ArbCom.
One supported suggestion was to allow third-party filings to ArbCom in an effort to minimize retaliation on the harassed/concerned editor.

Q3, on responding to allegations
This question follows on rather heavily from Q2, but focused more on the accused rather than the complainant. Many editors agreed that evidence should not be kept secret from the accused, except when it comes down to the safety of the complainant; if there are specific threats and/or information that could be used in retaliation, T&S should be contacted first (Q8). If there is private information, the complainant should be asked what information they would be willing to release publicly.
While the idea that "innocent until proven guilty" (Q5) was used a lot, significantly more people indicated that we (Wikipedia or ArbCom) are not a legal system, and so that should not be assumed; principles, not any specific rule or formulae should be used in relation to the accused. However, it was felt that there is an imbalance between accuser and accused, and that mediation (Q2, Q8) may be helpful to level that imbalance.

Q4, on unsubstantiated claims
This question had a fairly straight-forward consensus; all editors should be treated with respect and politeness, but there is nothing either the community or ArbCom can do to interrupt the "unpleasant dynamic" of unsubstantiated complaints and filings. A certain amount of "tough skin" is needed to edit Wikipedia, but ArbCom should not be used as therapy.

Q5, on plausible deniability
As mentioned in Q3, there is no "right" to a presumption of innocence. That being said, there was expressed a concern that there should not be any sanction unless there is a clear violation of policy; off-wiki links with no verification should be treated carefully. As every case is different, it is difficult if not impossible to write "rules" around this issue; ArbCom should use common sense and deal with limited available evidence on a case-by-case basis

Q6, on the arbitration environment
There was a fairly consistent response to this question advocating for more/better patrolling of ArbCom proceedings, in particular by the clerks. This includes word limits, lack of diffs (especially when accusations are made), and civility/arguing concerns; clerks should also be doing a better job of communicating with those who have "broken" the rules to get clarifications and/or indicate that their edits were removed for technical/procedural reasons rather than any sort of "point of view" suppression.
One supported proposal was to have ArbCom cases written in "c2:DocumentMode", where a case is presented more like an article (with clerks summarizing and updating a single document) and less like a half-threaded discussion between members (which can become heated/unproductive)

Q7, on unblockables
Much like Q2, there is no clear definition or easy solution to "unblockables"; everyone is cantankerous at some point, and we should all be treated equally. Opinions were highly variable, including many that felt there are no changes needed or that everything should be dealt with on a case-by-case basis, but the following were some of the most prevalent suggestions among the participants:
  • Admonishments and/or final warnings should be much more frequent, and actually enforced
  • Blocks should be handed out more frequently, but only as short-term blocks
  • Users with multiple (but un-sanctioned) cases at ANI, and/or those with lengthy block logs, should be looked at by ArbCom
  • More admin cases should be brought before ArbCom

Q8, on the relationship with T&S
Editors strongly feel that en-wiki issues should be handled "in-house", and only matters that affect the real world (Q2, Q3) should be passed to T&S. A better/improved dialogue between ArbCom and the WMF is also desired, with the Foundation and T&S passing along en-wiki-specific information to ArbCom to handle.
There was a desire from some editors, expressed in this section as well in previous sections, for the WMF to hire/find/create resources and training for mediation and dispute resolution, which would hopefully mitigate some of the most prevalent civility/harassment issues present on Wikipedia.

To reiterate, this close summarizes the opinions and feelings of those who participated, and are not binding; any proposals or suggestions that change policy will still need to go through the formal procedures as outlined in the Overview.

Signed,


1. Private evidence

edit

Many complaints related to harassment and abusive behavior involve private evidence. Most commonly this is because the alleged harasser is contacting the victim off-wiki. Sometimes there is personal context necessary to understand a pattern of behavior occurring in public. Existing community processes refer people with private complaints to ArbCom. However, in the past it has been very common that acting on these complaints and sanctioning community members based on them proves highly controversial, often prompting extensive discussion requesting more transparency. Negative features of these conversations include speculating unproductively on the victim's identity or the nature of the harassment, attempting to identify misbehavior in the victim's edit history, accusing arbitrators of abusing their power or having political motives, and making rhetorical comparisons to star chambers, secret trials, various historical dictators, Kafka, Orwell, etc. These behaviors can be a disincentive for taking action on complaints, or allowing complaints to be considered privately.

How can ArbCom more effectively communicate with the community about sanctions involving private evidence?

Discussion
  • In some cases, the community simply must accept that the arbs cannot comment publicly at all regarding sensitive matters including threats on one's life. For those with institutional memory, there have been two highly sensitive issues that affected me in real life quite severely (and still do). There was and remains no possibility of anyone publicly discussing any of those details. There are two current arbs who have the institutional memory to fill in the gaps on this, as I won't comment further. Because of my personal experiences, when the arbs say they can't discuss something publicly, I take their word for it, and suggest we continue on that path. SandyGeorgia (Talk) 14:33, 6 June 2020 (UTC)[reply]
  • Any body that publicizes a decision while concealing the evidence basis for that decision invites mistrust and suspicion. This will be vented in discussion spaces. Arbcom can't avoid this, so the arbs need to overcome any disincentive to take action. It's expected that arbs will have the character, grit and determination to do this.—S Marshall T/C 14:39, 6 June 2020 (UTC)[reply]
  • Either the community trusts the Arbs, or they don't, and some will always question and hound them over details, while the silent majority respects their actions and is willing to take it on faith. It is entirely more preferable that Arb handles these cases than WMF simply because Arb is accountable to the community, and the WMF isn't, although it should be in some way. Dennis Brown - 14:50, 6 June 2020 (UTC)[reply]
  • The “easy” way forward on this would be to expand allowing private evidence blocks to members of the functionaries team, which by its nature of being larger is more diverse, and in some cases is more trusted on these type of matters than ArbCom is (example: {{OversightBlock}}s even of established users are rarely controversial. {{ArbComBlock}}s can be.) You don’t get rid of the private nature, but by expanding the number of individuals who are able to handle it, you open the blocks up for more review and also make it come from a body that seems less monolithic. We already get private requests for stuff like UPE (which ArbCom has kinda referred to CUs in the past).
    Dealing with harassment of a private nature goes hand-in-hand with the +oversight permission, and that group contains some of our most trusted users. Policy already allows functionaries to block based on things other administrators can’t see on-wiki. There’s no fundamental difference between that and an email that can easily be shared with your peers. Opening up the private block policy a bit more would help with ability to review these things and also give more trust from a community perspective since there’s a larger group available to review. TonyBallioni (talk) 15:00, 6 June 2020 (UTC)[reply]
    This makes sense. MER-C 18:49, 6 June 2020 (UTC)[reply]
    Vaguely reminiscent of the former WP:AUSC, albeit only in terms of review by non-arbcom members, so it doesn't seem like it would be subject to the same issues (at least such as I was aware of them). This seems like a good idea; formalized or not, it would be good to have more functionary<->arbcom interaction. ~ Amory (utc) 15:45, 7 June 2020 (UTC)[reply]
    Would this be putting the oversighters in a position where they are not arbitrating, but can confirm that what the arbitrators claim matches up with the private evidence? · · · Peter Southwood (talk): 13:55, 8 July 2020 (UTC)[reply]
  • I do not think this is the most difficult and most controversial part (tracing on-wiki harassment is usually more difficult), but if off-wiki evidence is involved, it would be sufficient for ArbCom to state that off-wiki evidence is involved, without disclosing what kind of evidence this is and who provided the evidence. (And it is the role of ArbCom to make sure the evidence is genuine). A great part of the FRAM issue was that it was not clear whether evidence was on- or off-wiki, with people trying to analyze available on-line evidence and making all kind of crazy assumptions.--Ymblanter (talk) 17:12, 6 June 2020 (UTC)[reply]
  • By and large agree with Smarshall. More generally, I have long noticed that arbcom's communication problem is that it (the COMMITTEE) doesn't actually do after action communication -- it announces its decision, and then it is every arbcom member for themselves. Suggestion: In general, find a way to appoint one, two or three of you to speak for the committee (in each such matter) - have them announce that they are the point person in that matter - other committee members communicate through the spokesperson perhaps in hourly rotation (if something has to be taken back to the committee by the spokesperson say that and be upfront). I think this is especially important in privacy matters because how much to reveal needs to be a group decision, tightly held, not left to every committee member just out there dangling free in cross examination, so that each little reveal by this or that member becomes a torrent. Alanscottwalker (talk) 19:00, 6 June 2020 (UTC)[reply]
  • I want to begin by saying thank you to ArbCom for conducting this RfC. Now to the specific question, I agree with other editors that, unfortunately, a certain amount of uninformed criticism from some members of the community is simply something that comes with the job of being an Arb, and Arbs should try not to be inhibited from using their best judgment out of fear of criticism. And the community should accept that we don't know what we don't know, and should not WP:ABF the Arbs. On the other hand, I think that ArbCom should be reasonably strict on the point that something that does not need to be private, and is happening entirely onsite, should be dealt with onsite. Private evidence should be evidence that needs to be private, not accusations from persons who simply want to make accusations without community awareness of it. --Tryptofish (talk) 21:22, 6 June 2020 (UTC)[reply]
    • Adding: I disagree with the idea of oversighters and checkusers formally taking on additional roles as kind of like additional arbitrators. But there are perhaps some things that ArbCom could delegate to them, within their traditional roles. --Tryptofish (talk) 23:22, 6 June 2020 (UTC)[reply]
      • Tryptofish, yeah, I don’t think many of us, myself included would want to be a second ArbCom. I was just saying clarify WP:BLOCKEVIDENCE in regards to some of the more straightforward cases for CU/OS. How strictly BLOCKEVIDENCE is read/enforced changes from year to year based on the views of the sitting ArbCom (see my link above re: UPE/COI) and even outside of CU/OS in some cases seeing admins say stuff like “I’ve seen this email user sent from the Wikipedia email function that led to this block, I’ll forward it to any other admin” isn’t exceptionally rare.
        As an example JC7V7DC5768 is User:Jaredgk2008. I think that was figured out via emails and reddit messages obsessing over shoes (weird tick of that sockmaster) that was emailed to the functionaries list as well as some geolocation stuff and a bit of on-wiki behaviour comparison. The weird reddit and email stuff was what made the link though. There was back and forth on how to handle a block even though everyone involved in the discussion agreed it was him because of the BLOCKEVIDENCE bit. Eventually it was done as an ArbCom block, but from a practical perspective it really didn’t matter if it was an ArbCom block, CU block, or regular admin “appeal reserved to ArbCom block.” Stuff like email or other types of non-diffable harassment by known LTAs is something the CU/OS team can handle fairly easily. It’s different than something like a long-term user making a Twitter to harass people he considers his enemies and having to rely on judgement as to if it’s the right person. I don’t think anyone who isn’t an arb wants that role. TonyBallioni (talk) 00:58, 7 June 2020 (UTC)[reply]
        The checkuser and oversight teams are often better at doing investigations than ArbCom is, simply because they were specifically chosen for those roles, and usually have more experience. However, there is a fundamental difference between an OversightBlock and an ArbComBlock that should not be overlooked – ArbComBlocks require a majority vote of the committee, whereas Oversight and Checkuser blocks can be done by an individual functionary. Although the appeal process is the same, the process for applying an ArbComBlock is more robust. – bradv🍁 14:34, 7 June 2020 (UTC)[reply]
        • Sure, but a lot of these cases really don’t require a full discussion and vote. I’m not talking about things that are complex. Stuff like sending an abusive email to a user using Wikipedia’s email function is pretty straightforward, and depending on the composition of the committee in a given year, they may or may not care if someone else does the block. Formalizing some structure around when this would be okay would be useful. We already have admins who think “blocked based on email, I’ll forward if you want” is within policy (hang around CAT:RFU enough and you’ll see it.) Going ahead and delegating a lot of this is fairly feasible. It’s what was de facto done with UPE for years, so there’s precedent there in ArbCom motions and in reporting instructions (paid OTRS queue.) You have a 45 member oversight team, most of whom are pretty competent at this type of stuff. Letting them help deal with the obvious cases would reduce overhead. TonyBallioni (talk) 14:50, 7 June 2020 (UTC)[reply]
  • This has been understandably a really hard part of the arbitration process. The TDA and Soap bans come to mind, along with a few others where private, offsite evidence upset many. And while looking back at the ensuring discussions, while there is some completely ridiculous complaints that no one would agree with, the general sentiment- that there's a disconnect between "higher ups"- still rings true. The is likely an unsolvable problem, maybe give extremely vague descriptions for actions based on private evidence? But those have been provided before and people still didn't like them, even though that was a while ago. Thinking about two recent bans based on private evidence- Sk8erPrince and Icewhiz- the community reaction was pretty nonchalant, but maybe because the reasons for why were pretty well known.Moneytrees🌴Talk🌲Help out at CCI! 02:04, 7 June 2020 (UTC)[reply]
  • First off, be clear that a decision is based partially or entirely upon private evidence. No details need to be or should be provided as to what that evidence is, but I am reminded of the Ritchie disaster where it later turned out that a private decision was based entirely upon on-wiki conduct, and even the request "Tell us whether there even is private evidence" was met with cagey non-responsive responses. If it is made clear (preferably in ARBPOL) that private decisions may only take place when private evidence is a substantial part of the case, that would go a long way toward resolving any mistrust. Additionally, so far as is possible without compromising anyone's privacy, the Committee should still attempt to provide at least a generalized rationale for its actions when this would be possible. In some cases, "We're banning this person and can't tell you anything about why" will still be necessary, but that should be quite rare. Cases involving only public on-wiki evidence should be handled publicly on-wiki; private cases should still be handled with a balance between transparency and respect for confidentiality of private or off-wiki information. There will still of course always be some people upset, but I believe these measures would be convincing to at least most of the community that when a case is handled privately there is a genuine need for doing so. Seraphimblade Talk to me 02:30, 7 June 2020 (UTC)[reply]
  • I have no problem with ArbCom accepting private evidence, and can see numerous situations in which such a procedure is necessary. However, I do not believe that ArbCom should accept prima facie the claim that evidence needs to be private without explicit explanation as to why this is the case. Obviously, it is to the complainant's advantage for evidence not to be known to the alleged harasser, so ArbCom needs to be absolutely certain that privacy is necessary. Further, the entire body of evidence presented to ArbCom may not be necessary to be private, and ArbCom should, to the best of its ability, separate the necessarily private from the rest of the evidence, and release the remainder to the alleged harasser and the Wikipedia community under the circumstances it considers to be proper, but during the adjudication of the complaint. Beyond My Ken (talk) 02:58, 7 June 2020 (UTC)[reply]
  • Upon requests from the community, I think it should be possible to provide copies of sanitized evidence (with private information removed/blurred). --Piotr Konieczny aka Prokonsul Piotrus| reply here 04:28, 7 June 2020 (UTC)[reply]
  • Unfortunately there is a section of the community that has already made up its mind that the arbs and/or are evil (or corrupt or whatever) and nothing will convince them otherwise. Everybody needs to be firm in not allowing these people to bully private information into the public domain. More generally, I agree with Smarshal and TonyBallioni. Thryduulf (talk) 11:07, 7 June 2020 (UTC)[reply]
  • I do not think our current arbitrators are either evil or corrupt. I do, however, think that they, being human and subject to making errors, have collectively made some errors in the past. Stigmatizing everyone who disagrees with your position as having closed minds and accusing them of vilifying ArbCom hardly seems like the right attitude to go into an RfC of this seriousness with. Why don't you approach them as fellow editors who happen to have a different take on things and dial back on the rhetoric a bit? Beyond My Ken (talk) 21:57, 7 June 2020 (UTC)[reply]
    • It is my experience that there some users whose expressed opinions lead me to believe that my statement is an accurate reflection of their attitudes. This is not every group that criticises arbcom or arbitrators, I too believe they have made mistakes for example, but there is a difference between constructive criticism about specific things that you (generic) believe arbs and/or the WMF have got wrong and automatic rejection and denigration. There are examples of both on this page. Thryduulf (talk) 22:37, 7 June 2020 (UTC)[reply]
  • There are always going to be cases in which evidence cannot be made public. Short of asking the WMF to deal with all such cases the only solution is to have them dealt with by ArbCom or some ArbCom-like body of trusted community representatives. Anybody who is fundamentally opposed to actions being taken on private evidence should be reminded of this, and if it doesn't work then I don't think anything else will. Explicitly stating that some action has been taken on the basis of private evidence would help, as would posting high-level summaries or reasons when possible. Hut 8.5 11:57, 7 June 2020 (UTC)[reply]
  • I do not follow ARBCOM stuff closely, but my general impression is that so long as ARBCOM keeps the exception for private evidence narrow, and that ARBCOM is clear when they are/aren't relying on private evidence, that the community generally accepts the need and generally trusts it to be used narrowly. There will always be people unhappy with individual case outcomes, but I don't think the level of inevitable discontent has risen to a systemic problem. (Edit to clarify: Narrow means things such as off wiki evidence or oversighted content etc, not public on-wiki diffs.)
    On a related note, I will say that the Foundation has neither the competence nor the trust to handle routine claims of "harassment" or "incivility" where someone is unhappy that their edits are being strongly and persistently criticized or opposed. It is all too common that the strong and persistent criticism is warranted, and that the complainant is the problem. (That problem massively escalates if the individual brings in canvassed/meatpuppet/activist/COI/paid or other allies to attack and counter legitimate criticism.) It is appropriate for Trust&Safety to take jurisdiction over things like violent threats and escalate that as a police matter. However Trust&Safety is neither competent nor trusted to pass judgment on content quality, nor to pass judgement on the legitimacy of strong or persistent criticism that someone's content fails to comply with community policies and guidelines and standards. Alsee (talk) 14:49, 7 June 2020 (UTC)[reply]
  • The committee can change themselves, but they can never change the community.
    If the community wants to act like children and scream that's not fair, then that's on them. The fact that people can get surprised that the random strangers they only know from the Internet actually turned out to be an awful person is the most perplexing thing to me. Now, I love CaptainEek (using Eek because of their trademarked niceness); but if seemingly out-of-the-blue Arbcom de-sysopped and site-banned them based off private evidence, then I wouldn't think twice about it. I'd probably post how sad that is to see, hope the situation that led to that scenario has now been resolved, and then try to move on. People do bad things, and sometimes they are the people you don't expect them to be.
    If Arbcom is finding their work being negatively impacted by people who don't understand that and lob unfair accusations against them, then that is a major issue which should be addressed. If this is happening in arbspace, like in WT:AC/N, then the clerks need to be instructed that Wikipedia isn't a forum for people to play around in and remove offending comments from the discussion. –MJLTalk 21:13, 7 June 2020 (UTC)[reply]
  • Similar to MJL's comment above: Arbcom needs to accept that it is impossible to legislate the thoughts of others. When Arbcom makes a decision, the reaction to that decision is an expression of individual perceptions that are out of Arbcom's hands. Arbcom members are chosen, at least in part, to be the designated recipients of community outrage. Any member of Arbcom uncomfortable with that reality should either make peace with it or not run for reelection. The one thing that Arbcom could do to improve transparency while not violating privacy is giving an indication of what type of private information is implicated. It is already allowed to state, and Arbcom has itself sometimes stated, that private evidence is related to, e.g., off-wiki harassment or outing or information related to a minor. Formalize this. When a case involves private information, it should be assigned to pre-defined categories that let the community know why information is private without disclosing anything sensitive and there should be templates that announce the decision taken on that category. Moneytrees' examples above are instructive in this regards. I hope that helps. Eggishorn (talk) (contrib) 17:19, 8 June 2020 (UTC)[reply]
  • My thought is that the community should develop standards for when ArbCom should consider private evidence, with as little wriggle room as possible for discretionary exceptions, because that discretion will tend to undo the point of having the standards.--Wehwalt (talk) 19:57, 8 June 2020 (UTC)[reply]
  • Arbitrators are elected for the purpose of dealing with private evidence and taking decisions based on it. The breadth and diversity of ArbCom, and the fact that they are among the most trusted members of the community, is sufficient for me to ensure that they are unlikely to abuse their power. In exceptional cases, an appeal to Jimbo Wales remains an option. ArbCom needs to do nothing more than explain that and why a case was decided in part based on private evidence as part of their decision. Sandstein 16:41, 9 June 2020 (UTC)[reply]
  • How well does this scale? If, for example, we mandate this as an alternative to the usual process of googling obvious terms and joining the dots? Guy (help!) 20:59, 9 June 2020 (UTC)[reply]
  • Perhaps a standard "Findings of fact" section could be, in cases where it applies, "Private and sensitive information relating to this case has been used to inform the remedies agreed upon". I do not believe ArbCom should be any more forthcoming or "transparent" than they already are with sensitive personal information, even if based on on-wiki events. It should be an absolute necessity that information provided by a party under the implicit assumption that the information will not be made public needs to remain confidential. — Bilorv (Black Lives Matter) 09:31, 10 June 2020 (UTC)[reply]
  • I think I agree broadly with all comments above. TonyBallioni (talk) 15:00, 6 June 2020 (UTC) mkes good sense. However, I think *expanding* the list of people to an enlarged pool of "functionaries" is much less preferable to the creation of a special pool of people tasked with reviewing submissions of private evidence. This means more oversighters, but not all oversighters are tasked with reviewing submissions of privates evidence of harassment.
This pool of people responsible for reviewing private evidence of harassment.
  • Call them "Private evidence reviewers". Play down the "investigation" word. Of course, they have to do some investigation, but they should not become professional investigators.
  • They all require oversight permission.
  • They do not include any current arbs. Instead, the reviewers supply the arbs with the reviewed evidence deemed appropriate for ArbCom consideration.
  • They take a pseudo-legal approach, not a wiki-political approach. ArbCom plays the public politics, private evidence reviewers are to keep out of public discussions.
The use of private evidence must be limited by natural justice. If anybody is to be judged and penalised on the basis of private evidence, they have a natural right to see and respond to that evidence. In the practice of real world dispute resolution, the solution for this is for a mediator to never assert the truth of any private evidence, but to privately challenge the accused on what is known about their behavior, what they would do in certain (hypothetical (based on the alleged)) situations, and to elicit commitments to future behavior. Their commitment to future behavior becomes WP:ROPE.
ArbCom is particularly ill-suited to playing the role of mediator using private evidence. By being elected, they are wiki-political, and can be reasonably demanded in public to explain particular ArbCom actions. Better to separate ArbCom from private evidence reviewers, don't have Arbs ever engage privately directly with someone they are judging. If they have done, that is a reason to recuse. ArbCom can accept private evidence, preferably via the reviewers, but they must not engage directly with complainants in private channels.
How can ArbCom more effectively communicate with the community about sanctions involving private evidence?
Answer: By separating themselves from the private evidence correspondence and testing of that evidence. Only then can they be open on honest about what information influenced their decision.
--SmokeyJoe (talk) 01:19, 11 June 2020 (UTC)[reply]
@SmokeyJoe: To clarify: Under this proposal, Arbs would not have any direct access to nonpublic evidence, and would only deal with it via public summaries put together by the reviewers? --Yair rand (talk) 21:49, 11 June 2020 (UTC)[reply]
The Arbs may receive any private evidence/information that the "Private evidence reviewers" choose to provide to them. Ideally, the "Private evidence reviewers" will filter out the irrelevant information. The Arbs will not have any direct access to the providers of the nonpublic evidence; the Arbs will not correspond directly in private with people making private complaints or supplying nonpublic information, and will not personally investigate the truth or value of the supplied nonpublic information. Instead, the Arbs might ask their colleagues, the "Private evidence reviewers" to investigate, maybe to challenge the supplier of private information on apparent inconsistencies, if appropriate and necessary to be done privately. --SmokeyJoe (talk) 23:45, 11 June 2020 (UTC)[reply]
@SmokeyJoe: I do like this idea if the communication between the "private evidence reviewers" and ArbCom happens on the case pages. But if the "private evidence reviewers" can submit evidence privately to ArbCom, I fear it would only shuffle responsibility around without really resolving the trust issues. After all, in the Fram case WMF effectively wound up in this "private evidence reviewer with a private communications channel" role, and we see how well that turned out. Anomie 15:14, 21 June 2020 (UTC)[reply]
User:Anomie. Interesting point. Yes, maybe. My experience and knowledge of real life dispute resolution does not speak to your point, but it makes sense. The devil will be in the details. Private evidence reviewers must not pass the evidence in secret? They can’t reveal what the private evidence is, because that will reveal what was private. Should they summarise? Should they state a minimum fact: eg. “Private evidence has been received and passed to the Arbs”. This does not cross my main point, which is that the people who will pass judgement (the arbs) should not be the ones investigating private evidence, let alone negotiating with private evidence suppliers. —SmokeyJoe (talk) 12:19, 2 July 2020 (UTC)[reply]
At a minimum they should announce that private evidence has been received and that the oversighter considers it to be both valid and relevant and of acceptable provenance. The Arbs should be in a position to cross check provenance if they consider it necessary, and should be able to call on other oversighters to do so. I think the choice of whether or not a summary is provided for public consumption should be left to Arbcom, and similarly any explanation of why a summary should not be published should be up to Arbcom. This is a thing the clerks could handle? Having the oversighters validate evidence at a remove from Arbcom partly solves the issue of quis custodiet ipsos custodes. · · · Peter Southwood (talk): 13:55, 8 July 2020 (UTC)[reply]
  • Agree with a lot of the above re expanded role of functionaries, silent majority having faith in ArbCom, limits of being able to control speculation and disdain. I won't repeat that. One thing I wonder is whether this would be well addressed by a local ombudsperson. We have movement-wide ombudspeople tasked with investigating CU misuse, etc, but maybe developing a position local to this project tasked with auditing the committee would be helpful in keeping the trust of the community. They could be elected like arbs, sign the NDAs, and have access to the proceedings and mailing list, but would not participate. They could report actions they find troubling, perhaps to the movement-wide ombudspeople who can handle it as needed. This way the community doesn't have to feel like the only people who can monitor arbcom is arbcom, and so hopefully reduce the calls for release of private information in public hearings. Wug·a·po·des 23:09, 12 June 2020 (UTC)[reply]
    Wugapodes, If oversighters were to review private evidence, there would already be a group of trusted functionaries available for the task. Do you think there is a reason why this should not be considered sufficient? I would expect any oversighter to recuse if they were involved, which should go without saying, but will say it anyway, to avoid possible misunderstanding. · · · Peter Southwood (talk): 14:05, 8 July 2020 (UTC)[reply]
    @Pbsouthwood: I think I'm just wary of expanding the scope of what oversighters do. What I was thinking of is something more specialized, essentially someone (or someones) who would serve as an auditor. The skill set is similar of course, but it's more a political than a functional role which I think makes it more important to distinguish the two. The reason for not expanding the remit of oversighter is that, given the political nature of auditing, we would want to (1) have greater community control over the ombudsperson's selection than we currently have over functionaries and (2) insulate existing functionaries from the potential politicization of their work protecting private information. What you suggest would definitely work, but given the issues involved, I think we should keep in mind the harms of mission creep. Wug·a·po·des 19:44, 8 July 2020 (UTC)[reply]
    Wugapodes, That is worth considering. I do not have sufficient information at this point to agree or disagree with your concern, but it is plausible and worthy of looking into. · · · Peter Southwood (talk): 12:17, 9 July 2020 (UTC)[reply]
  • First off, don't use private evidence unless it's absolutely necessary (i.e. on-wiki evidence doesn't and can't cover essentially the same ground). Second off, clarify whenever you're using private evidence, ideally for each FoF and out-of-case action that relies on it. Third, verify that the private evidence is both accurate and germane. We've seen precisely what happens when these three tenets are not adhered to. en.wp has no appetite for a sequel. —A little blue Bori v^_^v 2020's a bust; thanks SARS-CoV-2 23:47, 13 June 2020 (UTC)[reply]
  • There are some things best left untold. If ArbCom is unable to protect a person's privacy in a harassment case because they feel pressured into transparency by the community, then they should probably deny such cases and turn the responsibility over to WMF T&S to handle it. Atsme Talk 📧 03:47, 15 June 2020 (UTC)[reply]
  • I forgot that there's an important comment to add. The WMF's UCoC accouncement stated that functionaries will be enforcing compliance and running an interim process. This isn't necessarily a good idea. If you look at the CU/OS appointments, the standards are 18 years of age and familiarity with policies. Then it's just about technical knowledge (CUs), or willingness to actively use IRC/OTRS and read the mailing list in a less common timezone (OS'ers). There is nothing about the ability to arbitrate difficult situations related to harassment. The community has little say who is nominated as a CU/OS, and there are no term limits. I think it would be highly problematic to have a walled garden of permanent volunteers arbitrate potentially complex cases with private evidence. Furthermore, many functionaries themselves are involved in many disputes as active editors. The role is technical, not a position of trust like the ArbCom.--Pudeo (talk) 21:01, 16 June 2020 (UTC)[reply]
    @Pudeo: I think that when the WMF uses the term "community functionaries" in this context, they're referring to ArbComs, actually. --Yair rand (talk) 21:06, 16 June 2020 (UTC)[reply]
    In the context of the English Wikipedia, yes. (There are plenty of alive projects without ArbCom, and also many without a single CU/OS).--Ymblanter (talk) 21:25, 16 June 2020 (UTC)[reply]
    Out of 310 Wikipedia language editions (as of 16:40, 17 June 2020 (UTC)), only 11 (including enwiki) has arbcom. (IMO) In this context, funct can be Arbcom, or lack of Arbcom, Admin/Crat/CU/OS (if any is existing), or maybe S if there is no such people with such positions. — regards, Revi 16:40, 17 June 2020 (UTC)[reply]
  • Arbcom is acting on all of our behalves by taking action and reviewing this evidence. That's why arbitrators are elected and not appointed. That's also why people are so pissed off at the WMF taking these actions. Even if this ban was perfectly legitimate and OK, the WMF isn't elected or accountable to the English Wikipedia editing body. While transparency does add quite a bit of legitimacy to the process, we shouldn't forget that fundamentally the reason why ArbCom is trusted to deal with private evidence is their accountability to the community. The secrecy of the evidence isn't the issue but rather the people who are chosen to see said secret evidence. I'm OK with secret trials in these types of situations but I need to trust that the people overseeing them are representative of the broader community. Chess (talk) (please use {{ping|Chess}} on reply) 08:14, 19 June 2020 (UTC)[reply]
  • There are, I think, four categories of evidence relevant here: (1) evidence that is publicly visible on-wiki, (2) on-wiki evidence that is not publicly visible (e.g. deleted or oversighted), (3) off-wiki evidence that can be connected to the parties without violation of WP:OUTING, and (4) off-wiki evidence where connection would be subject to WP:OUTING. The community generally expects that type #1 will not be considered "private", and that #3 is exceedingly rare but would also not be considered "private". In my limited experience the community seems to accept #2 and #4 being kept private, but still expects a summary of relevant evidence to the extent feasible even if only something like "An account strongly connected to User:Accused made threatining posts on an external forum", and generally seems willing to trust that ArbCom properly evaluated aspects like "strongly connected" and "threatening".
    IMO, the Fram case was particularly contentious because it seemed all the "private" evidence was type #1 (and it was eventually explicitly stated that there was no type #3 or #4 evidence), but the type-#1 evidence that could be independently located did not seem sufficient for the severity of the WMF's action. There was also some concern that the level of redaction on the part of WMF didn't allow ArbCom to properly evaluate the evidence for themselves. I think GorillaWarfare tried to do the right thing in the end in Special:Diff/915751333/915943801, but unfortunately enough other arbs did not go along with her approach. And that unfortunately damaged ArbCom's credibility overall with respect to use of private evidence in other cases.
    Also of note is that it's human nature to speculate on what is being kept private, and the more that's kept private the more people will speculate (cf. the Streisand effect). And it's unfortunately also human nature to accept speculation as fact and attempt to act on it, and for those speculated about to interpret even responsible speculation as an attack. The best defense, IMO, is to defuse speculation by providing as much information as possible, making it easier to use existing policies to crack down on those making attacks or unreasonable speculation. And, as unfortuate as it may be, warn accusers that they'll have to deal with a certain level of unpleasantness.
    Note this question is about evidence presented to the community as part of the case documents; privacy of complaints and of evidence presented to the accused is addressed in later questions, and I think there's benefit in keeping those cases separate. Anomie 15:14, 21 June 2020 (UTC)[reply]
  • I generally think concerns about private evidence are overblown. This is not a court of law where your liberty is at stake. This is a private website that no one has any particular right to participate. If a panel of elected users concludes that someone's continued participation is a net negative and that it would be a bad thing to disclose the evidence for whatever reason, I'm generally fine with that. Calliopejen1 (talk) 21:16, 22 June 2020 (UTC)[reply]
  • The Arbitration Committee is not without failures. However, it is still the best we got to handle private evidence because the members of the committee are trusted members of the community which are elected by the community for limited time periods. The committee is very limited in its options to reveal private evidence. I think that it is always helpful to communicate whether such a block or ban is based on private evidence relates to (possibly oversighted) on-wiki behaviour or to off-wiki harassment (as it was done in this case). Hence, I do not see a reason to change current practice in this regard. --AFBorchert (talk) 09:38, 23 June 2020 (UTC)[reply]
  • As opposed to my very firm view against secret evidence against the accused, I'm much more willing for ARBCOM to limit evidence to the general community (obviously the accused might leak it, but that would a straight block and all evidence would be OS-worthy, as may rediscussion of it). On-wiki evidence, non-OUTING off-wiki evidence, should all be given. Deleted (but not oversighted) content could also be provided or summarised in detail. Other evidence (that is, off-wiki or OSed) should only be provided in whatever detail ARBCOM thinks can be done reasonably. What should not be accepted in the future is accepting a case conditioned on reduced public transparency from the above (a la T&S obligations on the Fram case, despite the apparent absence of off-wiki evidence). It should be noted that ARBCOM must bear in mind that showing as much transparency as possible wins additional trust in cases that it can't (albeit with additional "oh god, it must have been extra bad", but I'm not sure how to avoid that). Nosebagbear (talk) 22:26, 28 June 2020 (UTC)[reply]
  • After reading this discussion I think that this question is, unfortunately, focused on the wrong thing. I think the core of the problem comes in from what is "evidence"? Diffs are clearly evidence. A threatening email sent to a Wikipedian is also evidence. What we on English Wikipedia struggle with is to what degree we consider personal testimony, that is the relating of a person's subjective experience evidence. I don't think anyone actually objects with a threatening email to be privately considered as evidence. I do think we have a substantial disagreement in what way we should allow a person to craft a narrative around publicly available diffs. Personally I think there needs to be a way for a person to do so. But there also has to be a way for ArbCom to reject that ask. We need to leave ArbCom discretion and then we need to elect people whose discretion for making such decisions we trust. And then, hardest of all, when ArbCom exercises that discretion we still need to trust them. I'm not opposed with ArbCom choosing, at times, to consult with oversighters, many of whom are themselves former arbitrators, when considering private evidence. But the decision making ultimately needs to stay with ArbCom, similarly to how OS are consulted about applicants for the user right but ultimately ArbCom makes the decision. Best, Barkeep49 (talk) 16:34, 1 July 2020 (UTC)[reply]
  • Invariably, "private evidence" that masks the accuser but names the accused is a one-way privacy policy that invites abuse. Whether the ArbCom can see through it, I don't know. We're volunteers, so are they. We like to believe people who claim to be victims, it's human nature; whether that can be overcome in a way that is satisfactory to those of us who like to see clear rationale for actions remains to be seen. There may be legal reasons we as a community are not aware of why "private evidence" must remain private, but naming the accused without naming the complainant seems to be prejudging who is in the wrong. Carlossuarez46 (talk) 18:00, 1 July 2020 (UTC)[reply]
  • Have you every worked with a good-faith editor, or have been watching one, and all the sudden in your watchlist, they get hit wtih {{checkuserblock-account}}? Turns out, despite making constructive contributions, they are a sock; checkuser evidence has it, it is private, and there is nothing you can do but acknowledge and deal with it. The same might go with SPIs, and now ArbCom. They can just present the summary of the evidence and sanction, and that is what we have to deal with. {{reply to|Can I Log In}}'s talk page! 22:37, 1 July 2020 (UTC)[reply]
    Though other checkusers can also review that evidence and raise complaints, just not publicly. Arbcom also has this access and reviews the logs periodically and also in response to complaints; there was a public incident quite recently where the committee disagreed with one functionary's use of the tool and removed their permission. Arbcom itself can be reviewed by the ombuds commission, though that commission is accountable to the WMF, not the communities. Ivanvector (Talk/Edits) 19:44, 24 July 2020 (UTC)[reply]
  • I'd have arbcom open a case page and put nothing on it, but it would demonstrate good faith to the community by allowing us to see what they're working on. That being said, I largely agree with S. Marshall. TomStar81 (Talk) 10:18, 3 July 2020 (UTC)[reply]
  • I tend to agree with what S Marshall wrote here some time ago now, and what TomStar81 wrote directly above this comment is relevant to that. When deciding cases that are based entirely on private evidence, the committee tends to simply announce that an investigation occurred and what the result is, without much in the way of rationale or thought process, which more often than not blindsides the community and invites distrust. It would enhance community trust if, when the committee takes a case based on private evidence, a case is opened on-wiki anyway, giving the community the opportunity to participate as we would in a public case. Evidence would have to be named in anonymized form to the extent necessary to preserve privacy but ideally well enough to support findings of fact, plus this would make visible to the community what the findings of fact were, the proposed and accepted remedies, and I think importantly would make visible the voting and recusal records of the arbitrators (this particular point seems to come up often in criticisms of the committee's private affairs). It's at least worth a try. If help is needed summarizing private evidence in a way which keeps private bits private while maximizing transparency, well there's a whole team of functionaries who pretty much do nothing but that. Ivanvector (Talk/Edits) 19:44, 24 July 2020 (UTC)[reply]

2. Fear of retaliation

edit

In some cases, a harassment complaint is based entirely on the public behavior of the alleged harasser and has no private component. Traditionally, complaints with no relevant private evidence are referred to on-wiki processes. However, some complainants are uncomfortable with this because they fear retaliation by the subject of the complaint or by people associated with them. This is also a common dynamic in cases of chronic incivility or long-term low-level problem behavior.

What can the community do to make on-wiki spaces where harassment is reported (AN, ANI, etc) less unpleasant? Would the community support having select examples of this problem handled privately by ArbCom instead? If so, how and which ones? How should complaints of chronic incivility be handled if the behavior doesn't rise to the level of harassment?

Discussion
  • I am of two minds on this. Only a few months ago, I would have answered an unequivocal "yes" to having ArbCom privately handle such cases, as I feared retaliation from one harasser. However, recent handling of arbcases have shown the other side of that problem. First, the person I feared was still able to affect me as wikifriends simply proxied for them in their absence, and harassed me in reprisal, so I now see that having everything as public as possible is much preferable. Second, the current Arb committee has allowed false charges of harassment to stand on arb pages, even when disproven, so again, it is better to have this information public so that all can see who lodges false charges, and who failed to address them. Third, the current ArbCom has shown itself unwilling to deal with unblockables, so giving those editors further leeway to extend their commentary in private is not likely to be helpful. SandyGeorgia (Talk) 14:39, 6 June 2020 (UTC)[reply]
  • If all the evidence is public and on wiki, 99% of the time it would be expected the case would be handled publicly. I say 99% of the time because there are always unforeseen circumstances that I simply can't imagine, which would fit in that 1%. There is no perfect system. Trying to create a 100% perfect system of "justice" at Wikipedia is a fool's errand. All we can do is the best we can do, and sometimes, we will fail. Accepting that fact is critical. Dennis Brown - 14:53, 6 June 2020 (UTC)[reply]
  • I don't think it's that much about retaliation, but the active act of making interacting with you as unpleasant as possible (with chronic incivility and hostility) so that you would step back from disputes. As long you recognize this intent, it won't work, but most editors won't have a thick enough skin to take the flak. Yes, private processes would fix this, but those have the downside of asymmetric participation. I don't know which is better. --Pudeo (talk) 15:45, 6 June 2020 (UTC)[reply]
  • To some extent, because we are a privately-owned website as opposed to the public square, we have to accept the fact that editing Wikipedia is not for everyone (the slogan of "the encyclopedia that anyone can edit" notwithstanding). So I don't think that ArbCom should agree to take a case privately simply because the complainant does not want it to be on-wiki. Whether something is actual retaliation or just ongoing nastiness, if it is happening onsite, it should not be tolerated, and we need to do better at making complainants believe that, if they do get blowback for having made a good-faith complaint, at least that will have consequences for the user who is bothering them. To make ANI less toxic would require some brave admins who are willing to close sooner the kinds of discussions that drag on and on past their expiration date. But it will still be toxic even then. And that should be something that ArbCom should be willing to take on: a history of multiple failed attempts to resolve a problem at the community level is the standard situation where ArbCom accepts a case request, and I think it would be desirable to accept cases based on chronic incivility and not just based on some more clear-cut violation of policy. And I feel very strongly that the answer to the question of whether chronic incivility should be treated as sanctionable, even if it is not harassment, is a loud and clear "yes". Personally, I'm really sick and tired of incivility onsite. The WMF is talking about "toxicity, incivility, and harassment", and whatever one thinks of WMF's competence at doing something about it, if one considers the spirit of that concern, it's a significant one, and it's about more than just harassment itself. --Tryptofish (talk) 21:47, 6 June 2020 (UTC)[reply]
  • I think on several points I agree with Tryptofish. Private handling isn't the way toward it anyway. If the accused really has been targeting someone, they'll know damn well who filed the complaint (and even if they're wrong, that individual still might become a target). ANI is good at handling clear and unequivocal cases, but is a substantially less good setup for dealing with difficult and complex cases with blame to be shared among many. I very much agree that if a matter has been repeatedly to ANI, and the matter has degenerated into shouting matches without any resolution, that indicates it is time for ArbCom to accept a case regarding the issue, where a structured format can allow for a much more in-depth analysis of the issues. But allowing private complaints of harassment based solely upon on-wiki conduct opens up the very real possibility that the body tasked with handling such complaints will itself become a vehicle for harassment via spurious complaints. When handled on-wiki, repeated filing of meritless complaints tends to result in the flight of boomerangs, discouraging the practice. At the end of the day, Wikipedia is a site whose community strongly values openness and transparency, and editors uncomfortable with those things may not be a good fit here. Seraphimblade Talk to me 02:37, 7 June 2020 (UTC)[reply]
  • The problem is the blurry standards. What is offensive to one person is just a joke to another. Another problem is the power unbalance. Recently I felt that I was a victim of some major ABF/NPA from an admin who thinks they are immune to criticism due to how well-known they are within the community, both online and quite likely through face-to-face meetups they attend (and I don't). I do feel that if I made a public complaint about them, I'd be piled upon by that person's friends and colleagues, and if any of them are vindicative, I'd be harassed one way or another for years (for example, by said admin and their buddies making comments critical of me in the 'uninvolved admin' sections or such, to show everyone what happens if you dare to criticize their clique). Now, maybe I am wrong and maybe that person and their friends would not act like this, and maybe if I made the public complain the community would side with me. But as a non-admin, I do think there is a large power imbalance here, and so I think that there is a reason for such cases to be made in private and reviewed by a selected body, to protect the less empowered members of the community from the real or perceived threat which otherwise makes them accept real or perceived harassment without complaining, due to fear of retaliation if they make their complaint public. Of course, eventually, the complaint is likely the made public, at least in some form, but I'd think this could be done only if the evidence is compelling enough to sanction the person against whom the complaint was made, so they could appeal it. But we should reduce the chances of 1) them being defended by their friends leading to the complaint being ignored because nobody wants to criticize a well-connected wiki-faction and make enemies, and 2) if the complaint is dismissed, making the complainer both look incompetent and open to indicative actions. --Piotr Konieczny aka Prokonsul Piotrus| reply here 04:39, 7 June 2020 (UTC)[reply]
  • I am not in favor of public disputes being handled in private by ArbCom if there is no private evidence to be considered. Public disputes need to be handled in public, in front of the Wikipedia community. Fear of unpleasantness or the (claimed) expectation of retaliation is simply not enough to justify a star court. Beyond My Ken (talk) 06:09, 7 June 2020 (UTC)[reply]
  • I can see value in some sort of semi-private semi-public dispute resolution body that accepts a complaint in private and makes a prima facie determination of whether there is substance to the complaint. If there is, then a neutral statement of the dispute is posted publicly soliciting evidence to be submitted privately. That evidence is then sifted to remove things like personal attacks, and irrelevant or unsupported material and then posted publicly, with (some parts) possibly anonymised. With this public evidence being the basis for a structured community discussion (with active moderation of off-topic and unsupported accusations, etc) of the issue for community resolution. This would be a lot of work for the group receiving the private evidence but it could reduce the toxicity or intimidation factors. (Details would obviously need to be worked out if this were to be implemented, and this is not the place to do that so please keep any discussion of this idea to a high level and don't oppose simply due to lack of detail). Thryduulf (talk) 11:24, 7 June 2020 (UTC)[reply]
  • ANI is generally pretty bad at dealing with complex or chronic issues involving experienced editors, and replacing it with an alternative process may well be a good idea. One thing which is problematic is that you can't file a complaint against someone without dealing with them or their associates, which is a big deterrent if that person is harassing you. If we could try to remove that element from the process then I think it would help. What is most desirable in these cases is input from previously uninvolved editors rather than discussion between the parties. I would also be comfortable with ArbCom handling select issues of this type. I know there is a requirement that arbitration shouldn't happen unless other community processes have been tried first, but this may not make sense if the problem is one which community processes are bad at handling. Hut 8.5 12:09, 7 June 2020 (UTC)[reply]
  • We would all love a magic wand to make dispute resolution less unpleasant. I regret that I have no easy or magical fixes to offer. I oppose taking routine on-wiki 'harassment' cases secret. In many cases people cry 'harassment' when their edits face legitimate, strong, persistent criticism and opposition. In many cases the complainant is the source of disruption and attacks against others. We of course want to deal appropriately and firmly with valid cases, but it would undermine both justice and trust-in-the-process to allow the process to become a weapon to slay opponents in secret. Alsee (talk) 15:15, 7 June 2020 (UTC)[reply]
  • We need a Wikipedia:Users for discussion (structured much like WP:MfD) where any user [including admins] can be nominated for discussion (the only immediate requirement being that they've edited within the last 72 hours). There'd be a strict criteria for speedy closes (for example, the filer was blocked within 24 hours of a nomination being made), and users who filed too many reports would be sent to WP:AN/I. Content dispute could be discussed freely (with WP:local consensus compromises given the chance to be heard), and where consensus can be found to sanction, de-sysop, or otherwise deal with problem users. The board could even serve as a place to give editors a chance to have their own conduct reviewed if they so choose, or to say good things about an editor who seems down.
    I solely propose this idea because there are no public forums where a user can report another user and actually know that said user would actually be discussed (WP:AN/I is structured in such a way, that boomerangs always detract from the filed report). However, this is a pipe dream because no one would ever want my ideal level of individual accountability. Also, knowing what I do, the vision I have in my head would never work in practice for this community and its culture. –MJLTalk 21:46, 7 June 2020 (UTC)[reply]
    If Arbcom wants to know why it receives so many requests handle onwiki disputes privately, it's because this community has given victims of harassment no ability to report their harasser without painting an amazingly huge target on their back. Even if a group of benevolent uninvolved admins came together to manage a specialized harassment helpline email list specifically made to privately hear out onwiki conduct issue, they still wouldn't be allowed to do anything about it without (at the very least) naming the user who filed the report (just to make a block that won't probably stick).
    If you were designing a system built specifically to protect harassers from accountability, I'd imagine it'd look similar to Wikipedia. We have a system where you (ie. victims) have to independently-- without help-- loudly and boldly proclaim the misdeeds done against you in a highly visible setting with rock-solid evidence to back up every single accusation you make while also being prepared to publicly defend yourself from any counter-accusations which could be made against you by any number of individuals during this process and having to justify why your accusations outweigh the perceived benefit of your harasser's contributions but still being calm and civil about the whole thing regardless what is said about you, your intentions, and your own ability to contribute to the very same project. Go through this process enough times, and you yourself risk being labelled "the problem" for inciting so much "drama". –MJLTalk 22:29, 7 June 2020 (UTC)[reply]
@MJL: Something like that existed for a long time at WP:RFC/U, but it was shut down in favour of dispute resolution through ANI and ArbCom. – Joe (talk) 13:36, 8 June 2020 (UTC)[reply]
@Joe Roe: I was aware of that process, yes. However, what I have in mind would seek to have a lower barrier to file reports, allow users to report themselves for generalized feedback (not all discussions need be negative), solely focus on the user being discussed, and have binding results. Many users during the WP:RFC/U abolition discussion expressed an interest for the creation of such a forum. –MJLTalk 20:52, 8 June 2020 (UTC)[reply]
While I think a revived and improved RFC/U might have a good place in dispute resolution, I don't think it would be a good answer to the problem raised in this section. If an accuser is afraid to bring a case to ArbCom for fear of retaliation, I can't see them being any more comfortable bringing one up at RFC/U. Anomie 16:05, 21 June 2020 (UTC)[reply]
  • AN/I is due to its structure (or lack thereof) incapable of seriously addressing harassment or incivility by popular, established editors. That's because it's mostly a popularity contest. I'd support ArbCom handling such cases privately and in an expedited manner. Sandstein 16:44, 9 June 2020 (UTC)[reply]
  • There ought to be a better place than ANI to discuss harassment. It's simple as that. ANI is one of the most toxic places on Wikipedia and people do not bring their concerns there because they know it will not help. Sir Joseph (talk) 20:36, 9 June 2020 (UTC)[reply]
  • I strongly agree with Sir Joseph and Sandstein and echo MJL's comment If you were designing a system built specifically to protect harassers from accountability, I'd imagine it'd look similar to Wikipedia. I believe that people who are afraid of harassment should be able to contact ArbCom directly without requiring precedent established by ANI reports that would only stir the pot. I understand that some are worried that in a private ArbCom case, unblockables and their friends will engage in harassment at not only the accuser but anyone who could feasibly be the accuser. The solution, however, is to not tolerate this behaviour. To decisively take action against any attempt to interfere with the private case. We have failed this in the past, such as the Fram ban which led to many admins and respected users committing serious harassment with impunity. We cannot blame the victims for these actions. I am disgusted but not surprised to find some allusions above to the idea that some people just shouldn't be welcome here if they can't handle brutish aggression from editors with effective immunity, met with a wave of harassment and callousness if you try to report it. I do, however, want to thank SandyGeorgia for providing their experience and giving their opinion, though it differs from my own. — Bilorv (Black Lives Matter) 09:48, 10 June 2020 (UTC)[reply]
  • Over the last year I've slowly come to the conclusion that we need something similar to what RFC/U should have been (not what it became). My thoughts on this are scattered—if I find time I'll compile them—but something in the spirit of what MJL suggested gets at the gist. Wug·a·po·des 23:17, 12 June 2020 (UTC)[reply]
    Yes (been saying that for a while now). SandyGeorgia (Talk) 23:23, 12 June 2020 (UTC)[reply]
    You're not wrong, but there is a non-trivial risk of exactly the problem that RFC/U generated - "votes for banning" with long dead horses brought out for yet another beating. Guy (help!) 15:04, 23 June 2020 (UTC)[reply]
    The risks are definitely non-trivial, if the idea were easy to implement I'd have proposed something already :) I like your idea further down of the WMF funding mediation training for community volunteers. Honestly, it may be the only way we could get such a RFCU/MEDCOM reboot to not turn into the processes we disbanded in the first place. Wug·a·po·des 23:38, 25 June 2020 (UTC)[reply]
  • Victims of harassment should be allowed to contact ArbCom privately, and ArbCom should be empowered to take out-of-case actions to deal with the harasser so as to prevent further harassment of the victim. AN/<foo> is not an acceptable venue for it, and in fact tends to exacerbate the issues more often than not. Ideally, the structure I envisage is this: The victim makes a confidential complaint to ArbCom (with all applicable evidence), ArbCom does its research, and then ArbCom quietly blocks the harasser if it's justified. If the harasser wants to refute the evidence, then the onus should be on them to do it in an unblock request to ArbCom. —A little blue Bori v^_^v 2020's a bust; thanks SARS-CoV-2 23:56, 13 June 2020 (UTC)[reply]
  • Absolutely ArbCom should take the case privately. No editor should be subjected to harassment for any reason, and action needs to be taken against the person doing the harassing. By taking such cases to committee, it frees a single admin from the burden of having to make a decision against an editor they consider a WikiFriend, and it also protects the victim from having to endure further harassment and humiliation at ANI where it becomes a popularity contest. A full arb committee reduces the chance that alliances will factor into a decision, although it is not 100% foolproof, which is why it may not be a bad idea to let T&S handle it in those instances. I don't see why ArbCom and T&S can't work together and coordinate their efforts for the benefit of the community. We're talking about real lives and mental state of mind, so yes, it is serious enough and certainly worthy of careful consideration. Atsme Talk 📧 04:12, 15 June 2020 (UTC)[reply]
  • I firmly disagree with the suggestions of moving so many harassment cases to ARBCOM (and private ARBCOM at that) that would come with a huge drop in transparency. I'm certainly open to if someone can make a viable RFC/U placement, or even (much like 3RR, AIV etc) a more dedicated harassment location that is clerked vastly more aggressively to at least reduce the in-discussion hostility. Nosebagbear (talk) 09:53, 18 June 2020 (UTC)[reply]
  • The problem isn't the publicity of these disputes at AN/I. It's the fact that consensus (among admins or regular editors) simply does not work in the case of chronic and intractable behavioural disputes where there's many people on both sides. Sometimes it's necessary to have a group of people such as ArbCom vote in order to definitively end one of these disputes. Both sides get to present their evidence without constant peanut gallery comments from people tangentially (or not at all) involved in the dispute, a group of people hears the dispute and then that group of people makes a final decision. AN/I often creates a long and drawn out process with endless comments in the vain hope of eventually finding consensus when everyone knows the dispute is probably heading to ArbCom anyways. Yes, there are numerous obvious situations where everyone can agree to sanction a specific editor; but there are a lot of high profile cases where this isn't possible. And to the people complaining that AN/I is mostly a "popularity contest" where editors who contribute a lot get leeway, that's often how the world works in general with respect to disputes. You can get away with rape & assault if you're good enough at playing sports or make enough money in the entertainment industry (not naming or referring to anyone specific due to WP:BLP; there's a lot of people this could apply to). While I certainly don't like that we live in a world where certain "high value people" are allowed to do bad behaviour, the fact is that this is a society-wide issue and changing the venue for on-wiki disputes from AN/I to ArbCom isn't going to get rid of the problem that editors with more contributions are allowed to act worse towards others. Chess (talk) (please use {{ping|Chess}} on reply) 08:35, 19 June 2020 (UTC)[reply]
  • I think the nature of AN and ANI makes it unlikely much can be done along these lines other than cracking down on bad behavior. But that sort of thing is a whole other set of issues, requiring dealing with things like ways of being provoking while maintaining a veneer of civility and cultural differences in register.
    I don't think going straight to completely-private ArbCom cases is a good answer though. I think Thryduulf above and SmokeyJoe in the previous section have the right sort of idea. If someone is afraid to come forward publicly, they might contact someone (not ArbCom) who would look into it, and if they find merit engage in a bit of parallel construction to build a case without reference to the original accuser. The case (whether ANI or ArbCom) would proceed from there like a normal public case, with the "private evidence investigator" serving as a proxy accuser to whatever extent an accuser is actually needed. I'm sure this wouldn't solve all the problems, but I think it would be significant.
    It's possible that ArbCom was hoping for parallel construction in the evidence phase of the Fram case, the community might have come forward with all the same evidence that was in the WMF's document which would have allowed them to dispense with the troublesome private evidence. But if that was a hope it doesn't seem to have played out. Anomie 16:05, 21 June 2020 (UTC)[reply]
  • Permit anonymous complaints, including by third parties uninvolved in the dispute. Then blame cannot be assigned to the "victim". We shouldn't make regulation of user behavior dependent on the willingness of victims to come forward -- problem users should be dealt with regardless, because their continued participation can affect the continued participation of other users (including potential new users). Calliopejen1 (talk) 21:15, 22 June 2020 (UTC)[reply]
  • As long as all evidence is public a case should be made public as it naturally profits from the comments and insights of all. Yes, there is an understandable fear of retaliation. But there is also a danger of an unfair judgment if the public is not invited to discuss this openly. As unfortunate this all may be for the involved, we as community can also learn from this to be more alert in regard to on-wiki harassment and to the patterns involved here. Private cases exclude this opportunity. --AFBorchert (talk) 09:46, 23 June 2020 (UTC)[reply]
  • Keep it public unless there is private evidence. There's sufficient scrutiny around these things that retaliation is unlikely to pass unnoticed. Guy (help!) 15:05, 23 June 2020 (UTC)[reply]
  • Our problem is that we have a lot of clear cut harassment by members of the community in bad standing (i.e. vandals, drive by trolls, or their more disruptive sibling the LTA). These are dealt with regularly and, with the exception of a few particularly troubling LTAs, relatively effectively. Because the barriers to claiming harassment are lower here than in say a workplace we also have more unjustified claims of harassment. These unjustified cases can either be with genuine or manufactured concern and it's hard for the community to discern that intent. Many, possibly even most, of the charges of harassment brought against established editors fall into this category on unjustified. These two facts already mean it can be harder for the community to recognize justified harassment by an established editor. And that is before we factor in the social power that an established editor can bring to a discussion but we need to factor that in because in justified cases of harassment power is a key element of the harassing. Our public methods is absolutely the right approaches to handling clear-cut harassment. And based on the current status of the community it is also very effective at dealing with unjustified charges of harassment. What it is ill equipped to do, incredibly ill equipped, is to pick out the cases against established editors where there might be something there and handling those complex cases well. The previous attempt to do so, RFC/U, failed and ANI doesn't even pretend to do that. This isn't to slam public community decision making. I cherish that element of Wikipedia deeply. But I do think it leaves us no good decision, as we're either compromising on that essential element of who we are as a community or tolerating justified harassment. Neither answer is good and frankly depending on the day I would swing towards one of those or the other more. Best, Barkeep49 (talk) 17:13, 1 July 2020 (UTC)[reply]
  • I agree with user:JzG. Carlossuarez46 (talk) 18:04, 1 July 2020 (UTC)[reply]
  • AN/I and many of our current approaches are inherently the wrong way to handle harassment. We ask that someone who feels that they are being harassed step into a public space, announce that they are being harassed in front of everyone with no direct support, require them to notify their harasser that they are making this complaint, and then require them to sit by while their harasser, their harasser's friends, enemies and the general community argue about whether or not it really was harassment, what should be done about it and which person out of the two should be punished. Imagine if we had to do the same thing in a workplace - if we required people being bullied and harrassed to stand in front of all staff and declare that they are being harassed while everyone argues about whether or not it is real. Therefore how we can make the current process at AN/I be less painful? We can't - this is not how we should be handling harassment at all. I think Bilorv explains it perfectly - "If you were designing a system built specifically to protect harassers from accountability, I'd imagine it'd look similar to Wikipedia." Instead, I'd look to how it is handled off-wiki, which as a first step is not always about investigations or passing judgement, but in the first case about providing support and advice. I work in education (not as a counsellor, but harassment cases are still occasionally brought to me if I'm the person someone is the most comfortable with). If a student feels that they are being harassed and approaches me, my job is not to convene a private trial behind closed doors, or to pull them out in front of the class and have them face off with their harasser, or even to act as an investigator. My job is to listen, respect, understand, support and advise. That's what it feels like is missing here. We ask people to go from nothing to public on-wiki declarations of hasrassment, with almost no steps in between. So I'd like a more fundamental change. Rather than looking at how to improve AN/I, or how ArbCom can handle things in private and be transparent at the same time, I'd like to see us make moves on the first stage - how to provide people who feel they are being harassed with support, advice, privacy and respect first, before they end up an AN/I. Which means we might need to develop a whole new process, and may need the WMF working with us, but I think a more fundamental change is better than trying to work out how to make a broken system slightly less broken. - Bilby (talk) 11:23, 2 July 2020 (UTC)[reply]
    • I think this is a good approach: let the aggrieved person consult with a neutral advisor, who can try to guide the person to the best next steps. isaacl (talk) 01:15, 3 July 2020 (UTC)[reply]
    • This is a conclusion I had just come to independently (as far as I can tell), so I agree with Bilby and Isaacl, though I am not sure how this should best be implemented. The problem is that this advisor should be a person who has both the skills and the time do deal with these problems, and it will not be a fun job for people who really just want to write an encyclopedia. On the other hand it is not a job for someone who does not have a fairly solid understanding of the community, the policies of Wikipedia, the aspirations of Wikipedians, and the terms of use. Maybe a combined effort by a professional arbitrator and a reasonably diverse and representative group of people elected by the community, who would be tasked with investigating complaints, assessing their validity, gathering evidence and where appropriate, presenting it to Arbcom for arbitration. The gathering of evidence and advising the complainant can be done privately at this stage, and the complainant can be advised when they are being unreasonable or frivolous, and can at any time bypass this stage and go direct to Arbcom. · · · Peter Southwood (talk): 15:13, 8 July 2020 (UTC)[reply]
  • I think harassment should be a public crime. To my surprise this is a red link... the concept exists here in Portugalin Portuguese and I presume just about everywhere too. I think I just don't know the English expression. I mean a crime, say homicide, which gets investigated and prosecuted by the State, even without the victim's complaint - because, well... homicide victims often can not make a complaint... OK, that is it. Harassment could be investigated after a private complaint, even if not by the victim. Most of the times, if there is harassment, the ArbCom investigation would find more victims. So, then they could open a public case "on behalf of the citizens of the Wikipedia". - Nabla (talk) 19:54, 10 July 2020 (UTC)[reply]

3. Opportunity to respond to allegations

edit

Arbitration policy specifies that for complaints being heard 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." However, it would often be possible for an accused harasser to identify the complainant based on the substance of the allegation.

In cases where a complainant fears possible retaliation, how should the ArbCom balance allowing an alleged harasser the opportunity to respond to allegations while protecting victims' privacy and safety?

Discussion
  • See my responses to the first two questions above. On the one hand, when lives are threatened and one has to write to the arbs, one MUST assume that information will remain strictly confidential. Risker, Casliber or Newyorkbrad can fill in details. We must allow arb discretion to know when information must be private. But separately from that, seeing that we have seen that not even the arbs are immune to preferential treatment of unblockables, there are certainly cases where one should be given a right of defense (and hope the arbs will actually read it). SandyGeorgia (Talk) 14:42, 6 June 2020 (UTC)[reply]
  • No evidence should be secret from the accused. It's unjust to punish an editor on the basis of evidence they haven't had the chance to challenge and reply to. Therefore complainants and witnesses must be willing to testify or else no sanction can be applied.—S Marshall T/C 14:45, 6 June 2020 (UTC)[reply]
  • Perhaps I'm western-centric, but I do believe that a person should be considered innocent until proven guilty. This means if someone is accused of serious policy violation, they should be able to know who is making the claim and have an opportunity to respond. "Private" implies off wiki material, which is easier to forge or fake than on wiki, and not allowing any access is unreasonable. Dennis Brown - 14:59, 6 June 2020 (UTC)[reply]
  • I’m skeptical of any attempt to create due process on Wikipedia. We are not a legal system and there is no inherent right to edit. Actions taken are for the best interest of the project, and sometimes when dealing with sensitive situations you do not want to tell all parties precisely because ArbCom is not a court and can only block someone. It can’t protect them in real life or even on other Wikimedia projects. The goal on Wikipedia should never be procedural fairness but rather substantive fairness. In some cases substantive fairness might give individuals the right to respond. In other cases it absolutely should not.
    We should not be giving someone where there is legitimate concern about real life stalking (such as messages on Facebook, Instagram, etc.) the opportunity to learn that their victim has requested they be blocked and explain the actions. If they are charged criminally, they can do that in a real court. This is not a real court, and no human rights are violated by blocking someone without giving them a chance to respond. These situations should be handled case-by-case but substantive fairness must be the standard, not procedural. A body of volunteers should not be making procedural rights on a website when it comes to dealing with harassment complaints. Sometimes specifics should be given, other times not. It depends and the answer is not absolute. TonyBallioni (talk) 15:13, 6 June 2020 (UTC)[reply]
  • You don’t give someone the right to respond to allegations that they took pictures of someone’s house and sent them to them on Instagram. They could easily be doing that to multiple people and by giving them the right to respond you could put someone’s life at risk. In a court of law, where there are armed guards and someone can go to jail for harming a witness and is at risk for losing their real life liberty, they have a right to confront their accuser. On a private website where the worst penalty is being told to find a new hobby? No, of course we don’t let someone see evidence that could put the life of someone else at risk. We block them and let T&S evaluate what needs to be passed to law enforcement.
    As I said, it’s case-by-case, and sometimes substantive fairness will mean letting someone respond. Other times it means blocking them without comment. In these types of cases it very much depends on the specific facts of the case. TonyBallioni (talk) 15:47, 6 June 2020 (UTC)[reply]
  • Whether to block should be a decision for the authorities. They might want the alleged abuser to be given more rope. In all honesty there aren't many decision points for a volunteer body in a case like that. I'm talking about the cases where Arbcom is rightly the deciding body.—S Marshall T/C 16:22, 6 June 2020 (UTC)[reply]
  • No, the victim calls the police; the arbs still have the issue of blocking or desysopping the editor, and still should not reveal the information publicly. The police rarely act, and the arbs still have to take action on the website particularly when disruption and threats are also happening on the website. SandyGeorgia (Talk) 16:43, 6 June 2020 (UTC)[reply]
  • While I agree that a case-by-case approach is superior to formulating a rigid, absolute answer, we all need to be mindful that the specific case that eventually led to this RfC did not involve real life stalking. Rather, it was a matter of the WMF protecting their own from editorial scrutiny at the expense of a volunteer who was merely seeking to uphold encyclopedic standards. We are not a real court and there will probably be cases in which some evidence needs to be kept private in order to safeguard the well-being of victims, but recent precedent has shown that at least some authoritative bodies cannot be trusted to know when private evidence is not appropriate. Therefore, it may be necessary to establish specific rules that restrict the use of private evidence. LEPRICAVARK (talk) 15:54, 6 June 2020 (UTC)[reply]
  • That is fair. I think it should be principles based rather than formulaic, and I was responding in part to S Marshall and Denis Brown, both of whom gave a fairly US Bill of Rightsy response to something that is more complex precisely because ArbCom isn’t a government. I’d generally support the principle of When there are legitimate concerns about safety, information may not be shared with all parties. In other cases information will be shared as far as is practicable within the confines of the specific case and respecting the privacy and moral rights of all involved. I think that’d deal with the concerns on both sides while also keeping us clear of the “ArbCom as court” mindset. TonyBallioni (talk) 16:02, 6 June 2020 (UTC)[reply]
  • Lepricavark, the recent case is not the only case and we should not hamstring the arbs from dealing with much more serious and clearcut cases than the recent case. In my case, the situation one time was handled perfectly and competently by the arbs and there was no other way it could have or should have been handled. The other time was less expedient and competent, but handled nonetheless based on plenty of public evidence, with other private evidence that merely corroborated what was public. SandyGeorgia (Talk) 16:51, 6 June 2020 (UTC)[reply]
Pls do not forget that "going to the court" is a purely US-based notion. In most other countries, including highly civilized ones, one can not go to the court for this kind of offences. I recently received threats related to my Wikipedia activity as admin, and the police said they could not do anything.--Ymblanter (talk) 17:39, 6 June 2020 (UTC)[reply]
  • I think we need a category of trusted user who can see the allegations and help mount a defense in cases where the accused seeing all of the allegations would out the accuser and/or expose them to further harassment or other dangers. The accused could be allowed to choose from a list of such trusted users someone they also personally trust, and that person would be able to see all evidence and investigate it from the accused's point of view. —valereee (talk) 18:01, 6 June 2020 (UTC)[reply]
  • @Valereee: please take severe mental illness into account, and recognize that there are cases where ANY information revealed does expose the target to further danger. Please let's not set up something that leaves a victim no choice but to vanish. SandyGeorgia (Talk) 18:50, 6 June 2020 (UTC)[reply]
    SandyGeorgia, I'm not following...how would a category of trusted user who could see the evidence so as to help an accused mount their defense cause any more risk to the accuser than having ArbCom see the evidence? —valereee (talk) 18:55, 6 June 2020 (UTC)[reply]
    @Valereee: perhaps I am misunderstanding your proposal (as in, which party is which, and which you are referring to as the "accused")? But I have already said too much here. If the stalker/accused is mentally ill, and needs to be blocked as well as desysopped, then the arbs are involved. SandyGeorgia (Talk) 19:00, 6 June 2020 (UTC)[reply]
    SandyGeorgia, my proposal is if someone who has been accused of harassment (the 'accused') cannot be given for whatever reason the opportunity to see the evidence themselves, we assign an advocate who can review the evidence as an advocate for the accused. These "advocates" should be people the community have decided can be trusted with confidential information. —valereee (talk) 19:10, 6 June 2020 (UTC)[reply]
    @Valereee: OK, I earlier misunderstood you. Yes, that works, except if the stalker is an admin, ArbCom will be involved anyway. And we can't globally designate such people, because when the matter is extremely sensitive, few people will want to talk to a "stranger". In my case, the arbs were also the people I trusted. SandyGeorgia (Talk) 19:13, 6 June 2020 (UTC)[reply]
    @SandyGeorgia and Valereee: Feel free to correct me, but I think that you may still be talking past each other. My understanding of Valereee's suggestion is that it would work something like this:
    1. Arbcom receives a private complaint based on private evidence, and after review determines the complaint is not baseless.
    2. Arbcom appoints someone (probably an oversighter, maybe someone else who agrees to keep everything secret) to represent the accused. That appointee examines the evidence, and presents the accused's case on their behalf.
    3. Arbcom considers the evidence and arguments, and reaches a decision on the merits.
    If this is what Valereee is proposing, I think it would be a good idea, or a minimum, I don't think it would hurt.Reconsidering this --Mdaniels5757 (talk) 19:58, 6 June 2020 (UTC) --Mdaniels5757 (talk) 19:47, 6 June 2020 (UTC)[reply]
    @Mdaniels5757:, if that is the proposal, I would opt out and would have left Wikipedia, because yes, it could hurt. Now, if you alter #2 to be someone both ArbCom and I trust, it could work in general. I'd like to stop commenting on this. @Risker, Casliber, and Newyorkbrad: (who might not want to comment either, but I hope that in general certain voices will be heard wrt certain circumstances that go beyond the current cases, so that we don't hamstring others). SandyGeorgia (Talk) 19:55, 6 June 2020 (UTC)[reply]
    Fair enough. Thank you for your response. --Mdaniels5757 (talk) 19:57, 6 June 2020 (UTC)[reply]
    Mdaniels5757, yes, that's the process I was thinking ADD: but with the accused being able to choose their advocate from a list of trusted users. SandyGeorgia, I see this as a completely optional step for those who want it, not as something that would be forced on anyone. —valereee (talk) 19:59, 6 June 2020 (UTC)[reply]
    SandyGeorgia, my original explanation included The accused could be allowed to choose from a list of such trusted users someone they also personally trust —valereee (talk) 20:02, 6 June 2020 (UTC)[reply]
  • It only matters that the advocate/representative is trusted by the accuser. It doesn't matter if they're a functionary or trusted by Arbcom. The "private evidence" will be the private evidence submitted by the accuser, so if the accuser trusts their advocate/representative with seeing the private evidence, that's all that matters. I think all editors should be allowed to have an advocate/representative, whether it involves private evidence or not. Almost every dispute resolution system in the real world allows people to have advocates/representatives, and for good reason. We should, too. Levivich[dubiousdiscuss] 23:40, 6 June 2020 (UTC)[reply]
  • As for the idea of a trusted user to act as a sort of pubic defender, I agree with what other editors have said about why that won't work well. Instead, I think ArbCom should take seriously the role of looking at both sides, and I think ArbCom already does that well enough that parties do not need to have representatives speaking for them. As for being able to respond to allegations, I'd like for the approach to be to default to giving the accused as much opportunity to do so as practical – except when the situation is one where the accused clearly cannot be trusted. So things like off-site harassment, child welfare issues, threats from mentally ill persons, and threats of harm, are situations where the accused is not entitled to any such access from ArbCom to accusations – and of course, there are things that cross the line into being an actual legal problem, that are above ArbCom's pay grade. When it seems likely that something is happening that is a police matter or a WMF legal matter, ArbCom should be able to block or IBAN someone without otherwise dealing with the problem, as a sort of preventative act somewhat like WP:NLT blocks, until the problem is resolved. But for problems that are less than that, for problems that are basically onsite bad behavior, the accused should have access to the accusations, minus private information (as the harassment policy defines "private information"). The accused may indeed be able to figure out the accuser's on-wiki identity (username), but not the accuser's real life identity unless posted onsite, so long as the accused is not someone who is engaging in those kinds of very serious dangers that I listed just above. If someone is simply being bad on-wiki, and not anything else, they should be able to respond to allegations in that way (and they can be sanctioned if they retaliate onsite). It is offensive to envision that the accused would be left in the dark. --Tryptofish (talk) 22:12, 6 June 2020 (UTC)[reply]
  • Having been on ArbCom myself, I'm certainly aware of cases where revealing information to the one accused would very much put the complainant at risk, so I would not have a blanket rule that it must be done in all cases. However, it should be the default, especially if it's just an accusation of garden-variety misconduct such as inappropriate use of email (and, as above, questions regarding one's on-wiki conduct should always be handled publicly). A good rule of thumb might be that if the matter is serious enough that a reasonable person would consider referral to T&S for law enforcement followup, that's when one should seriously consider whether disclosure is appropriate. But in most cases, the accused should have the opportunity to respond. Seraphimblade Talk to me 02:44, 7 June 2020 (UTC)[reply]
  • Consult with the person making the complaint. They need to balance what evidence they want to make public, which can reveal their identity, and which evidence they don't want to make public, even to the harasser. Of course, if they chose to withhold the evidence to the point the alleged harasser cannot reply to it, such evidence is not really evidence, and should not be a factor, since in the end or has to be able to defend themselves - also from fake accusations. --Piotr Konieczny aka Prokonsul Piotrus| reply here 04:46, 7 June 2020 (UTC)[reply]
  • I think here ArbCom should take their cue from a well-tried system of jurisprudence: the Anglo-American justice system. The alleged harasser should be considered "not guilty" until otherwise proved, and that means that the mere allegation of wrong-doing should not be sufficient to deny them access to the evidence against them. If ArbCom has seen the evidence, and it;s unequivocally strong enough that the alleged harasser appears to ArbCom to be guilty, then ArbCom should just apply a sanction and close the case. Anything in between, and the evidence must be shared (although I would have no objection to specific facts being redacted if they don't appreciably affect the case and are likely to be used against the complainant). Beyond My Ken (talk) 06:15, 7 June 2020 (UTC)[reply]
  • I think arbcom should be very cautious about blindly applying rules and procedures from legal systems to its own procedures because, fundamentally, it is not (and should not be) a court. The goal of arbitration should be to do what is best for the encyclopaedia first and foremost, while being as fair as possible to the individuals. In most cases that will be sharing (a summary of) evidence with accusers, but there will be times when that is neither possible nor desirable. For example if it is clear from a user's twitter profile that they are intentionally significantly disrupting the encyclopaedia, and there is no doubt that twitter user and Wikipedia editor are the same person but no on-wiki link has been made, then that editor needs to be blocked and all they need to know about why is that there is evidence that they are intentionally damaging the project. The same is true if the off-wiki evidence is that they are intentionally harassing one or more editors - they absolutely should not be told who complained or what evidence specifically was, because we don't want them to increase the harassment or tell them how to cover their tracks better in future. These will be rare cases though, and we need to trust the judgement of the arbitrators (who we elected for this purpose) about when it is necessary. Thryduulf (talk) 11:47, 7 June 2020 (UTC)[reply]
  • Look at all the language which is used in relation to ArbCom, and it's patently obvious that it is a court, albeit of a specific type with rules that actually don't protect people accused of misbehavior as well as actual courts do. Looking at this RfC, it's quite clear is that the concern is primarily about protecting and accommodating the person who is alleged to have been harassed, and little or no concern is shown for the accused person. That imbalance needs to be addressed, but I don't see much effort being put into it. Beyond My Ken (talk) 21:28, 7 June 2020 (UTC)[reply]
  • I agree with Seraphimblade: people should be given an opportunity to respond by default but there shouldn't be an absolute right to it and it can be suspended if there's a good reason to do so. Hut 8.5 12:18, 7 June 2020 (UTC)[reply]
  • Arbcom's disgusting and reprehensible behaviour when it actively collaborated with the WMF in not allowing Fram to a)know who had made allegations against him, and b)see the totality of the evidence used against him (collated by a WMF team made up of some of the worst individuals to be involved in harrassment) needs to be explicitly forbidden in future. No you do not get to make allegations against someone without them knowing who their accusers are. No you do not get to submit evidence against someone without them being able to see it and respond to it. These are fundamental rights of every person which the WMF trampled over and Arbcom rolled over and enabled. The problem with private evidence was not that it was kept from the general editing public, but that it was kept from the person accused. Only in death does duty end (talk) 14:29, 7 June 2020 (UTC)[reply]
    Dont want to be called out for denying a person their basic rights under natural justice? Dont take actions that require challenging. Fram was denied the basic rights that murderers and child molesters are accorded when evidence is submitted against them. Arbcoms decision to refuse to show Fram the evidence against him in full, its refusal to name his accusers to him, was a disgusting and despicable act of cowardice and a blatant abuse of basic fairness. You call that a personal attack? As long as you take actions that are unethical and immoral dont complain if people take exception to them. Dont complain in an RFC that is meant to address the sitation, those actions are called out for what they are. Wikipedia is neither a criminal nor a civil court, so there is absolutely no excuse for using procedures that deny the accused the chance for a proper rebuttal that are only used in cases of terrorism to protect intelligence assets. Gratuitous? I could write 3000 words without blinking on the necessity of challenging the methods used in the Fram case, why it needs to be publically aired, and that everyone involved should be ashamed for enabling such despicable actions. The above is the *minimum* I can state without using profanities, or my true thoughts on ther individuals involved in it. Only in death does duty end (talk) 16:30, 7 June 2020 (UTC)[reply]
  • I wouldn't use "disgusting" and "despicable" and "reprehensible", but I do agree with OID that the actions of WMF and ArbCom in not allowing Fram to see the totality of the evidence presented against them was a very poor one, albeit made under stressful circumstances. Simple justice would have allowed it, and there's no excuse for the decision to withhold the evidence. Beyond My Ken (talk) 21:32, 7 June 2020 (UTC)[reply]
  • @Only in death and Beyond My Ken: I would like to point out that decision not to show Fram the full evidence was not ArbCom's. We didn't see the full evidence, and we were contractually obligated, under the non-disclosure agreement that we're required to sign to be on the committee, not to share what the WMF did send us. During and after the case, there was a strong consensus amongst the remaining members of the 2019 committee that that situation was untenable and should never be repeated. However, under ARBPOL, we still have the responsibility to handle private complaints somehow. That conflict is one of the main reasons we organised this RfC. – Joe (talk) 13:48, 8 June 2020 (UTC)[reply]
  • Joe: Unfortunately, that reads a lot like "We negotiated the conditions, so don't blame us for following them." As I said, I know it was a very stressful and trying time for everyone, but, still, don't you, collectively, need to take responsibility for the conditions under which you accepted the case, when those conditions denied to Fram what every accused person in the criminal justice system receives, i.e. the evidence that's been levied against them? In both the Fram case and the recent Bbb23 fracas, there is a distressing unwillingness on the part of the Committee to take responsibility for their actions, and to be transparent with the community about them. Beyond My Ken (talk) 17:49, 8 June 2020 (UTC)[reply]
  • @Beyond My Ken: I will (and have) admit to many mistakes we made with the Fram case. However, even with a year of hindsight, I can't say that I consider accepting the case to be one of them. At the time, it was made clear to us that if we wouldn't take the T&S evidence redacted and under a non-disclosure agreement, there would be no community review of the office action. Not only would that have been even more unfair to Fram—who would now still be banned for what we concluded were unjustified reasons—but it would undermine the strong statement we made in our open letter: that enwiki has the right and capacity to handle harassment complaints internally. Also recall that we only accepted T&S' conditions after obtaining the compromise that we could pass an anonymised summary of the private complaints to him, which I still maintain contained all the salient information in the document to which we ourselves had access. The case we ended up with was, absolutely, procedurally flawed and unfair to Fram, but looking at the bigger picture I think we did the best we could at asserting enwiki's self-governance whilst not totally burning bridges with the WMF (which would have been even more damaging to our self-governance in the long term).
In any case, whilst I see the value in dissecting what went wrong with Fram, the point of this RfC is more to figure out how we can deal with private cases better and avoid it happening again. – Joe (talk) 19:01, 8 June 2020 (UTC)[reply]
Fair enough, and I've already expressed my opinions in the various sections. Beyond My Ken (talk) 20:53, 8 June 2020 (UTC)[reply]
  • Utter self serving bullshit frankly. And this attitude is why everyone who willingly kowtowed to the WMF has my disgust. Arbcom willingly chose to hold a case where it withheld evidence from Fram. Arbcom had the option to say to the WMF 'we won't accept any evidence that can't be shared with the accused'. It chose to knowingly and willingly violate Frams basic right and ability to defend himself, and to try and shift that decision on to the WMF is nonsense. Only in death does duty end (talk) 14:16, 8 June 2020 (UTC)[reply]
    Only in death, dial it down, please. Vilifying ArbCom for negotiating a resolution and then holding up their end of the deal is not a productive criticism, and detracts from the purposes of this RfC. – bradv🍁 14:26, 8 June 2020 (UTC)[reply]
  • You might not think it's productive to call out abusive behaviour by those in positions of authority. I do. But suffice to say, this discussion is unlikely to be productive with people still in positions of authority who think it is acceptable to withhold evidence from those accused. I will save further comments for any re-election or other relevant community processes. Only in death does duty end (talk) 16:04, 8 June 2020 (UTC)[reply]
I find it unusual, if not frightening, that the accused can be denied access to the details of the accusation, because that can lead to anonymous character assassination being the rule. Some onus has to be borne by the accuser to prove their case. The community at large might not have to be apprised of the titillating details, but accusations of wrongdoing can't be completely anonymous. Dhtwiki (talk) 23:41, 8 June 2020 (UTC)[reply]
  • It seems to me there was no issue here until the Foundation screwed things up, and proceeded to twist ARBCOM's arms trying to get themselves out of the crisis they made. Depending how you look at it this response either fails to answer this question, or fully answers it. Alsee (talk) 15:42, 7 June 2020 (UTC)[reply]
  • I take strong exception to the comments of, e.g., S Marshall. In the real world, that the accused have access to the evidence against them in a criminal trial is crucial because of power imbalances: the state has access to police, jails, a professional prosecutor corps, etc., and access to the evidence is part of a package of rights to ensure that this power is not mis-used. In the civil setting, cases are adjudicated by judges who again have access to the power of the state to ensure that all participants engage appropriately. There is simply no analogue of these state powers in the context of Wikipedia self-governance: being blocked or banned are entirely dissimilar from being arrested or jailed or fined, and the community/admins/ArbCom/the WMF have no mechanism to enforce any punishments more severe than banning. Meanwhile harassers have access to a wide array of harmful tools outside of WP, regardless of any blocks or bans. This different power dynamic means that naive translation of the rights of the accused is going to be very problematic. I think the comments of Seraphimblade are very sensible. --JBL (talk) 16:13, 7 June 2020 (UTC)[reply]
  • I no longer trust Arbcom to fairly judge evidence that the subject of the complaint is not given an opportunity to rebut. There have been too many cases of bias among arbs. But the WMF is worse. We need to be able to trust Arbcom, which means arbs setting aside friendships and keeping each other fair, and ideally, I'd like to see a way to impeach Arbcom; the need for it was horribly apparent during the Fram case. The subject of the complaint should be given all possible access to specifics on the evidence behind the accusation. Yngvadottir (talk) 17:02, 7 June 2020 (UTC)[reply]
  • The right of an accused party to hear the allegations against them so as to mount a defense is fundamental. Carrite (talk) 00:08, 8 June 2020 (UTC)[reply]
  • It's worthwhile considering in this "anti-harassment" RfC, that allegations of harassment can themselves be a form of harassment, considering the trouble tha the accused editor could get into. If only for this reason, for the sake of fairness and not judging the accused as guilty before the evidence has been evaluated, both the accuser and the accused should be treated equally, at least until ArbCom has seen the evidence and can make a preliminary evaluation of where the onus lies. It is not a foregone conclusion that every accuser is a straight shooter, and ever accused is a harasser, but that seems to be the assumption that many of these questions -- and some of their answers -- start from. Beyond My Ken (talk) 04:00, 8 June 2020 (UTC)[reply]
  • I'd say there is very little space in the venn diagram where non-showing of evidence is permissive, but it would still be within ARBCOM's remit. Obviously threats to life/limb are a T&S issue. One possible area is if there would be an appreciable chance (that is, both possibility and risk) of the accused using the evidence to reach out to the accuser's employer of similar. However, outside of niche cases like that, I believe the accused must have the right to see any evidence against them. I do not believe the "trusted defender" route (or ARBCOM handling both sides itself) is sufficient for the following reason: the accused might well have information they don't know to provide without being asked fairly explicity - enough to render hidden evidence pointless. So a defender wouldn't have all the information, so they aren't a sufficient substitution. They might be worthwhile in their own right for reducing "arbcom process advantage" issues, though. Willfully choosing to be less fair than we could is still immoral, even when the consequences are reduced. Nosebagbear (talk) 09:56, 8 June 2020 (UTC)[reply]
  • I think that giving the alleged harasser an opportunity to respond is an inevitable requirement of any fair process. But since we're a website and not a legal system, we should allow ArbCom to make exceptions when there are reasonable grounds to believe that exposing the complainant's identity could result in actual harm to the complainant (physical or reputational, such as calls to employers, etc.). Otherwise I agree with Piotrus: the complainant should be allowed to determine how much information to make available to the other party, with the caveat that unrevealed evidence may not be grounds for action. Sandstein 16:51, 9 June 2020 (UTC)[reply]
  • We are not a courtroom. The substantive difference is that we do not have the power to enforce our decisions. While a person who commits a serious crime gets a fair right to respond to claims because they are locked up if found guilty and (supposedly) will not be able to harass the people who they find out provided information, Wikipedia does not have anything of the sort. Off-wiki harassment and doxxing can seriously damage a person's health, and sockpuppetry is so widespread that we don't really have a way to even stop banned users from continuing to edit. A person who provides information to ArbCom should be in control of whether that information is provided to the accused. ArbCom can choose to weight such evidence less or even discount it if they believe it cannot be fairly considered without the accused replying and providing their side. We already have such a widespread issue with people not coming forward and providing evidence (generally based on justified true beliefs that they will be harassed and abused) that we need to change the system to encourage evidence to be provided first and foremost. I agree with comments made by TonyBallioni and JBL. — Bilorv (Black Lives Matter) 10:00, 10 June 2020 (UTC)[reply]
  • The Police and judiciary are not everyone's role models at present. But one thing they can get right is the ability to investigate based on a tip off. That anonymous tip off need not come to court, but it could inspire an investigation that does. Equally we should be clear with our anonymous complainants, if you aren't prepared to publicly complain then the evidence you give an admin, checkuser or other functionary needs to hold up just by looking at the difs provided. Sometimes those who aren't prepared to publicly "press charges" have to accept that that leaves others with insufficient evidence to act. I appreciate that this can leave people uncertain whether their complaint will be dealt with fairly if they complain. But the solution to concerns about fairness can't be to be deliberately unfair. Has any society or organisation ever done less harm than good by agreeing to act on anonymous denunciations that can't be shown to the accused? ϢereSpielChequers 18:11, 10 June 2020 (UTC)[reply]
  • I get the arguments everyone is making for absolute due process, but I most agree with TonyBallioni and Bilorv who point out why pretending to be a government just isn't smart. The reason courts and governments can guarantee the right to face your accuser is because the state has a monopoly on legal violence, and they will use violence to protect the victim if the proceedings place them in danger. We have no such ability. If we cannot make the same guarantee of safety for victims as a government, we cannot make the same guarantees of process for the accused as a government. Consider someone reported for off-wiki stalking; if the victim reported this to ArbCom and ArbCom was forced to either provide all evidence to the stalker, what are we to do if the stalker decides to up their aggression in retaliation? A government could place guards at the victim's house pending resolution, and imprison the stalker to prevent retaliation. We cannot, and so it is incredibly irresponsible to pretend that we have a duty to guarantee due process for the accused but no duty to guarantee safety to the accuser. One without the other is a recipe for protecting abusers as we have seen. We must move beyond thinking of Wikipedia and its processes as a government and trial courts. We're not a government; ArbCom is not our Supreme Court; we're an encyclopedia; Arbitrators are here to help us get on building an encyclopedia without having to deal with painful people, not help us build a functioning city-state. Wug·a·po·des 23:45, 12 June 2020 (UTC)[reply]
@Wugapodes: But their job is to remove painful people, not to remove those who might not, or would not, were more transparency and fairness in place. Very few, if any, of the participants above advise an absolutionist approach, with all information shared, in all circumstances. But being blocked when evidence might indicate otherwise is its own form of abuse. Insufficient sharing of information limits not merely a defence an individual might make on the information provided at that point, but also seals off evidence arbs may not be aware of. Additionally, any reduction in fairness on the site does damage well in excess of the loss of those incorrectly blocked. Nosebagbear (talk) 14:47, 13 June 2020 (UTC)[reply]
@Nosebagbear: I don't think I understand what you mean by the first sentence? But beyond that, I agree that many of the comments above are well considered, but I disagree that they correctly appeal to due process as a value in this consideration because we cannot achieve due process—doing so requires us making guarantees we simply cannot make, and we must work within that reality. States can guarantee high levels of fairness in proceedings because they can violently suppress those who threaten a fair hearing for both sides. We do not have that ability, and as a result pursuing "fairness" without regard for all aspects of due process actually makes the system less fair. If we cannot guarantee the safety of reporters, we are less likely to get reports and so abusive editors are more likely to continue their behavior; how is that fair? If we get fewer reports, the few we do get only show part of the picture so abusive editors are less likely to be sanctioned; how is that fair? If we cannot guarantee freedom from reprisal, we likely will receive less evidence and so sanctions will be weaker if they occur at all; how is that fair? If there are sanctions and we provide enough evidence to identify a reporter, we cannot guarantee safety from reprisal by the sanctioned editor or their allies; how is that fair? These are all protections governments can guarantee through imprisonment, orders enforced by armed agents of the state, or relocation and protection of witnesses following trial. We can offer none of these, and it is unfair to guarantee a level of protection to abusive editors that we cannot guarantee to reporters.
Due process guarantees fairness for both sides, and focusing on whether we're being fair to the accused without due regard for whether we're being fair to reporters subverts the principle of due process which we're supposedly trying to uphold. Even governmental systems acknowledge that in some instances, due process requires us to accept some false positives in order to guarantee fair outcomes (in aggregate) for both sides of a dispute. This is why civil trials and administrative hearings have evidentiary standards other than reasonable doubt such as clear and convincing or preponderance of evidence. For the sake of argument, even if we can agree that this constitutes "reduction in fairness", I still fundamentally disagree with your conclusion for three reasons: (1) you have provided no reason to believe that we risk a spate of unjust blocks unless we give unfettered access to evidence—in fact the case that brings us here is an example of the exact opposite: an unjust block was removed despite limited chance to respond to evidence; (2) the WMF has already shown a willingness to intervene if we fail to design a process that adequately handles harassment and abuse, and the harms of the WMF having to step in should this supposedly fair system continue to protect abusive editors are significantly greater than the harms you insinuate; and (3) we are already experiencing significant harm by protecting abusive editors who scare potential new editors away from the project, and any benefit we gain from protecting abusive editors is outweighed by the continuing harm we have in editor recruitment and retention.
If we want to make progress on this problem and retain community control, we need to go deeper than surface-level arguments about what due process means in criminal procedure and consider our actual material position. We do not have the means to act like a government; and the issues at stake are far less than at any judicial proceeding. Being fair to everyone requires us to acknowledge and act on that basic fact. We, intentionally or unintentionally, protect from sanctions editors whose behavior would be unacceptable in any other workplace; yet we pretend that the harms of being kicked from a volunteer encyclopedia organization are as drastice as imprisonment or loss of income. I simply cannot buy arguments about what constitutes fairness which proceed from that false premise. Wug·a·po·des 22:01, 13 June 2020 (UTC)[reply]
  • TonyBallioni, JBL and Bilorv have expressed the same views I have about this, —PaleoNeonate06:29, 13 June 2020 (UTC)[reply]
  • I agree with TonyBallioni and in particular with Wugapodes' articulation of fairness—they said it better than I could have. buidhe 02:43, 14 June 2020 (UTC)[reply]
  • These types of cases should be handled more along the line of a whistleblower who needs protection. We're dealing with anonymity so why all of a sudden is it important to reveal identities, especially when someone feels threatened and is being harassed to the point they are no longer able to enjoy their editing WP? Any arb who cannot make a decision based on solid evidence (diffs), doesn't need to be on the committee. The defendent is entitled to some information but not any of the details that would out the victim. For example: A complaint of harassment has been filed against you by another editor. ArbCom is currently reviewing the evidence. We may be asking you questions during this phase of the process....yada yada yada. Atsme Talk 📧 04:24, 15 June 2020 (UTC)[reply]
  • Banning the accuser from seeing evidence against them is Kafkaesque and means that ArbCom is going to be much more of an inquisitorial process than an adversarial process. However, we need to consider that ArbCom isn't a court of law and doesn't have numerous powers a court of law does. If ArbCom indefinitely blocks an abuser from Wikipedia, they have no power to prevent the abuser from expanding their harassment or retaliating against the accuser, nor do they have the power to "gag" someone from outing their victim or private evidence. A real court has the ability to create and enforce restraining orders with actual jail time. It will sadly be necessary to impose secretive measures in some cases.
About the only potential solution to this kind of situation is to force accused harassers to sign a version of the access to non-public information policy with their name and physical location so WMF-legal or victims can take action against people that abuse their access to the evidence against them. I expect this solution to be untenable though as it would add significant administrative burdens to the WMF and the Wikimedia community does not in general appreciate legal solutions to on-wiki problems; as well as the practical difficulties in drafting or enforcing some sort of standard for the accused in this quasi-legal process. Chess (talk) (please use {{ping|Chess}} on reply) 08:46, 19 June 2020 (UTC)[reply]
  • If we want any claim to fairness, the accused needs to be able to respond to the allegations against them. They may not need to know the identity of the accuser, but they need to be able to respond to the facts of the accusation. While it's possible that other concerns might be able to outweigh this, that would have to be an extremely high bar. And just like liberty versus safety, if we want to claim safety over fairness we're likely to deserve and get neither.
    I find the arguments above that we don't need fairness because a ban from Wikipedia isn't depriving anyone of life or liberty unconvincing. The need for fair dealing with others isn't limited to cases with extreme consequences. I also find the arguments that we can't provide an accuser with physical or off-site protection unconvincing. It's unfortunate, but just because we can't guarantee the complete safety of an accuser under even the unlikeliest of circumstances isn't a good excuse for going to the opposite extreme. Real-world police exist for the extreme cases, and if a potential accuser is that fearful they should go to said police. That said, the ideas above about parallel construction to bring an accusation separated from the original accuser could help to focus the allegations on the relevant facts rather than the identity of the accuser.
    I also don't buy the argument that WMF will force something if we don't adopt biased policies. Sure, they might try, but if they want to take a turn to the left where Conservapedia took one to the right then that's on them and m:Right to fork is an option for the rest of us if other methods of protest don't suffice. Anomie 16:58, 21 June 2020 (UTC)[reply]
    The police often disregard serious felonies, so I'm rather dubious as to why they would take an interest in any of the internet harassment that is within their jurisdiction, or how they could improve the situation if they did. — Bilorv (Black Lives Matter) 22:19, 22 June 2020 (UTC)[reply]
    I'm very dubious as to how that means we should disregard any attempt at fairness in our own practices. Anomie 00:25, 23 June 2020 (UTC)[reply]
    This is a pithy response, but do you think anyone genuinely holds a belief and also thinks that belief to be unfair? — Bilorv (Black Lives Matter) 07:00, 23 June 2020 (UTC)[reply]
    I think some people believe that the ends they promote are so "good" that they outweigh basic fairness to all parties involved. Anomie 12:35, 23 June 2020 (UTC)[reply]
  • I generally agree with TonyBallioni above. I'm less concerned with fairness and more concerned with safety and maintaining a good collaborative editing environment. If there are rare incorrect blocks/bans (caused by a lack of opportunity to respond), that is a reasonable price to pay for the protection of other editors more broadly. Calliopejen1 (talk) 21:19, 22 June 2020 (UTC)[reply]
  • Allow a response, and if the response is inappropriate, take action. I have been the victim of harassment in the past. I've had to get a Norwich Pharmacal order to find out the identity of someone who was posting death threats against me and my children, there was an arrest but no prosecution in the end. These were not credible threats, but they were sinister and they were in the context of intimidating cyclists on the road, which can easily lead to serious injury or death even if the threat is intended to be hyperbole. It was above the level of "I KILL U BRO!" but below the level of a clearly articulated and malicious plan: the perpetrator was an angry idiot, basically. What I learned was that (a) these things are best dealt with in the open, because most bullies are in the end also cowards, and (b) the help and support of a community of friends is desperately important. I am not a fan of Wiki-lawyering, but I think any case like this that gets taken, there should be guidance for those on both sides on de-escalation, mental health, and peer support, and we should really be trying to boost the profile and resourcing of mediation. The less adversarial the process is, the better it will feel for those involved. Our mechanisms for collaborative resolution have become markedly less effective over time, for a lot of reasons not all under our control. For a community that is supposed to be here to write an encyclopaedia, there is an ironic lack of nuance in our meta discussions - sometimes people strive for brevity but lose clarity in the process, sometimes we just write clumsily or in haste, and often we read to reply rather than to comprehend. There are some who are good (Newyorkbrad, as a lawyer, is a stand-out here) and some who can be shockingly bad (including me). Word limits do not necessarily help with this, though they serve a useful purpose in other ways. Guy (help!) 15:25, 23 June 2020 (UTC)[reply]
    This is an excellent response, which I wholeheartedly endorse. JzG, I'm sorry you had to go through that; nobody should. Thank you for talking about it.
    Mediation is incredibly important as a first step in resolving disputes, and since the demise of the Mediation Cabal and RFC/U, there's not really been any way of discussing conduct issues other than arbitration and ANI - two environments which are incredibly hostile for all involved, by their very nature. Creating separate avenues for mediation not only allows more people to engage with the processes, but it also allows it to be done in a transparent, collaborative manner - hopefully one that focuses on mutual improvement, rather than punishment. Naypta ☺ | ✉ talk page | 15:31, 23 June 2020 (UTC)[reply]
    Naypta, RFC/U was adversarial, too, IMO. MEDCAB was a genuinely good idea and I mourn its passing, we should try to rebuild it but with tangible support from WMF in the form of training. Guy (help!) 16:45, 23 June 2020 (UTC)[reply]
    @JzG and Naypta: - the WMF phrased training in the strategic recommendations (plus the not particularly clear minimal engagement from WG members in the consultation) has been somewhat...unclear with a shade of odd compulsoriness. However, training for MEDCAB members in mediation might well be helpful - I'm positive that someone has worked out mediation training focused towards online usage (rather than say, company contract mediation). That could well be of use. Of course, what killed MEDCAB wasn't the competence of those involved but the unhelpful rule nature - I certainly think there's interest in bringing it back, with a lower barrier for usage. Nosebagbear (talk) 21:50, 28 June 2020 (UTC)[reply]
    @Nosebagbear: You make a good point about this being a potential area for collaboration between enwiki locally and the WMF - I think a version of MEDCAB brought back would actually work quite nicely alongside the UCOC that's being introduced. Mediation-trained volunteers could step in to resolve disputes and to de-escalate problematic situations, and could also perhaps take on part of the role of civility enforcement where mediation fails and/or is rejected, in line with the UCOC, and with WP:CIVILITY etc. Naypta ☺ | ✉ talk page | 21:59, 28 June 2020 (UTC)[reply]
    @Naypta: - I'm not sure you'd want mediation to be enforced, mediation with even the appearance of compulsion becomes vastly less effective (something that has real data to back it up generally, and I'd believe it to be the case here), indeed to get the best requires real openness, which is why mediation evidence is exempt from being considered in ARBCOM cases unless the mediators think it was being engaged in bad faith. I'd also see real INVOLVED issues. Nosebagbear (talk) 22:06, 28 June 2020 (UTC)[reply]
    @Nosebagbear: That's fair; I'm just spitballing ideas   Having some sort of informal mediation for disputes between editors that aren't 3O/DR compatible (often because they're not content disputes) I think is helpful for sure, either way. Whether it be through the mediation route or through something else, though, I do think we ought to have some system of dealing with civility complaints that's not just "take them to ANI" - the more stuff we lump into the category of "take it to ANI", the more overworked, tired and fed up everyone there gets; the less people feel their concerns are treated seriously; and the less actually gets done. Naypta ☺ | ✉ talk page | 22:18, 28 June 2020 (UTC)[reply]
    @Naypta: that's certainly fair. A separate MEDCAB discussion is probably needed to discuss: what went wrong, what are possible fixes, would the actual process (as opposed to the triggering process) be the same etc, before any formal consideration on re-creation. I absolutely agree that MEDCAB2 would need to cover both content and conduct cases (particularly since the worst cases usually involve both, after all, content is what we do) Nosebagbear (talk) 22:30, 28 June 2020 (UTC)[reply]
    While mediation committee cases were held in private and so its discussions were not used in evidence for arbitration committee cases, I don't believe there was any restriction on the evidence discussed in mediation cabal cases. The circle that needs squaring for any type of mediation is that it works when all parties are earnestly working towards a common solution. Unfortunately, there isn't a lot of incentive for recalcitrant editors to engage, given English Wikipedia's current consensus-like decision-making process. (In the real world, there is often a direct financial cost to the parties not reaching an agreement, which provides an incentive to participate.) There are some discussions that benefit from someone guiding the thread of conversation, and some editors try to help keep discussions focused and progressing forward. Unfortunately all too often the participants aren't willing to be co-operative and place a greater premium on winning the discussion instead of finding a true consensus that most people can live with. We should structure the decision-making process to encourage co-operation and looking for the best outcome for the project as a whole. isaacl (talk) 23:06, 28 June 2020 (UTC)[reply]
    Naypta, this feels right to me. Guy (help!) 23:41, 28 June 2020 (UTC)[reply]
  • JzG touched on an idea I've toyed with, & if I could trust the Foundation to act thoughtfully, sensitively & without the same favoritism that reappears here on the Wiki (& I've seen on other online fora), I'd seriously suggest it. Simply put, the function of T&S in situations of alleged harassment by established members would be to do a soft intervention. By this I mean that a professional mediator would examine the evidence, then reach out to the person & try to explain to the accused that there is a problem & how the person should modify their behavior. By this, I don't mean a "one-&-done" exchange, but a series of meetings without any leveling of blame. If the professional can't effect a reduction in harassment, then that person takes the case to the ArbCom where, although the case is made public & anyone can offer input, the mediator's evidence is private.
    This has a number of advantages. One is that it avoids the issue that FRAMGATE had -- someone gets banned out of the blue & everyone gets upset. (I'll note here that the vast majority disputes end up at ArbCom after a lot of storm & flames beforehand: the ArbCom can make a bad call in that situation, & it won't create a backlash half as fierce as what happened in FRAMGATE. Another advantage is by separating the intervention from the issuing of sanctions, if there is a case of false accusations of harassment -- which do happen -- there will be at least two stages where this can be reviewed. A third is that not only can the private information be kept private, the person reporting can be kept anonymous. By having the mediator file the case with ArbCom, there will be a firewall between the original reporter, the public, & the responding party. A fourth advantage -- & I'm speaking from my experience as a member of the first Mediation Committee back in the Stone Ages of Wikipedia -- by having a professional involved at some point, we benefit from experience instead of struggling with naivete. Right now, any conflict serious enough to come before ArbCom has generated tens of thousands of words, most of which aren't relevant, then try to figure out the best solution -- all in their spare time. And we are here to create an encyclopedia, not to be experts in handling interpersonal conflicts. (I suspect many of us wouldn't be here if we had better social skills in the first place.) An expert would be able to sift thru all that & look for key words or patterns to figure out what is going on much more quickly & more accurately. Maybe the accusation of harassment is simple misunderstanding; maybe it's unintended insensitivity; maybe the person accused is simply a jerk who needs to be removed from Wikipedia.
    Of course to do something like this means diverting funds from pet projects like renaming the Foundation, & making an effort to find someone with experience in handling online disputes successfully -- &, no snark intended, it may not be easy to find such a hire. (I doubt any social networking platform has anyone currently employed with that kind of experience -- such a person would be seen as an overhead cost, at best.) But speaking from my experience both here & on other online fora, this may be the most effective way to handle this problem of conflict & be fair to all parties involved. -- llywrch (talk) 22:42, 29 June 2020 (UTC)[reply]
    I would suggest that a hired person would bring in professional expertise, but will probably be severely lacking in local knowledge, so a professional on their own would probably be as prone to error as our usual experience with WMF intervention, which would be a bit counterproductive at best. This could be mitigated by having some trusted members of the community elected as representatives/advisors to make sure that the proceedings don't go instantly down the tubes. Such community representatives would have to be trusted by the complainant, so there should be a few to choose from, and they might have to recuse if involved, though not necessarily. · · · Peter Southwood (talk): 16:11, 8 July 2020 (UTC)[reply]
  • I agree with Tony that there can and should be limits to the amount of DUE PROCESS we afford. What a state can inflict on a person needs greater safeguards than a private website. I also agree with Guy/Llywrch in terms of skilled, formal, mediation being necessary. Whether that's through a WMF contractor/employee or through trained Wikipedians is too far in the weeds for me to decide here. But I support the principle - though I will note that in the mediation I've been trained on, discussion with the two parties in the same room is not the first step. Instead one or more discussions, privately, has already occurred prior to bringing them together. Could this be done effectively onwiki? I would hope so but that hope doesn't mean it's what it should happen. If there are RS about how to do effective online mediation I'd love if someone could point me towards them so I could read and form a more intelligent opinion on public/private rather than merely hoping we could be public. Best, Barkeep49 (talk) 17:42, 1 July 2020 (UTC)[reply]
  • WP is not a bureaucracy and ArbCom is not a court. No one has a right to edit here; everyone can be kicked out for any reason or none at all. There are not rules of evidence. But for a community to thrive, volunteer members should be accorded the dignity of not being kicked out without a valid reason that they have been able to address head-on. For the few instances that have been hypothesized above where someone may be at risk of physical harm, I posit that ArbCom cannot do anything really to prevent that. We have police forces for that. I wonder whether we want editing privileges contingent on one's off-wiki conduct? What off-wiki conduct could we begin to police? If harassment is bad, what about convicted felons? We don't inquire into that and probably ought not - if someone is contributing to the project, they get to stay is the norm. If we allow ourselves to remove from our allowed editors those whose off-wiki behavior we disapprove of, we should (a) state what behavior qualifies for removal, and (b) give a person accused of such behavior the opportunity to respond to the evidence. We don't have to, but it's the right thing to do. Carlossuarez46 (talk) 22:13, 1 July 2020 (UTC)[reply]
  • Agree with S. Marshall. TomStar81 (Talk) 10:27, 3 July 2020 (UTC)[reply]
  • Of course the accused should be given a chance to respond. --GRuban (talk) 15:21, 8 July 2020 (UTC)[reply]
  • Allowing the accused access to details of the evidence against them is desirable when the risk is acceptable. This is the general case in a court of law where they know who the accused is, which allows the associated authorities to follow up when necessary. On Wikipedia the accused is often anonymous. Providing information to an unidentified/anonymous person may increase risk to the complainant. If the accused is willing to positively identify themselves to Arbcom it may reduce the risk to the complainant. This could be considered as an option in some cases, but remaining anonymous must remain an option. · · · Peter Southwood (talk): 16:27, 8 July 2020 (UTC)[reply]

4. Unsubstantiated complaints

edit

In some cases an editor reports that they are being harassed, and upon investigation, the conduct in question is arguably unpleasant or rude, but is not a violation of the harassment policy. Occasionally, this is the result of a deliberately false complaint, but most often the complainant genuinely feels victimized, and is even more distressed after learning the result of the investigation. We want people to feel comfortable contributing to Wikipedia even if they make mistakes; likewise we want people to feel comfortable correcting mistakes even if the person who made them reacts poorly.

Is there anything the community or ArbCom can do to interrupt this unpleasant dynamic? Are civility warnings appropriate for editors who are making legitimate criticisms at a frequency or in a style that the subject finds too unpleasant to deal with effectively?

Discussion
  • No, "civility warnings are not appropriate for editors who are making legitimate criticisms at a frequency or in a style that the subject finds too unpleasant to deal with effectively". This is too easily gamed, as we have seen recently. Those who don't want any criticism of policy violations-- even those that comment on the content, not the contributor-- are too often prone to cry "civility", and too often unwilling to address the problems with their contributions. But yes, there is something ArbCom can do: actually read the "real" evidence in arbcases. SandyGeorgia (Talk) 14:46, 6 June 2020 (UTC)[reply]
  • If someone's making a mistake we need to be able to tell them so, and sometimes it's necessary to say it clearly and frankly. If the subject finds that unpleasant, then they may not be well-suited to Wikipedia. Compared to the rest of the internet Wikipedia is hardly a bear pit.—S Marshall T/C 14:54, 6 June 2020 (UTC)[reply]
    @S Marshall: I don't think there's a question at all about legitimate concerns with regards to someone's conduct or mistakes, but rather the way in which those concerns are conveyed. It's inappropriate and unhelpful to try and convey concerns to someone by means of a barrage of expletives - which wouldn't come under our current harassment policy. That's fair enough, because it's not really "harassment" per se - but it is incivility, and that's a problem. Naypta ☺ | ✉ talk page | 11:35, 9 June 2020 (UTC)[reply]
    I totally understand why some editors swear at others, and imo it's frequently the target, rather than the swearer, who's displaying the more problematic behaviour. Swearing isn't OK. But we need to see it as a symptom of stress.—S Marshall T/C 13:13, 9 June 2020 (UTC)[reply]
    @S Marshall: I can understand that, and sure, it might well be a symptom of stress. However, this is an encyclopedia; we don't let vandals off because their vandalism is a symptom of their stress, youth, sense of humour or any other factor - to do so would be rather absurd. If someone is not capable of channelling their stress into anything other than insulting other editors, then they should be warned over it - and eventually, if they continue to do so, should be civility blocked. Naypta ☺ | ✉ talk page | 10:20, 10 June 2020 (UTC)[reply]
    Right now, the simplest way to win a content dispute is to provoke the other bloke until he does something you can report on AN/I. Novel rules about civility blocks would make that problem worse. It's also cultural bias against Australians, and in favour of Canadians.—S Marshall T/C 10:52, 10 June 2020 (UTC)[reply]
    @S Marshall: I disagree that that's the way it is; I would, however, argue that that's the way that it should be. It should be an awful lot easier to block someone for incivility than anything else. People should not be "provokable" in the first instance to start attacking and insulting other editors. If an editor does something that you think is wrong, standard process should be followed; discuss it with them, if you fail to reach a conclusion bring it to dispute resolution or to ANI. Nowhere in that process should "provocation" come into it - and if it does, that in and of itself is incivility and should be subject to the same set of rules. And, much as I assume you're joking about Canadians and Australians, as a Brit, I have faith that everyone can be brought to the table on this. Nobody, from any country or cultural background, wants to be treated poorly; that's why the Golden Rule is so globally recognised. Naypta ☺ | ✉ talk page | 11:51, 10 June 2020 (UTC)[reply]
  • Wikipedia is not therapy. If someone is overreacting to an event, there isn't anything we can do except explain that to them. We aren't here to be the civility police, or to help overly sensitive people cope. Not everyone can contribute to a collaborative project comfortably, and while we try hard to accommodate and compromise as much as possible (including voluntary interaction bans), some will just not fit in. It is impossible to make Wikipedia a perfect environment for everyone, and trying to do so will just run off more people than it accommodates. Dennis Brown - 15:04, 6 June 2020 (UTC)[reply]
  • Editing is a rough business, and one needs to learn to not take it too personally when someone else kills your darling. Frank and direct assessments of other editors’ contributions are vital to the improvement of the encyclopedia. New editors may need to learn to detach their emotional response to having their work edited. Rudeness that goes beyond “frank and direct” on behalf of revising editors is counterproductive to that goal. I don’t like the phrase “civility warning”; if the conduct isn’t sanctionable, what are you warning them of? But if ArbCom’s conclusion is that someone was unproductively rude but not harassing, by all means say so. Frank and direct assessments of contributions go both ways.--Trystan (talk) 15:15, 6 June 2020 (UTC)[reply]
  • Being the civility police has never worked out well for either ArbCom or individual administrators. Let the noticeboards handle these issues. TonyBallioni (talk) 15:17, 6 June 2020 (UTC)[reply]
  • Personal attacks should be met with warnings and/or blocks, depending on the specifics of the situation. However, editors often fail to distinguish between personal attacks and criticism of content. If any editor finds that they cannot handle having their work scrutinized even when that scrutiny is within the bounds of NPA, then perhaps this editor needs to find a different hobby. LEPRICAVARK (talk) 16:01, 6 June 2020 (UTC)[reply]
  • If someone has a pattern of being unpleasant or rude, without good reason, then that goes against various conduct policies. I know the community is generally terrible at enforcing these but that's a bug, not a feature. Hut 8.5 16:41, 6 June 2020 (UTC)[reply]
  • Yes, false claims of "harassment" made by unblockables needs addressing. I was, for example, blocked for (nearly) two years on the basis of the sockpuppet User:Sagecandor claiming that I was "hounding" them. Though I have repeatedly asked for that "block record" to be amended, it has not been. More recent examples of people using that tactic may be explored someday. (I have in mind the removal of a sanction (for another unblockable) by dei ex machina, and the fact that I was "convicted" in one area before being allowed to present a defense. It may be that some unblockables do more good than harm. However, my experience shows that pointing out the harm is dangerous if you wish to continue editing on certain subject matter. It appears that pointing to someone's timecard is especially frowned upon. It would be interesting to have comment from ArbCom on whether pointing out, using data from WMF tools, that 1) some people have dominated TP discussion about a topic, and/or that 2) some people are on the project full-time, and/or that 3) coordinated activity / tag-team tactics are being used; constitute "harassment". In general, the tactic at AE (which is an extension of ArbCom in principle at least), seems to be block or ban first, and read the evidence later, if at all (when it concerns defense against "unblockables" using their social capital to bring -- mostly or even "slightly" ^^ -- frivolous cases). -- SashiRolls 🌿 · 🍥 17:01, 6 June 2020 (UTC)[reply]
  • If we aren't willing to call out rude and unpleasant behavior, how will we recognize a pattern of behavior that may be pushing editors away? If you're being complained about by multiple editors for your style of providing correction, you're doing something wrong, but we'll never realize that Editor X is doing that if we're just saying, "Oh, well. Editing's a rough game. Not for sissies." —valereee (talk) 18:09, 6 June 2020 (UTC)[reply]
  • In the US, there has been a story of "birding while black" that has been prominent in the news lately, that is an illustrative case of an unsubstantiated complaint. And on-wiki, it's pretty common for users to allege a personal attack or assumption of bad faith when nothing of the sort has actually occurred. (And I've noticed that such users are sometimes quite casual about making personal attacks or the like, themselves.) If something is unsubstantiated, then it's unsubstantiated, and that's that. The accuser should have the decision explained to them in a considerate and emphatic way, but if they cannot accept it, that's their problem. But the threshold should not be that it has to be full-throated harassment. If the accused is being chronically unpleasant, but not really doing what they were accused of, it's appropriate to issue a commensurate warning or advice. --Tryptofish (talk) 22:25, 6 June 2020 (UTC)[reply]
  • I've said before that both experienced and new editors need to show understanding. The expectations on experienced editors has been much discussed. New editors need to appreciate that the Wikipedia community is imperfect, and is certainly going to respond in less-than-optimal ways at times. We should strive to be as welcoming as possible, even when providing negative feedback, but the nature of any community is that they'll be conflicts, and people often aren't going to respond to you in the way you think they should.
    Noticeboards aren't a great way to provide feedback to editors set in their ways. Even respected editors have trouble with providing a quiet word of feedback in these situations in a way that will produce in a positive result. It's just all the much harder for a large group of editors of varying levels of interpersonal skills to do so. To defuse behavioural problems, we need content dispute resolution procedures that can deal with disputes more effectively, thereby reducing the incentive for poor behaviour such as unsubstantiated complaints. isaacl (talk) 23:59, 6 June 2020 (UTC)[reply]
  • The answer is in the question itself: legitimate criticisms. If someone is legitimately criticizing something which you are doing wrong, the solution is to start doing it right, not to complain that you keep getting caught or seek sanctions against the one who does so. The solution is to tell the complainant both that criticism is par for the course when one participates in a collaborative endeavor and that a thick enough skin to hear it is necessary, and that if you want to stop getting legitimate criticism about something you actually shouldn't be doing, stop doing that! Seraphimblade Talk to me 02:48, 7 June 2020 (UTC)[reply]
  • Sigh. WP:PAIN was closed years ago, we don't really have a way to enforce civility. Some people are civil, some are much less so, and the line is too blurry. Which is why in real world you don't get sentenced for being uncivil. There may be some laws we could adapt and issue warning based on clearly identifiable transgressions such as saying 'fuck', but this would be rather... amusing, to say the least. Frankly, I'd rather see a strong enforcement of the rule 'don't discuss other editors outside AN(I)/AE or relevant boards'. That might be more useful. But again, we would need to consider why PAIN failed. --Piotr Konieczny aka Prokonsul Piotrus| reply here 04:50, 7 June 2020 (UTC)[reply]
  • "Are civility warnings appropriate for editors who are making legitimate criticisms at a frequency or in a style that the subject finds too unpleasant to deal with effectively?" No, civility warnings should only be issued for real, strong instances of incivility, not for unpleasantness, or sarcasm, or rudeness, or scatology not directed at people, or simply to make another editor feel better. Wikipedia exists to help create and improve an encyclopedia it's not a therapeutic community or a sociological "safe place" or your mother's bosom. As long as editors are improving the encyclopedia, and generally behaving themselves within tolerable limits, then they should be allowed to go about their business without being hassled. Harassment cases should be about real, actual harassment, and not about someone's bruised feelings because they're not being treated like the royalty they believe themselves to be. Beyond My Ken (talk) 06:24, 7 June 2020 (UTC)[reply]
  • Editors behaving as rudely as the rules allow shouldn’t be subjected to harassment complaints, but they also shouldn’t be given validation of that behaviour, because it is detrimental to the encyclopedia. They should be given constructive feedback on how they could improve their on-wiki interactions.--Trystan (talk) 16:46, 7 June 2020 (UTC)[reply]
Part of that problem is identifying the level of rudeness allowed by the rules.· · · Peter Southwood (talk): 12:25, 9 July 2020 (UTC)[reply]
  • There are multiple aspects to this question, and while I generally agree with much of the above it is not always going to be as black and white as some are painting it. If editor a finds editor b uncivil, it might be because editor a is being uncivil, it might be because editor b is being overly sensitive about their work, or it might be a bit of both. For example, if editor a is giving other users templated warnings faster than they can be responded to or following responses of "I don't understand" with just another template then then both editors need to be told to adjust their behaviour in a manner that they understand but editor a's problem is not one that civility warnings will likely help with. Thryduulf (talk) 11:59, 7 June 2020 (UTC)[reply]
  • I am surprised to see little mention of interaction bans above; surely many interpersonal conflicts could be resolved if they were employed with greater frequency? --JBL (talk) 16:18, 7 June 2020 (UTC)[reply]
  • The first part to have to say is that while we do want to be welcoming, Wikipedia is not a daycare or therapy center for people who get their feelings hurt by valid criticism or being told that Wikipedia isn't their personal blog to say and do whatever they want. That goes triple for anyone who's feelings are hurt because they may lose a paycheck when their edits reverted. We do (and should) have a low tolerance for paid or COI editors who selfservingly create a drain on volunteer time and labor - I'm not getting a paycheck to clean up your poor work.
    Now... if someone is making valid-criticism in a somehow egregious manner... that generally doesn't require ARBCOM and if it does rise to ARBCOM it absolutely not a case that should be handled in secrecy. One of many reasons is that the community still needs to clean up the correctly-identified bad-work by the complainant in the past or the future. Alsee (talk) 16:40, 7 June 2020 (UTC)[reply]
  • This topic has two parts: aggressive actioning being interpreted as harassment & non-harassment incivility. I was spectacularly unimpressed with the WMF indicating that legitimate actions on Wikipedia could became harassment (or, here, at least incivility) just because they had to take place frequently. The WMF suggested going and finding another editor to avoid suggestions of harassing an individual, but that's not always going to be viable. I would suggest seeing if some slightly more in-depth help than the Teahouse usually provided could be offered here - someone who knows the right words to reach out to the actioning individual and try to help. Sort of ultra-short term mentoring, perhaps? Nosebagbear (talk)
  • As to non-harassment incivility (that is, unpleasant comments that don't violate our rules, at least as one-offs, not just being a pushy correcter). I think here it's more on bystanding editors to tell the rude one(s) to cool it. This helps, but does have the issue I see in ANI where the accuser often feels that their "harasser" has been let off with a slap on the wrist. I would suggest the Community suggest a couple of areas in Civility to enhance, without going full workplace like the WMF would like. Nosebagbear (talk) 10:07, 8 June 2020 (UTC)[reply]
    • I think the best course is to patiently explain the situation in a calmer tone without issuing reproofs to rude editors, so the object of the incivility can be assured that the dismissive attitude on the part of one isn't widely shared. I often agree with the snarkiness that is directed at what seems to be a willfully obtuse editor, without thinking that's the best way of handling it; so, my attempt at being a scold would be both hypocritical and implicitly condescending toward the offended editor. Dhtwiki (talk) 23:51, 8 June 2020 (UTC)[reply]
  • A collaborative encyclopedia is, by its very definition, a collaborative effort. Successful collaboration requires mutual respect and understanding - that's why we have WP:AGF. To my mind, it is an absolute absurdity that the argument is repeatedly advanced that, effectively, "people should be allowed to be (almost) as rude as they like, so long as they are operating within the limits of strict policy". As has already been mentioned by several above, this is a real issue in and of itself. Someone producing expletives directed at you when you are trying to have a productive conversation is actively detrimental to the work of our encyclopedia, and (as is rightly pointed out in the premise of this part of the RfC) discourages editors from contributing.
    So what to do about it? Well, we already have WP:CIVILITY - but for a "pillar of Wikipedia", it seems rarely even read, never mind enforced. Personally, I would be in favour of setting up a WP:3O style project but for civility - allowing uninvolved editors to come in and discuss a civility complaint between the two involved users. Beyond that, I think there needs to be a greater emphasis given to that "pillar" at both ANI and during arbitration. Naypta ☺ | ✉ talk page | 11:32, 9 June 2020 (UTC)[reply]
WP:civility us used as a tool to conduct harassment...it covers only superficial choice of words. It needs a broader replacement.North8000 (talk) 12:21, 9 June 2020 (UTC)[reply]
@North8000: What sort of replacement would you propose? Naypta ☺ | ✉ talk page | 12:31, 9 June 2020 (UTC)[reply]
@Naypta: A broader "don't abuse editors" type policy. WP:Civility should become a section within it. The theme for another section would be "Processes (such as Arbcom, wp:an, wp:ani, RFCU) should not unfairly abuse editors, and should not be used to conduct warfare. Trying to cleverly, unfairly use these venues (in combination with policies and guidelines) to "get" or deprecate someone is a common and destructive problem in Wikipedia." North8000 (talk) 12:47, 9 June 2020 (UTC)[reply]
@North8000: With exception to the last sentence (I think it's worth leaving out specific statements as to what is or isn't a common problem, because hopefully the aim of policy is to make sure it isn't in future a common problem) I could get behind that. I'd be in favour of something like Quora's "be nice, be respectful" policy, which is far more broadly construed than Wikipedia's current civility policy. Naypta ☺ | ✉ talk page | 12:49, 9 June 2020 (UTC)[reply]
@Naypta: You are right. I guess my thought process was too much on trying to build consciousness in that area. Wanna build an essay that would be a draft for the policy? North8000 (talk) 13:03, 9 June 2020 (UTC)[reply]
@North8000: I'm not sure it's much of a policy draft, but I've put some thoughts down at WP:EXCELLENT which you might be interested in reading. Naypta ☺ | ✉ talk page | 16:09, 9 June 2020 (UTC)[reply]
@Naypta:Nice! What a great way to come at it from! I'll post something at your talk page. North8000 (talk) 20:15, 9 June 2020 (UTC)[reply]
Based on some remarks I made on respect during last year's contretemps, I have created a draft essay along similar lines: User:Isaacl/Be respectful of others. isaacl (talk) 06:09, 10 June 2020 (UTC)[reply]
@Isaacl: It's a very good essay - well done! I think the part of it which is most germane to what's going on here actually is You can be frank about your opinions without belittling others. Many people seem to be concerned that what people are looking for is to prevent them expressing their opinions, which is of course not the case. Naypta ☺ | ✉ talk page | 09:20, 10 June 2020 (UTC)[reply]
Clearly, it's proven difficult to identify a precise and useful definition of conduct that is harmful but falls short of harassment. WMF has been using the word "toxicity", which has a lot of shortcomings. I've been trying for some time to come up with a better word, without success. But when I saw the word "belittling" here, that caught my attention. There is probably no single word that is similar in meaning to "incivility" while being more precise and less prone to "in the eye of the beholder", and yet also comprehensive. But "belittling" comes close, and fits well with what I've seen in cases where the community has done poorly in resolving the issue. Intentionally speaking to another editor in a way that makes them sound small or inferior is not harassment, and may not quite be a personal attack (if said skillfully), but it's very typical of the kind of conduct we need to be less accepting of. It's OK to say: "You need to know that WP:(fill in the blank) is something that is very important around here, and you really need to stop doing things the way you've been doing them." or "Maybe you don't know about the now-archived discussion at (link to talk page archive), but here's a link to it and I hope that you will take a look at it." But it's not OK to say: "I know how things work around here and you don't." or "That was a bad edit, and those of us who have been editing this page for a long time know that the right way to do it is such-and-such." And we really should not allow anyone to make fun of someone else unless it's just friendly banter. --Tryptofish (talk) 19:55, 10 June 2020 (UTC)[reply]
To me "belittling" has connotations of "punching down". Incivility can involve punching up, down or sideways. I'm not so worried about incivility from those who are punching up, though yes I'm aware that sometimes both parties think they are "punching up". However there is a fine line between criticising someone's edits and belittling them, perhaps finer than between disagreeing with someone and incivilly disagreeing with someone. ϢereSpielChequers 20:08, 10 June 2020 (UTC)[reply]
Those are good points, although I've definitely experienced belittling comments that were directed sideways at me but written as though they were punching down. Also, most instances that I can think of as punching up tend to be grosser forms of incivility, as in personal attacks, or clumsier, as in claims that Wikipedia is censored, and we usually are able to deal with those. On the other hand, if I think about the kinds of cases where en-wiki fails to adequately deal with the problem, they tend to be either punching down, or acting as if they were punching down. As a PS, something that I might have added to my previous comment is that getting upset after being goaded may sound like belittling, but really is something else, whereas the goading itself, if looked at more critically, actually is often belittling. --Tryptofish (talk) 21:08, 10 June 2020 (UTC)[reply]
Although I appreciate the distinction between punching up and punching down, my essay is intended to make a broader point: we can disagree with others in the community while still being respectful of them, regardless if they might be considered in a position of greater or lesser influence. Making personal insinuations or accusations doesn't make you look better in comparison. isaacl (talk) 21:41, 10 June 2020 (UTC)[reply]
What makes it impossible to manage on English Wikipedia is that the community tries to do it in a large group discussion using a consensus-like decision-making process. This is impossible without strong alignment of goals with all participants, and that in turn can't be maintained as a group grows beyond a (fairly small) size. We need to reduce the incentive for poor behaviour by making it a losing strategy for disputes. (And I know people don't like to hear this, but interpersonal issues will only get dealt with effectively by introducing some form of hierarchy.) isaacl (talk) 21:58, 10 June 2020 (UTC)[reply]
  • We're back to the civility debate with this. My view has long been that a collaborative project cannot function well without a level of civility and collegiality that we would expect in a professional work environment, and that civility should be much more strictly enforced with blocks and bans than it is now. But in practice this has not been the case because of differing expectations of civility and the "unblockables" problem discussed below. This will not change until we come to a workable consensus on this as a community or ArbCom decides to use the banhammer very liberally for incivility in the course of its cases or indeed proceedings. Sandstein 16:58, 9 June 2020 (UTC)[reply]
  • I don't think this question is a particularly good one. I could agree with all above by saying "yes, genuine incivility should lead to preventative actions and no, mistaken accusations of incivility should not". I think many above display quite a weak sense of emotional awareness; the style of comments reminds me of ANI and what makes it the absolute hellhole it is. If someone feels genuinely upset then that person deserves sympathy. Instead many of our editors prefer to abrasively accuse them of time-wasting, being a cry-baby, trying to make a civility case out of a content dispute, being too emotional to be able to edit, being unwanted on this website if they can't handle criticism etc. There are lots of situations where boomerangs are deserved because the person making a complaint is actually the one behaving the worst. But in non-boomerang cases, the person making a report should always be treated respectfully and have politeness shown to them, even if their case is not actionable. I actually don't want anyone on this site to "grow a thicker skin". I want our long-term editors to start treating other people like their contributions are valuable and their valuable free time spent volunteering is appreciated, even when you need to provide constructive feedback to someone who is making a pattern of mistakes. — Bilorv (Black Lives Matter) 10:13, 10 June 2020 (UTC)[reply]
    +1. Asking people to "grow a thicker skin" if they want to edit is a fantastic way of ensuring that our pool of editors doesn't grow, because nobody wants to work with people who refuse to treat them with respect - especially when they are doing so for free, volunteering their spare time. Naypta ☺ | ✉ talk page | 10:18, 10 June 2020 (UTC)[reply]
  • Civility warnings are appropriate for people who are incivil to others, including when they are in the right but uncivilly so. That's why we have templates, carefully nuanced prewritten messages to communicate negative messages to people. The templates aren't perfect, but some effort has gone into getting them right. They shouldn't be made generally compulsory to use, as tailored messages are generally better. But there are some editors who should be encouraged, if necessary obliged, to use templates when communicating with others. If those templates are abrasive then we can potentially involve the complainant in refining the template to put the same message in a more diplomatic way. I'm relieved if deliberate false reports of harassment are rare. But be aware, if you shift the scales so that the perception of harassment is up to the complainant to define, every troll, spammer and PR adviser will learn PDQ that claiming harassment is a way to troll or spam Wikipedia. Where there is a pattern building up of tension between two editors we can consider interaction bans, but we need better software for them. Enabling editors to know when they are are delivering more than x warnings to the same user, or when they are about to edit a page that a certain editor has edited, would make interaction bans easier to comply with, and possible to nuance. There might even be editors who should be restricted to only giving a level 3 or 4 warning to someone who had previously been warned by someone else. If an appropriate throttle was written into twinkle it might be a useful tool for Arbcom. Not everyone who finds legitimate criticism difficult to deal with is going to be at home here, but with a bit of thought and some extra tools we should be able to keep more such people, without losing some of our best patrollers. ϢereSpielChequers 16:56, 10 June 2020 (UTC)[reply]
  • Painful people suck. TonyBallioni and others wrote a wonderful essay—Wikipedia:Not compatible with a collaborative project—that I think captures this issue well. ArbCom and other editors should feel empowered to warn abrasive editors about their conduct in a kind manner with actionable advice, but I've come to agree with Tony that if people routinely cannot play well with others we should not fall into the anti-pattern of unchecked meatball:RadicalInclusiveness. Wug·a·po·des 23:54, 12 June 2020 (UTC)[reply]
  • I have found "Are civility warnings appropriate for editors who are making legitimate criticisms at a frequency or in a style that the subject finds too unpleasant to deal with effectively?" problematic since I initially saw it in a draft. This is very subjective and seems to imply that anyone could complain about legitimate processes and use this as a form of pressure to circumvent its policies. —PaleoNeonate06:34, 13 June 2020 (UTC)[reply]
    These are not new issues, but the RFC needs refinement. Even vandals are best handled by escalating blocks that are sufficiently far apart for the vandal to have time to have read one warning before they are given the next (there are times when it is appropriate to skip one or more levels of warning. But normally you take people through ascending levels of warnings, spaced far enough apart that they can be read). We certainly do have people who are overly aggressive in the way they enforce various policies, and telling such people to tone it down is something we already do. I assume that what they really mean is that some people are stressed by a tone or frequency of warning that we regard as normal. My concern is that instead of trying to change things by proposing that we slow down some of our enforcement processes and that we rephrase some of our warning messages, there is a risk of people trying to achieve change by sanctioning people for actions that are in accord with current procedure and processes. We should not forget that trying to change policies and procedures by harassing those who follow or apply those policies and procedures is a form of harassment, and inherently toxic behaviour; even when you really really disagree with that policy or procedure but can't be bothered to try and get consensus to change it. ϢereSpielChequers 07:47, 13 June 2020 (UTC)[reply]
    My concern is that instead of trying to change things by proposing that we slow down some of our enforcement processes and that we rephrase some of our warning messages, there is a risk of people trying to achieve change by sanctioning people for actions that are in accord with current procedure and processes. when writing the above, this is the type of concern I also had in mind. —PaleoNeonate08:21, 13 June 2020 (UTC)[reply]
    Maybe I'll also add: being a participant at WP:SKEPTIC and WP:FTN I commonly see editors complain and post their own essays (or userspace polemics) about how Wikipedia and its editors are sold to industry (a common conspiracy theory), that their rights are violated, that they are prevented because other editors have too much time on their hands, that WP processes are flawed, how it's even more disreputable or unreliable for not giving in (it can be for something as simple as WP not being a journal for research), etc. That may be after using every other venue possible (not to mention unimportant socks and campaigns that are already sanctionable, but also via valid civil discussions on article talk pages and noticeboards, policy pages and ARCA). When that still failed, it would be a mistake to provide gratuitious backdoors to game the system and sanction patrollers who may already be strained by their thankless work (that includes receiving negativity as gracefully as possible regularly and sometimes even harassment, the very topic). While "the resistance", either via conflict of interest or for whatever reason, may well feel or pretend to be harassed by warning messages or reverts of other editors (from "the cabal" or "the empire" to go with "the resistence"), yet that is of course subjective. These are not new issues, indeed. Ultimately, if a patroller is always really an edit warrior against consensus and policies, wouldn't they eventually face sanctions themselves using the current processes (other than frivolous complaints)? Previous text above of course suggests unblockables, but shouldn't WP:HERE, WP:CONSENSUS, WP:RS, WP:NPOV (and WP:PSCI) be more important things to focus on, considering that they affect the encyclopedia's reliability? —PaleoNeonate08:58, 13 June 2020 (UTC)[reply]
  • If there's a chance of an unsubstantiated complaint, then ArbCom should probably remove the ambiguity from our harassment policy and make it more precise. Harassment also includes hounding, and I'm not aware of a minimum threshold. Perhaps there shouldn't be any because we are communicating with text in cyberspace, so it's a mental condition, not physical. How do we gage a person's mental state over the internet or by reading text messages? Perhaps if it is serious enough, it may warrant a group phone call. Civility is more easily defined and controlled but the least acted upon. No editor should have to deal with profanity being thrown at them or abusive language and insults. Each time it's allowed to happen, it becomes a bit more unmanageable. I'm sure gender plays a role in how all of this should be approached. Female editors will probably feel more threatened by abusive language, bullying, hounding & harassment. Age would also be a factor. I think the bar for harassment/hounding should not be set too high, and the same for civility. It may even need to be lowered. Atsme Talk 📧 04:50, 15 June 2020 (UTC)[reply]
@Atsme: I don't think ARBCOM should be responsible for deciding our harassment policy (and adding "precision" can be done in a way that adds quite a lot of meaning), though I'd certainly (and am) paying a lot of attention to views of current and former arbs. We've made fairly clear that swearing at another editor is not permitted (though we've let several editors go with only mild rebuke at ANI when they faced serious aggravation and it was a rare outburst), but more general non-targeted swearing (e.g. "FUCK IT") has had case by case consideration. In terms of "no minimum threshold for hounding", that's disputable, not because there's a permitted level of hounding, but there are legitimate actions that are only hounding when ongoing, or at least repeated. Nosebagbear (talk) 19:09, 30 June 2020 (UTC)[reply]
  • Civility warnings for the kind of thing described here? No. But an FYI that doesn't come with implications of potential sanction, sure. The person making the complaint could similarly get feedback, although I'm feeling pessimistic that the type of person likely to get to ArbCom over something along these lines would actually take it to heart when so much of today's society seems to be pushing the opposite message. Anomie 17:02, 21 June 2020 (UTC)[reply]
  • I would not consider this an "unsubstantiated complaint" so much as a complaint about a different sort of behavior. If someone is chronically incivil, they should be warned/blocked/banned. I am happy for arbcom to do this if the community cannot. Even legitimate criticism of edits can be a net negative if made by someone who is chronically uncivil and unsuited for work on a collaborative project. Calliopejen1 (talk) 21:12, 22 June 2020 (UTC)[reply]
  • I discuss my complicated feelings about unsubstantiated complaints in my answer to Q2 so will provide a diff rather than repeating. I think the discussion Guy started in relation to Q3 really is the answer to this as well (or even more this than that) and so I won't repeat that here either. Best, Barkeep49 (talk) 17:47, 1 July 2020 (UTC)[reply]
  • One problem with unsubstantiated complaints is the demotivating effects it has on the wrongfully accused. Also, we know that complaints are easily manufactured. There is often disparity between the accuser being a throw-away account vs. an established contributor. I would suggest that all complaints be reviewed before being allowed to proceed to weed out junk ones. Without a sufficient basis, justified by diffs, of a repeated and long-term problem, they should be archived rather than pursued. Very few long-term editors here haven't on occasion run afoul of some rule or another or used a swear word or such. We have to try to behave ourselves, but we're not in church. Carlossuarez46 (talk) 22:22, 1 July 2020 (UTC)[reply]
    @Carlossuarez46: Very few long-term editors here haven't on occasion run afoul of some rule or another or used a swear word or such strikes me as more of the "net positive" versus "net negative" thing, which I find to be quite problematic. Someone contributing loads of great content oughtn't allow them a free pass to be horrid to other contributors; if anything, I'd argue we should hold our long-term editors to a higher standard, because they are likely to interact more with other editors. I'm sure there've been situations in which we've all unknowingly broken some policy at some point, but the test isn't "never breaking policy"; rather, it's learning from the experiences that one has, and changing and improving to avoid those issues in future. Naypta ☺ | ✉ talk page | 22:27, 1 July 2020 (UTC)[reply]
    @Naypta: Remember we're here to write an encyclopedia not right all the world's wrongs. I didn't suggest a "free pass" to someone who is "horrid", but we should balance the positive against the negative and there are many dials to turn to make that happen (including interaction bans, etc.) Regardless of what one learns to "avoid...issues" there will be numerous admins who must wade into the fray of emotion-laden conflicts that will no longer do so if the second time the make an error throw-away accounts complain they haven't learned. Not fair. Carlossuarez46 (talk) 23:16, 1 July 2020 (UTC)[reply]
    Carlossuarez46, also some things that are essential to maintain credibility (e.g. calling homeopathy out as nonsense) will cause advocates to be hurt badly in the feels, and this leads inevitably to frivolous complaints. It is reasonable to judge such cases based on relative experience. A 100,000 edit Wikipedian explaining policy to someone with 50 edits all to the same hot-button topic is likely to be indicative that the experienced Wikipedia knows more about our policies than the newbie. Guy (help!) 23:24, 1 July 2020 (UTC)[reply]
    @JzG: Agreed. Similarly, there are no newbies on ArbCom. Carlossuarez46 (talk) 23:43, 1 July 2020 (UTC)[reply]
    @Carlossuarez46 and JzG: There's a difference, though, between calling out homeopathy as nonsense and personal attacks or incivility; one can do the one without doing the other. I don't pretend to think that admins mediating disputes is an easy thing to do (indeed, part of this discussion is about trying to find better ways to do DR than just handing all conduct complaints off to admins!), but I don't think edit count can be used as a barometer of civility. Policy knowledge is a different thing. Naypta ☺ | ✉ talk page | 09:10, 2 July 2020 (UTC)[reply]
  • Agree with S. Marshall. TomStar81 (Talk) 10:28, 3 July 2020 (UTC)[reply]

5. Plausible deniability

edit

The Arbitration Committee has encountered some situations in which a person was likely harassing another user, but there was some amount of plausible deniability. For example, some statements can be carefully worded so they can be claimed to not be an attack. As another example, accounts used for off-wiki harassment are not always able to be definitively tied back to an alleged Wikipedian, and the Arbitration Committee does not possess the ability to compel information from other websites that could potentially link an account more definitively. Some argue that the accused should be considered innocent until proven guilty; others believe that harassment cannot be effectively stopped when the burden of proof mirrors that of a criminal trial, but when evidence cannot be compelled in the same way.

How should the Arbitration Committee balance the presumption of innocence with the sometimes limited amount of available evidence?

Discussion
  • A central principle operating behind criminal burdens of proof (innocent until proven guilty beyond a reasonable doubt) is that it is better to let a hundred guilty people go free rather than send one innocent person to prison. That is not the case with harassment enforcement on Wikipedia. The goal here is not justice, but building a good encyclopedia. Blocks are not punitive, but designed to prevent disruption. The appropriate standard of proof is a balance of probabilities.--Trystan (talk) 14:51, 6 June 2020 (UTC)[reply]
  • WP:AGF is a real thing for a very good reason (IMO, it is our single most important policy). If Arbs cannot subscribe to it, they shouldn't be arbs. "Plausible deniability" is speculation, and we have recently seen arb decisions based on favoritism and pseudo-evidence, so to extend speculative powers further would be unhelpful. For a sitting arb to make this kind of statement (impugning one's character and motives on an arbcase page) is unacceptable, and that we have such statements from a sitting arbs gives a good indication of where this "plausible deniability" could lead; what on earth is meant by "usually knows how to avoid wording it as a direct personal attack"? I don't make personal attacks, period. Since we have recently seen a decision where the arbs gave clear indications of not having read the evidence, with one arb acknowledging a decision based on personalities and not evidence, we must accept that arbs are human, and not extend further rights to base decisions on presumptions rather than evidence that everyone can see. SandyGeorgia (Talk) 15:00, 6 June 2020 (UTC)[reply]
  • Typically, if there is an attack, there is some behavior that triggered, or appears to have triggered the attack. If the "victim's" editing history is problematic, and the "attack" can't be tied definitively to the alleged attacker, the Arbitration Committee should always consider the possibility that the "victim" manufactured the attack to get rid of the alleged attacker. There should be no presumption that an attack actually occurred and the problem is merely one of figuring out which editor launched it. Jc3s5h (talk) 15:02, 6 June 2020 (UTC)[reply]
  • With the carefully-worded statement that has a chilling effect, Arbcom need to decide whether they're confident the perpetrator had mens rea. With the off-wiki stuff, Arbcom need to make the best determination they can on the basis of the evidence they can get. This is going to be messy and involve judgment calls, and Arbcom will make mistakes.—S Marshall T/C 15:03, 6 June 2020 (UTC)[reply]
  • The same as we would with SPI or any admin block. Without reasonably clear evidence of a policy violation, you don't sanction. Dennis Brown - 15:18, 6 June 2020 (UTC)[reply]
  • As I said above, Wikipedia is not a legal system, ArbCom is not a court, and there is no inherent right to edit Wikipedia. We should absolutely not be creating procedural rights or standards when it comes to blocking. There is no presumption of innocence in either the blocking policy or the harassment policy precisely because on a website there is almost always going to be plausible deniability and arguably reasonable doubt. Proof beyond a reasonable doubt is not a reasonable standard for a private website. It should be based on weighing probabilities and judgement, but there is absolutely not as high a standard as in a criminal case. TonyBallioni (talk) 15:25, 6 June 2020 (UTC)[reply]
  • I don't think the legal standard of proof is relevant here. In the two situations cited it would be perfectly reasonable to sanction the user. It is often going to be impossible for us to definitively prove that external accounts are tied to a certain person, expecting that would make harassment policies toothless. Deliberately skirting harassment rules also shouldn't be a reason to evade punishment. Hut 8.5 16:35, 6 June 2020 (UTC)[reply]
  • DUCK-like standards, because editing Wikipedia is comparatively 'no big deal'. Besides, even in their most rigorous use in the legal system, legal standards are no fail-safe, to begin with. Alanscottwalker (talk) 19:48, 6 June 2020 (UTC)[reply]
  • I think it has to be a case-by-case thing that rests on common sense. There is the possibility of false flags. Obnoxious stuff at other sites, really is the responsibility of those sites and not us, and we do not have to enforce a user's "right" to participate at other websites. But for that, or for more serious forms of off-wiki harassment, it comes down to a good-faith commonsense judgment for ArbCom to make, whether someone is the user that they appear to be. --Tryptofish (talk) 22:33, 6 June 2020 (UTC)[reply]
  • For anything off-wiki, the link between the off-wiki and on-wiki account would have to be absolutely definitive and proven. Any lesser standard is an invitation for editors to joe job those who they would like to be rid of. On the other hand, when the conduct is on-wiki, personal attacks are not the only grounds for sanctioning an editor. Behavior which is clearly intended to be disruptive and provocative, even if it does not quite cross the line into blatant personal attacks, is still disruptive conduct and still grounds for warnings or sanctions. Seraphimblade Talk to me 02:54, 7 June 2020 (UTC)[reply]
  • I agree with Trystan, TonyB and Hut. --Piotr Konieczny aka Prokonsul Piotrus| reply here
  • ArbCom should use good old-fashioned common sense: if it looks like a statement has been carefully crafted for its plausible deniability, assume that is the case, and act as if the statement is meant negatively. The very act of creating such a statement is itself a piece of evidence which supports the alleged harassment. Beyond My Ken (talk) 06:26, 7 June 2020 (UTC)[reply]
  • I generally agree with Seraphimblade and Hut 8.5. If someone is making harassing comments on wiki then that editor needs to be sanctioned on wiki, regardless of who they are in real life. If the harassment is off-wiki then we need to be as sure as we can be about a link before taking on-wiki action. If a link is found we should, as far as allowed by the privacy policy, share that with the anti-harassment bodies on external sites (who we must do all we can to encourage victims to engage with). Regarding joe jobs, we absolutely need to be careful here, but part of that is assessing the liklihood of that - is there any evidence to suggest someone would want to frame someone else and that person specifically, and is there any evidence to suggest that a specific person might be behind that. Thryduulf (talk) 12:37, 7 June 2020 (UTC)[reply]
  • I'll just say I expect ARBCOM to be careful, without putting any hard slant on it. I'm not sure what kind of "carefully worded" statements you have in mind, but I'd be careful about shoving bad-faith interpretations inside someone else's head. Regarding offsite accounts, we already match on-wiki editing by the DuckTest, although with offsite accounts there more risk of something like a joe job. Alsee (talk) 17:26, 7 June 2020 (UTC)[reply]
  • "Joe jobs" are a real thing, and bureaucratic structures with secret evidence and limited right to face one's accuser makes for a prime situation for this sort of abuse. I disagree with the view expressed above that "proof beyond a reasonable doubt" is somehow excessively burdensome. Carrite (talk) 00:04, 8 June 2020 (UTC)[reply]
  • So a couple of components here: 1) I get irked when individuals say just go with balance of probabilities since editing is both no big deal and our ultimate job is to create an encyclopedia. While insignificant compared to jail time, please consider the time you've spent and the reputation acquired, and then losing both, and still viewing it as no big deal. 2) That said, beyond reasonable doubt is clearly impossible. I also believe balance of probabilities is too low. Without being formally equivalent, I believe something like clear and convincing evidence "highly and substantially more probable to be true than not" is the way to go. That's moral, but also fairly pragmatic (here, I believe some level of "due process" is ultimately beneficial pragmatically as well. It's also somewhat akin to AGF. 3) As part of this, I would love to see editors above, and ARBCOM's, views on how to handle those requesting unblock who have never conceded guilt in these unclear cases. We concede we will make mistakes. But our functional unblock requirements require conceding guilt. It's like the parole situation - there are individuals who pose no threat to Wiki but are forced to either lie and concede fault or not be allowed back in. Nosebagbear (talk) 10:17, 8 June 2020 (UTC)[reply]
  • In the first case, typically users who end up at ArbCom have a long history, and it should be considered as a whole. For example, if there's just one comment that could potentially be considered an attack, then we should err towards a more charitable interpretation per WP:AGF. But if they repeatedly make such statements over a period of time, it's clear that they are gaming the system and trying to wikilawyer their way out of sanctions. In the second case, probabilistic balance of consequences is the standard I would go by, i.e. maximizing the expected value, which is not the same as balance of probabilities. Indefinitely blocking a user based on a joe job is significantly worse than throwing out evidence of off-wiki harassment in most cases IMO, so we would require a probability much higher than 50% that the user was responsible for the off-wiki harassment, something like clear and convincing evidence. In other cases the threshold probabilty would be closer to 50%, and sometimes it could even be less than 50%. For example, WP:AC/DS is frequently applied to articles where most of the editors are not disruptive, because the inconvenience suffered by a good editor is dwarfed by the potential disruption caused by a bad editor. -- King of ♥ 18:30, 8 June 2020 (UTC)[reply]
  • These cases have to be decided on their own merits, case by case, depending on whose account is the most plausible. I agree that because we are not a legal system we do not need to apply legal standards of evidence such as the presumption of innocence or "proof beyond a reasonable doubt". Sandstein 17:01, 9 June 2020 (UTC)[reply]
  • Largely agree with Seraphimblade. In the era of the JoeJob we should not be opening ourselves up for even more trolling. The day we decide we are prepared to humiliate Wikipedians and end their Wiki careers even when there is a good chance they are innocent, is the day we give up on trying to neutrally cover certain quarters of the internet. The solution to a lack of communication and cooperation between websites is better communication between websites - something the WMF should be investing in. As for snarkyness with a diplomatic veneer, sometimes the appropriate response is to make people aware that what they thought was honest and frank needs to be phrased differently. English is a language rich in invective, but also in words with double meanings and even different meanings in different parts of the world. Warnings and requests for clarification are important here, and also a commitment to globalism, and a recognition that a really important part of our community are people for whom English is not their native tongue. Something that all of us, and especially T&S, should try to remember. ϢereSpielChequers 17:20, 10 June 2020 (UTC)[reply]
  • I agree with a lot of what's been said, perhaps predictably at this point, especially the opinions of TonyBallioni and Seraphimblade. What I want to add is how these should affect the weighing of evidence and innocence the question brings up. I think the principle factor should be meatball:LimitDamage. There should not be an absolute threshold for "guilt", rather, ArbCom should seek resolutions that avoid the worst outcomes, even if it may not be the best outcome. We want to avoid disruption to the project above all else. Wug·a·po·des 00:07, 13 June 2020 (UTC)[reply]
  • If there's no plausible evidence, I think that there's not much that can be done, beyond what existing tools permit. This reminds me that the unefficiency of the current SPI+Checkuser processes (and the limits on checkuser usage) may not benefit the project, with many LTA socks undetected. Unavoidable technical limitations related to how the internet works will also subsist, unless non-anonymous editing was enforced (I don't advocate for this, but theoretically, nicknames could still be used while a small trusted group would handle real-world credential checking; projects with such restrictions have never achieved Wikipedia's popularity and likely never will). —PaleoNeonate06:51, 13 June 2020 (UTC)[reply]
  • Plausible deniability results from not properly evaluating passive-aggressive behavior, or turning a blind eye to it when it's a WikiFriend, and if the latter, recuse. Look for the motive - if there is none, then either the accused or the accuser has an issues that arbcom needs to address if they don't do anything else but explain to accused (or the accuser) why their behavior is problematic but not actionable. Examine the behavior of both parties - look at motive first, a misunderstanding second, potential personality disorder third. Atsme Talk 📧 18:44, 15 June 2020 (UTC)[reply]
  • Even leaving aside the possibilities of joe jobs others have raised above, I find myself extremely wary that one person's "obvious plausible deniability" is another's "different culture" or "neurodiversity" so sanctions based on "obvious plausible deniablilty" could easily increase discrimination against these groups. Anomie 17:07, 21 June 2020 (UTC)[reply]
  • I don't know that there should be a "presumption of innocence"--this is a strictly criminal-law concept. If you're likely harassing another user, even if there is some possibility you're not, then we should keep you from editing here. I'd say a preponderance-of-the-evidence standard is probably the right one. (Or even stricter... Even if there's only a 30% chance you're harassing someone, perhaps you should be shown the door. The damage done by permitting harassment may be such that a quite-strict standard is best.) Calliopejen1 (talk) 21:23, 22 June 2020 (UTC)[reply]
    • Even if this site hadn't had a history of having to deal with trolls and attempts to compromise the accounts of known Wikipedians, a standard that 70% of Joe Jobs should succeed in taking down active Wikipedians is not going to reduce harassment. More likely by creating open season for people from certain badsites to harass Wikipedians, such a policy would increase total harassment on this site. ϢereSpielChequers 14:35, 23 June 2020 (UTC)[reply]
  • This is the problem of off-wiki policing by ArbCom. ArbCom cannot find out about it and cannot stop it. What one may decide is an attack to another is pointing out the obvious. I've seen dozens of times comments and edit summaries along the lines of "please learn about the subject before editing this article". Attack? Intimidation? and importuning of ignorance or malice/bias? Or a helpful suggestion? We have enough trouble with on-wiki conduct that knowing what the "culture" is on various off-wiki fora is just a mess to deal with. No one should feel hounded here. If it happens here, we'll see it here. If it happens elsewhere, we cannot stop it. Carlossuarez46 (talk) 22:29, 1 July 2020 (UTC)[reply]
  • Arbcom in not a criminal court, Balance of probability and shared responsibility should be considered. Also, some people do not seem to recognise when they cross the line into personal attacks. That is their problem, and they should be held accountable for their actions whether they understand or not. That way they may eventually learn. The same applies to people who see personal attacks where their work is criticised, whether the criticism is justified is a separate issue. Others appear to not understand the distinction between incivility and personal attack. Both should be avoided, but PA is usually more serious, including innuendo and aspersions, even when made in polite language. · · · Peter Southwood (talk): 16:56, 8 July 2020 (UTC)[reply]

6. Arbitration environment

edit

ArbCom cases and other pages (case requests, ARCA, etc) have historically been areas where difficult behavior is common and is somewhat tolerated to a degree, on the grounds that the parties involved in a dispute are stressed or frustrated. On the other hand, it may worsen participants' stress to be exposed to this behavior, and it may discourage submission of evidence. When clerks and arbitrators attempt to manage pages in arbitration space, they are sometimes attacked by the participants.

Should the environment in ArbCom proceedings be more closely managed? If so, how? Should this also apply to noticeboard discussions (eg WT:ACN)? How can arbitrators or clerks perform these functions without attracting commentary that they are suppressing legitimate criticism or refusing to hear a particular point of view?

Discussion
  • I would say the question is not fully elaborated or correct: in some cases, when "clerks and arbitrators do not attempt to manage the pages" (in fact, do not even read or respond to questions on the pages), the likelihood that they will be "attacked" (that is, held responsible) increases. Yes, they should more actively engage, read, answer questions, and remove false claims that are not based on evidence. The problem I have seen is a complete lack of engagement and adequate clerking. If someone presents a statement that is not based on evidence, a clerk should be querying the poster or removing the text. Arbs should be actively monitoring their noticeboard and all case pages, and actually respond to questions asked of them. They do not and have not recently. SandyGeorgia (Talk) 15:04, 6 June 2020 (UTC)[reply]
  • In the most fraught cases, consider using separate subpages for each party, where the parties talk to clerks rather than each other.—S Marshall T/C 15:07, 6 June 2020 (UTC)[reply]
    Wouldn't this mean either a lot of people miss out on context or happenings, or it basically boils down to threaded discussion? ~ Amory (utc) 16:01, 7 June 2020 (UTC)[reply]
    If what is suggested it similar to police investigations questioning individually to evaluate discrepancy, my impression is that this would put an unnecessarily heavy burden on the volunteers running the process, other than being very different to the current method. —PaleoNeonate07:01, 13 June 2020 (UTC)[reply]
  • They are longer and more formal versions of ANI, and the atmosphere is no worse than ANI. It is fine to have more clear cut rules, but at the same time, you have to expect that emotions will pour out. From what I have seen, the clerk have the authority and ability to "trim the fat" when needed. Clearer rules, ok, but clamping down is likely to have the opposite effect, and just push problems to many areas of the encyclopedia, including personal talk pages. Dennis Brown - 15:09, 6 June 2020 (UTC)[reply]
  • To some extent, there is a difference between understandable stress and the continuation of disruptive behavior onto case pages. I think people can be allowed a significant amount of venting on pages like WT:ACN. But I'd very much like to see more active patrolling of case pages. For evidence pages, any word limits or the like should be enforced, and people who, for example, post something without diffs should get a polite message from a clerk at their usertalk, advising them how to do it right. I think that workshop pages are where things have the greatest risk of going badly, and I'd like to see more active enforcement of civility expectations. There is a real problem of some good-faith editors who are reluctant to participate in cases, out of fear of it becoming too unpleasant to do so, and ArbCom loses out on useful input when that happens. Workshop pages have spaces for comments by Arbs, and Arbs should use that to caution against bad conduct on the workshop page. (There should always be a warning before banning anyone from further participation in a case. The concern about suppressing a legitimate point of view is largely solved by that.) And Arbs should look at these things in context: the editor who is nasty to other editors on case pages is different from the editor who becomes upset in reaction to that nastiness. (Although it's not the question asked, Arbs and clerks, but particularly Arbs, also need to be more attentive to answering questions in a timely way.) --Tryptofish (talk) 22:53, 6 June 2020 (UTC)[reply]
  • Generally I think the issues are two-fold - the more direct, obvious and frequent the engagement clerks and especially arbs have with a case the less likely it is to get out of hand, for example if obvious errors or non-starters in proposals are called out officially then they are less likely to be repeated, if a back-and-forth is shut down before it gets out of hand, it wont get out of hand. The second aspect is that there should be a strongly enforced policy of two-strikes and you are out for misbehaviour during a case - misbehave and you get a warning, do something similar again and that's the end of your participation in the case, regardless of whether you are a party or an observer, with blocks also a possibility for egregious examples. However certain things should not need a warning, e.g. the most obvious personal attacks, reverting an official clerk or arb action without explicit permission, or other unquestionably bad faith contribution should see removal from participation in the case as a first step with further sanction to be discussed if desired. Poor behaviour during a case should also feature in the final decision, with the possibility of sanctions like a standing warning for future arbitration cases or prohibition on participation in cases where they are not a party being considered. Arbcom should also be free to pass official comment and/or sanction on the behaviour of everyone who has participated in a case, whether a party or not. I'd also suggest that expulsions from the case should be recorded as findings of fact, even if no further action is felt desirable at that time, so that it is on record. Thryduulf (talk) 12:54, 7 June 2020 (UTC)[reply]
  • Yes, they should be more closely managed.
    • First, civility and decorum should be rigorously enforced. I have seen multiple case requests where editors commenting on the request have insulted other editors, sometimes with vulgar profanity, or cast aspersions, or even called for the filer to be indef'd for filing the case request (I can provide examples if that's helpful, but I'm not doing it so as not to drag up old stuff). Predictably, this has a chilling effect. The most I've ever seen is one editor be warned once. This sort of behavior should result in, at least, a formal warning, every single time it happens. Some have expressed the opinion that because Arbcom cases are high-stress areas, more leeway should be given to editors to blow off steam. I think the very opposite should be done: we should be more formal, more professional, and give less leeway for incivility, because it's a high-stress environment. Keeping it clean helps keep it orderly. This is why the standards of decorum are rigorously enforced in every court. Active management of decorum on case pages will make the environment less stressful, and will encourage productive participation from more editors.
    • Second, Arbs should engage much more directly with participants, and much earlier in the process. An Arbcom case is kind of like a black box. You present the case request, and maybe it's an accept or maybe not, then there are evidence and workshop phases where many arbs stay totally silent, then the always-exasperating wait for a PD, and then--BOOM--a PD issues, and we all hear Arb's comments for the first time. I've heard some Arbs say they don't read any of a case until the PD is posted. This almost always results in a disconnect between participants and arbitrators. I'm surprised every time I read a PD, to see how the Arbs view the evidence and workshop phases. It seems in every case, we find out, very late, at the PD stage, that large swaths of the evidence and workshop pages are viewed as not relevant by a number of Arbs. Boy, it'd be useful to know that at the outset, instead of finding out at the end.
    • What arbs should do is to provide feedback early in the evidence phase, at the end of the evidence phase, and throughout the workshop phase. Arbs should engage in Q&A: they should ask for the evidence they want to see. They should say what evidence they think is irrelevant. They should say which workshop proposals they think are worth pursuing, and which are non-starters. Interaction, not word limits, is what will bring efficiency to the process. I'm sure of this much: Arbs don't want to read 100k, and editors don't want to write it, either. But when you're presenting evidence and workshop proposals into a black box and get no feedback, you don't know what to do other than to send everything, in the hopes that in there will be something that helps the Arbs. Active feedback from arbs throughout the case presentation will help editors focus their efforts, and save everybody time.
    • Clerks can enforce decorum. Only arbs can give substantive feedback on a case. I'm sure both groups could do this without suppressing legitimate points of view. It doesn't have to be heavy-handed: clerks shouldn't ban participants from case pages for one unkind word, nor should arbs prohibit a workshop proposal from being posted simply because they disagree with it. But clerks could give a lot more warnings, and arbs can give feedback that is helpful instead of stifling (e.g., say why you like or don't like a proposal, or find evidence useful or irrelevant, rather than simply declaring things "good" or "bad"). Someone will always complain, but the goal shouldn't be to avoid all complaints, but rather to have an open, efficient, and fair process. Levivich[dubiousdiscuss] 18:46, 7 June 2020 (UTC)[reply]
      @Levivich: Some have expressed the opinion that because Arbcom cases are high-stress areas, more leeway should be given to editors to blow off steam. I think the very opposite should be done: I could not agree with you more. Not only at Arbcom cases, though, but everywhere else too: being under "high stress" is not in any way, shape or form an excuse for incivility. If editing is putting someone under so much stress that it causes them to consistently lash out against other editors, then their presence is not conducive to creating a healthy environment in which people can edit. Naypta ☺ | ✉ talk page | 10:35, 11 June 2020 (UTC)[reply]
Although I very much share the view that case pages should be under more control against battleground conduct, I have to push back against the idea that stress is never a valid consideration. In the real world, with real people, unpleasant situations lead to stress. We cannot expect editors to be anything other than real people. On the one hand, someone who is habitually disruptive should not be able to habitually hide behind a claim of stress. But on the other hand, even the kindest of editors can find themselves momentarily in a genuinely stressful situation that is not their fault, and especially so during ArbCom cases. Ideally, we want those editors to step away from the computer for a brief time. But that doesn't always happen. Someone who lets loose with an angry post one time, but does not do so habitually, should be entitled to having the post examined in context. All too often, I've seen civil POV pushers and even trolls provoke a kind and gentle editor into an outburst, and that editor gets blocked while the provoker walks away smugly. --Tryptofish (talk) 16:27, 11 June 2020 (UTC)[reply]
@Tryptofish: Of course, it's understandable that people will not always be at their best; as you rightly say, editors are humans, after all. However, I'd argue that self-declared stress or lack of stress ought not to be included in considerations regarding incivility. The majority of the time, it would be inappropriate to block someone for incivility on a "first offence" basis - the flip side of that is that after two or three times, there's a problem that needs serious addressing, stress or no stress. As many people like to point out in these arguments, Wikipedia is not therapy - which goes both ways; if an editor is being repeatedly incivil to other editors, whatever the reasoning for that may be, they should be prevented from doing so. Naypta ☺ | ✉ talk page | 17:51, 11 June 2020 (UTC)[reply]
With that clarification, we pretty much agree, and were just saying the same things in different ways. My concern here is principally on behalf of editors who are not chronically incivil. I would slightly adjust "first offense" to "first offense in a long time", and I do think that context always matters. --Tryptofish (talk) 21:53, 11 June 2020 (UTC)[reply]
I want to come back here to say that I think that the suggestion about Arbs being more active on case pages earlier on, and giving more feedback about what is or what is not useful, is an important idea that really should be followed through on. I've been noticing a trend in which editors deflect from the issues at hand by raising clearly bogus complaints framed as boomerangs (à la Roy Cohn). This is not at all the same thing as genuine boomerang situations, and it's becoming something that seriously disrupts case pages when it goes on and on and on. Arbs stepping in before it gets out of hand is not the same thing as suppressing legitimate discussion, and would really be helpful. --Tryptofish (talk) 21:07, 7 July 2020 (UTC)[reply]
  • On WT:ACN, don't restrict that further - it's got to be emotional, as otherwise venting erupts in lots of different places. Arb actions are quite hard to directly do anything on, so emotions can't just be put to good use. On case pages, hmmm, there are some egregious comments that need better handling. For "darker" or more substantive aspersions or accusations (that is, not just saying someone's an F-ing idiot), then Clerks need to start demanding diffs. Those diffs only need to meet an ultra-low level (judging their true worth is an Arb job - if we're going latin, call it prima facie evidence), but filtering out completely unjustified dross (when diffs aren't provided on demand) will improve things without over-egging it. Nosebagbear (talk) 10:24, 8 June 2020 (UTC)[reply]
  • If we're looking for solutions, I think we need to look at how discussions are structured and the ways a structure can influence behavior. I want to put forward a rather radical change to how arbitration discussions are structured in the hopes that it can reduce current problems and give future benefits. Rather than policing discussions for problematic behavior, clerks actively refactor discussions from c2:ThreadMode to c2:DocumentMode. This may seem weird, but I think we're already halfway there. Arbitration is structured to discourage threaded discussion because threads quickly become flame wars. The problem, I believe, is that case pages are still in ThreadMode; even if you have to respond to others in your own section you're still engaging in all the trappings of a thread. So why is ThreadMode the problem? (see also c2:ThreadModeConsideredHarmful)
    • ThreadMode emphasizes the trivial: if Arbs need to know who said what when, they have the page history. We don't need to save every comment on the page forever.
    • ThreadMode prevents tensions from defusing: reading discussions, we get a sense of the tone and are influenced by it. Because a tense thread remains on the page forever, new editors to the discussion may revive a tense discussion even one that died out. Letting old threads stick around increases the chance that someone will reignite tensions.
    • ThreadMode incentivizes conflict: clerks essentially have three options to deal with a problematic comment, none of which are good: (1) do nothing, (2) redact part, or (3) revert the whole comment. None of these things defuse tensions, and some of them make tensions worse. Structuring discussions in a way that gives clerks less confrontational options will reduce conflict.
    • ThreadMode is not a useful end product: who wants to read through an entire arbitration case page? The threads all contain information that is useful, but they also contain a bunch of stuff that takes up space and is useless. Ideally the initial phases of a case would produce some sort of neutral summary of the dispute and issues that is useful in both deciding the case and understanding what happened years later. Thousands of threads do not do this.
To resolve these problems, we could have clerks actively refactor stale threads into DocumentMode. Using the statements in the thread, the clerks would extract the important points and summarize the issues discussed, and this summary would replace the thread. Subsequent editors could opine by adding a signed comment below or by directly editing the summary. Subsequent threads can then be refactored in the same way. The benefits of encouraging DocumentMode are that
  • DocumentMode dissuades interaction: with minimal chance to interact, there are fewer opportunities for editors to get into conflicts.
  • DocumentMode encourages intervention: editors don't like reverting comments, and they don't like their comments being reverted. DocumentMode gives clerks more options to handle disruption than wholesale reversion. Rather than completely reverting an edit to the summary, clerks can edit it like an article and keep the useful stuff. Editors will still feel heard and clerks won't have to control discussions through direct conflict.
  • DocumentMode leads by example: because the process of refactoring and editing the summary are similar to how we build encyclopedia articles, it provides an example to the parties on how they should behave in articles after arbitration concludes, reducing the chance of further problems.
  • DocumentMode encourages a neutral tone: because heated threads do not stick around, new editors to the discussion will not feel as if they need to get involved in a heated exchange or as if the place is a battleground. Instead, the summaries and limited amount of reading encourage outside participation from those who currently avoid cases due to their heated nature.
  • DocumentMode encourages collaboration: parties can work together to develop a shared understanding of the dispute and get to feel heard as their comments get worked into the summary.
  • DocumentMode produces a useful end product: arbitrators would have a useful and neutral summary of the dispute to work from (perhaps speeding up turnaround time) and years later the community can understand the dispute without having to read hundreds of comments. a useful final product for arbitrators and the community.
It would not be a panacea, and there probably need to be some procedural rules. To prevent conflict, I would suggest only clerks or arbs could convert threads to DocumentMode, and as normal editors who engage unproductively can be warned or sanctioned.
It sounds strange, but it is actually very similar to how we already build articles; it's similar in spirit to Wikipedia:Edit this proposal and is how essays were often written and discussed on MeatBall. It won't be a panacea, and there will likely be growing pains as people adjust to the format. However I think such a change would be a useful experiment to see if it can remedy the climate issues in arbitration cases brought up by this question. Wug·a·po·des 01:10, 13 June 2020 (UTC)[reply]
Once upon a time, I made a similar proposal to have designated persons collect the evidence and feedback from interested parties and consolidate it together. This would reduce repetition and the number of threads that have to be followed, making it easier for people to follow the case with less time investment and thus freeing them up for other things. isaacl (talk) 01:49, 13 June 2020 (UTC)[reply]
  • Someone (a clerk, an arb, an admin) needs to oversee each case for the purpose of enforcing proper WP:DECORUM, which would include things like removal of aspersions, warnings for incivility & PAs, hatting irrelevant comments, and doing whatever else needs to be done to maintain civility and stay on point. It actually should apply to all the drama boards. There is no shortage of POV pushers and bad actors on WP who are experienced at picking their target, freezing it, personalizing it, and polarizing it - see WP:POV railroad - and that includes harassment and hounding. If it is happening during an arbcom case, arbs need to recognize the signs and quickly take action to nip it in the bud or at least appoint someone who can. I doubt the purpose of WP:IAR was intended to allow for incivility, especially when determining a case about problematic behavior. Arbcom was created to find remedies/resolution for difficult cases, so what purpose would they serve if they can't control disruptive behavior during the cases they are reviewing? One other thought - allow editors who have been wronged/harmed the freedom to present their case without the chilling effect of a potential boomerang. Maybe the first person who suggests it should be the target of the boomerang, not the parties in the case. They are already going through hell. Atsme Talk 📧 19:24, 15 June 2020 (UTC)[reply]
@Atsme:, I'm not sure if there are more rhetorical parts of this paragraph, but I'd like to address the latter part. I get concerned when I see proposals like this - it is fairly frequent, on any conduct consideration location, for a case raiser to have made actions that absolutely warrant a boomerang, though certainly they are sometimes used as a weapon (both "offensively" and defensively), and someone has to raise that and do so before we perhaps lose the other side to a block (someone can be both victim and legitimate accused). In terms of ARBCOM cases and case requests, perhaps we should make it a binding aspect that someone who submits a case can only be judged for a boomerang action either as part of a case or (if the case is rejected) by an arb acting as a regular administrator. The latter is a little complicated since we don't normally allow individual admins to rule on non-simple cases. More thought required, perhaps. Additionally, misbehaviour on the part of the submitter does not invalidate their accusations. Significant delays in boomerang consideration offer an appreciable first-mover advantage. Nosebagbear (talk) 19:18, 30 June 2020 (UTC)[reply]
Nosebagbear - I'll start with the following quote by MalcomX because it fits in so many ways: “The media's the most powerful entity on earth. They have the power to make the innocent guilty and to make the guilty innocent, and that's power. Because they control the minds of the masses.” Apply it to WP as it relates to groupthink, and an editor who is charged with disruption by his/her detractors at a controversial article (in AP, GMO, GG, PS, etc.). However...the few who take the time to actually read the diffs in context may very well find that the editor wasn't being disruptive at all; rather, they were consistently effective in discussions/RfCs and the opposition simply wanted them gone. How does the accused present that case to arbcom, knowing the accuser is lying, exaggerating, giving unsupported reasons under the guise of disruption which is immeasurable or ambiguous because our PAGs are unclear and subject to one's POV for interpretation, such as Stonewalling, CIVIL POV pushing, WP:Gaslighting, Wikilawyering, IDIDNTHEARTHAT, CIR, etc. There are no distinguishable lines drawn in the sand. The outcome of any case depends on the number of editors who show up and what power they yield in the community. When I say power, I mean level of popularity in the community, including support from admins who are quick to defend them, right or wrong. It doesn't help the project one bit - we're losing editors as a result - we've seen it played out in the Jytdog case - bully editors who get away with things that others have been site banned over, some for less. Sadly, WP has become a mirror of mainstream media with its opinionated journalism. We also reflect some of the troubling issues we've seen on university campuses regarding safe spaces and free speech. It's a problem. Perhaps disallowing involved editors & admins to participate would help, and instead of using the boomerang force the accusers to open an entirely different case. Atsme Talk 📧 21:25, 30 June 2020 (UTC)[reply]
  • Arbitration is advocacy, which is emotional and persuasive writing. You cannot divorce emotion from it. That said, I would expect that any indecorous words can be edited by ArbCom. Carlossuarez46 (talk) 22:56, 1 July 2020 (UTC)[reply]
  • Right now much of arbitration space, but especially WP:ACN just lets any cockwomble say what they want. The complete lack of boundaries undermines not only arbitration procedures, the arbitration commitee, but the expectations of what is acceptable onwiki in general. Things that are tolerated in some places, you know like replying to people and having threaded discussions, as we are here, are not tolerated at ArbCom. Yet, for some reason, ArbCom isn't actually any more decorous than any other conduct forum and many times a good deal less. People do, of course need to have space to blow off steam. So I do understand why Arbs would turn a blind eye towards a lot of thing directed at them. But I would hope the arbitrators and clerks would use their judgement, the judgement by the way that we elect arbs for in the first place, about where the line is. Because right now there is no line and there is no judgement being used. Anything goes. And that permissive attitude is, from what I can tell, a big reason we got FRAM in the first place. I would hope that we could learn something from that. Best, Barkeep49 (talk) 03:24, 19 July 2020 (UTC)[reply]

7. "Unblockables"

edit

There is a theory on the English Wikipedia that some users are "unblockables": long-term users who routinely act in ways that would get other users blocked, but who are either not blocked for it or are blocked and then quickly unblocked. This has presented a particular issue for the Arbitration Committee when deciding on sanctions, because choosing a sanction short of an ArbCom-imposed siteban often requires trust that the community will adequately enforce it if it is breached.

How should the Arbitration Committee handle users who have behaved in a way that is sanctionable but not risen to the level of a ban, when the user exhibits the characteristics of an "unblockable"? How can the community better avoid its administrator community treating various users as "unblockables"?

Discussion
  • I think admonishments still carry weight, serving as a final warning by the Committee before actual sanctions are applied. El_C 14:44, 6 June 2020 (UTC)[reply]
  • The Committee has recently solidified that there are in fact priviledged editors for whom they will not even put forward findings of fact such that arb enforcement has something to act upon should the behaviors continue. In a case where it was well evidenced and understood that the community had not and would not act, the arb committee failed to even put forward enough findings of fact so that some brave admin might be able to act in the future. So, we have a real problem here, not only at the community level, but also at the arbcom level, that can only be addressed via more careful attention to arbcom elections and more opposes at RFA of editors who simply have not demonstrated time in the trenches building articles, as those kinds of editors tend to turn into admins and arbs who make such mistakes. SandyGeorgia (Talk) 15:10, 6 June 2020 (UTC)[reply]
  • Eric Corbett is the example for this question, I'm sure. I don't think any change is needed. How this is handled varies from year to year, depending on who is at Arb, and I don't see any rule change as making a difference. What makes someone "unblockable" (and I disagree with that assessment) is solely the support they have from the community, not any restriction or limitation of Arb. Dennis Brown - 15:12, 6 June 2020 (UTC)[reply]
    • I am not sure Eric can be characterized as "unblockable" considering how often he was blocked, and not always with good reason. There are true unblockables that we should be considering-- for example, admins who editwar at great length across many articles and many years after an arbcom block for edit warring, and yet are never blocked. Or who insert POV into articles based on COI and aren't even admonished. That is the definition of "unblockable", that Eric certainly was not. SandyGeorgia (Talk)
      • I only meant he was heralded by many as the poster child for "unblockable" for many years, but I get your points. Odd enough, I worked with him on many articles and found him a delight to work with, but a pita in many respects, as well. Dennis Brown - 15:25, 6 June 2020 (UTC)[reply]
  • Simple: accept user conduct cases that have had several AN/I threads with a low threshold. Recognize that "unblockables" are protected with premature AN/I closes, non-policy based calls for a boomerang and counter-pile-ons by friends who pop up every time the user is being discussed. --Pudeo (talk) 16:03, 6 June 2020 (UTC)[reply]
  • Everything that Pudeo just said. If a user has a long history of noticeboard reports that almost always end without action being taken (often with a few blocks that were quickly lifted somewhere in the mix), there is a high probability that the community has become incapable of dealing with the editor's behavior. It is for situations like these that the Arbitration Committee was created. LEPRICAVARK (talk) 16:09, 6 June 2020 (UTC)[reply]
    +to what Pudeo said.Moneytrees🌴Talk🌲Help out at CCI! 02:16, 7 June 2020 (UTC)[reply]
  • If someone has a pattern of unacceptable behaviour then it needs to stop, one way or another. If they aren't willing to change and we aren't able to come up with an effective sanction then a ban is the only remaining option. I agree with the suggestion that ArbCom should be more willing to take this type of case, the arbitration process isn't nearly as vulnerable as the noticeboards where it's almost impossible to deal with unblockable editors. Hut 8.5 16:48, 6 June 2020 (UTC)[reply]
  • This is complicated. I think in response to FRAM ArbCom has, understandably, been more active than probably needed. While every ArbCom candidate says they don’t think a case necessitates sanctions, I’ve yet to see that actually happen. Once a case is opened at the very least someone is getting admonished and usually a TBAN or desysop.
    I think the recent portals case is a good example of the type of case where the structure of the proceedings resulted in an outcome that wasn’t in the best interest of the project (the BHG desysop.) Noticeboards are more visible so while people’s friends are more likely to comment, so are more uninvolved community members. The “pre-mature” closes discusses above aren’t always pre-mature, and usually the community is fairly capable of pointing out bad conduct on all sides. ArbCom cases are by design away from the community boards in other to give a structured space to present evidence. This has the negative ramifications of usually involving only people who have some interest in the case scope. For civility cases, it can easily turn into a railroading. The committee in good faith acts on the information presented, and then people get mad after the fact because of an unfair outcome that they weren’t aware of that impacts someone they respect.
    Tl;dr: there is a role for ArbCom here, but I don’t think it should be more active in enforcing it than it currently is. Honestly, I think that in reaction to FRAM this ArbCom has probably become more active than desirable in these cases. This is a project full of people with issues over power structures. Having the closest thing to a power structure we have more actively policing day to day fights is not going to end well. TonyBallioni (talk) 17:12, 6 June 2020 (UTC)[reply]
The use of the term 'Unblockables' not only is incorrect but causes the problem to be misunderstood. The example of EC is a good example. These editors are frequently blocked, and the block is then always lifted. What typically happens is that one administrator lifts the block, and the lifting of the block is widely condemned, but the block cannot be reinstated, because blocks are preventive and not punitive. Some other term should be used, but I am not sure what it is. ArbCom needs to take up cases of editors who have lengthy block logs. A very lengthy block log is evidence that the community is not dealing with the user effectively. Robert McClenon (talk) 18:24, 6 June 2020 (UTC)[reply]
For an example, I cite the two ArbCom cases of Joefromrandb. The first ArbCom case was declined, because the subject editor agreed, after pleading by multiple admins, to try to change their behavior. Concerns about their behavior had been documented for more than four years. The second ArbCom case was accepted after their behavior did not change. An earlier ArbCom case might have permitted the editors who were taunting the subject to be identified and also sanctioned. I realize that this example is not about harassment. He wasn't harassing other editors, just being a net negative. Robert McClenon (talk) 18:24, 6 June 2020 (UTC)[reply]
  • I think there's really two categories here as to why someone could be an "unblockable" - high edit/high content users who are sometimes uncivil, and administrators. As far as the latter, there has always been a stigma at blocking administrators who do something that would have gotten a non-admin blocked. Perhaps this should not be. --Rschen7754 18:51, 6 June 2020 (UTC)[reply]
    Three categories: Unblockables I are high-quality /volume content users, Unblockables II are administrators, Unblockables III are users who have a lot of wiki-friends, either via affiliates / ,eetups or via activity in Wikiprojects.--Ymblanter (talk) 17:52, 7 June 2020 (UTC)[reply]
    You left off the most truly unblockables: WMFers. SandyGeorgia (Talk) 18:27, 7 June 2020 (UTC)[reply]
    They are not unblockables. They are not even Wikipedia editors (at least not under their WMF accounts, and I can not recall any problem with any of their volunteer accounts in the last five years). In this sense, they are just bureaucrats (not in the Wikipedia crat sense) performing some functions. Whereas I agree with the sentiment below that the WMF culture must change (and it would be great if ours could change as well), the WMF problem has no relation to the problem of unblockables.--Ymblanter (talk) 18:32, 7 June 2020 (UTC)[reply]
  • The problem is that, through a bunch of practices, coalitions etc. admins are defacto exempt from any review, accountability, guidance etc. except by arbcom. And the high bar and big deal that it takes to be an arbcom case means that 99% of the needs for admin accountability, review, guidance etc. will never get taken care of. We need an additional venue to handle the other 99%. Probably a group of 5-10 people with immense experience, thoroughness, self-control and objectivity that could individually handle these cases. North8000 (talk) 21:51, 6 June 2020 (UTC)[reply]
    • One of the first action after becoming an admin was starting Wikipedia:Request for Admin Sanctions along with a former Arb. Participation was marginal, even with heavy notifications. There have been many other proposals, each getting about the same response. The community has a history of getting very luke warm once you put an idea in front of them. I have no idea why, but the idea of an accountability board for everything short of an Arb case has been shot down every time, which is why you seldom see action against admin. And yes, most of the time, admin hesitate before blocking another admin, no matter how much they would like to say otherwise. Dennis Brown - 18:19, 7 June 2020 (UTC)[reply]
      • @Dennis Brown: My idea of a subset of admins with the above qualities would solve that and if expanded a bit later would even 1/2 solve the RFA problem. (the other 1/2 is more easily solvable) They could also handle cases involving experienced editors so that giving someone the tool belt is less of a big deal. The right 2-3 people could just start listing them and the criteria as an essay and then I think it could slowly gain traction /evolve into place without some gigantic decision. Wanna try? North8000 (talk) 11:27, 8 June 2020 (UTC)[reply]
        • I'm happy to participate if it gets moving, but honestly, I find I'm happier if I avoid taking the lead on anything political around here, and anything to do with admins is always political. Dennis Brown - 16:41, 8 June 2020 (UTC)[reply]
  • To some extent, the word "unblockables" is what is in the eye of the beholder. There's no single answer for all cases. It's appropriate to consider an editor's overall contributions to the project when determining remedies. But the same policies need to apply to everyone. This has been hard for the community to deal with effectively, so ArbCom definitely has a role here. Also, I disagree with the opinion that ArbCom has become overly harsh after Framgate. I've followed some recent cases where editors have railed against ArbCom for some decisions, but I think ArbCom got those right, and if anything ArbCom is too hesitant to deal with incivility that falls short of clear-cut personal attacks. --Tryptofish (talk) 23:01, 6 June 2020 (UTC)[reply]
  • Judging from some comments directed at me on AN/I, I believe that there are editors on en.Wiki who see me as "unblockable", but in need of being shown the door. On the other hand, other editors (or even the same editors at other times) will point to my (sadly) all-too-extensive block log as proof that I'm a bad dude and need to be given the boot. (I can't win for losing.) This, perhaps, gives me a somewhat unique perspective on this question.
    I, too, have perceived a small number of other editors as being "unblockable", but over the course of years, many of those editors have been substantially blocked, or indeffed, or topic banned from their primary subjects and have walked away on their own. In truth, then, it seems that being "unblockable" is a matter of perception and context. It's really just the age-old balancing question writ large: is this editor a "net positive" or a "net negative"?
    In most cases, it's fairly easy to evaluate that: brand new editors who misbehave or are disruptive without having made any substantial contributions to the project may be given a second chance, but is still often blocked easily. Obviously, it's a lot harder to make that judgment with an editor who's been here for a long time and has a substantial corpus of productive editing. Further, longevity confuses the issue because -- no matter how much we long for a collegial editing community -- the reality is that many of us can be, at times, cantankerous or ornery, and others of us can be oblivious to the effect they are having on other editors. This means that the longer one is here, the more likely it is that one is going to get into disputes with others. This underscores the perception of being "unblockable"; if one edits a lot, and therefore is seen at numerous places across the project, and gets into disagreements, but never seems to get the "just desserts" that would seem to be coming, the "unblockable" tag would be almost inevitable.
    Unfortunately, I have little by way of recommendations as to what to do in regard to "unblockable" editors, except to continue to evaluate their worth to the project in terms of productive contributions, and attempt to weigh that against their behavior. There's no easy answer here: it would be foolhardy to eject excellent content creators willy-nilly simply because they rub some editors the wrong way, but it's also detrimental to allow them to have free reign to misbehave. Admonitions, warnings, interaction bans, topic bans, and short blocks should all be utilized as seem appropriate under the circumstances. One thing should be avoided, though, which is to block "unblockable" editors simply because they are perceived to be unblockable. There's generally a reason that they've been around long enough to be labelled that way. Beyond My Ken (talk) 00:40, 7 June 2020 (UTC)[reply]
    I enjoyed reading this thoughtful and self-reflective comment. --JBL (talk) 16:24, 7 June 2020 (UTC)[reply]
    I also liked reading it and think that it reflects my perception of WP:HERE, which is likely a better perspective to evaluate this topic than the unblockable concept. —PaleoNeonate07:40, 13 June 2020 (UTC)[reply]
  • The answer to this is beyond simple: Arbcom should siteban said unblockables to save the community from the trouble of having to enforce these elaborate half-measures which merely delay that inevitable outcome. –MJLTalk 02:29, 7 June 2020 (UTC)[reply]
  • ArbCom should evaluate the behavior of an "unblockable" the same way as anyone else. If that would call for a site ban, site ban them. If that calls for restrictions, restrict them. If that calls for no action, do nothing. This is a collaborative project, and no one should be above expectations to treat others decently and civilly. Sure, nobody's perfect and we all have bad days, but if someone's constantly being an ass toward others, that can't be allowed to happen with impunity. Seraphimblade Talk to me 03:04, 7 June 2020 (UTC)[reply]
    • I disagree. Whether Wikipedia likes it or not, and whether or not it's acknowledged officially, Wikipedia is a meritocracy -- a project such as this could hardly be anything else and succeed. Because it is a meritocracy, evaluations of behavioral problems need to be made in conjunction with evaluations of the editors' worth to the project, that's just simple good sense. Having a zero-tolerance policy is almost always a bad idea, since it shifts responsibility for thinking up the food chain to an authoritarian figure farther removed from the front lines. This is why I have an almost-zero-tolerance for zero-tolerance policies. Beyond My Ken (talk) 06:00, 7 June 2020 (UTC)[reply]
      • I believe I was clear I was not talking about "zero-tolerance": Sure, nobody's perfect and we all have bad days.... And no one's suggesting we should ban excellent editors for a few intemperate comments over a period of years. But what I'm talking about is when people think that, because they write good stuff, they can behave however they like toward anyone else anytime they like. That, I will absolutely not go for. So, certainly not "zero tolerance", but no one should get unlimited tolerance either. Seraphimblade Talk to me 06:09, 7 June 2020 (UTC)[reply]
      • Your comment, BMK, helps to focus the problem: those who have lots of free time become more equal than those who can only work occasionally for the projects. Now, some people have this free time due to retirement or other handsome settlements from their former employers, while others are, in fact, paid for their participation here. (cf. the recent investigation at fr.wp mentioned in this month's Signpost) The wiki-merit of this class of "under-employed" knowledge workers is likely perceived to be quite high in terms of gnoledge production.
Those who dominate the drama boards in specific content areas also tend to be among those who have the most free time to give. I look forward to the brave tomorrow where anyone who calls fellow contributors pigs, dogs, cabal goons, or liars -- (without providing the glossy diffs with circles and arrows showing the gleeful wallowing, loony howling, dextrous cloaking, or fulsome fibbing) -- will be auto-blocked for a spell... no matter who they are or aren't. However, when those glossy diffs are provided, it shouldn't be possible to shout down the evidence by calling up a posse of notice-board minions chanting about golden boomer-rings. -- SashiRolls 🌿 · 🍥 10:05, 7 June 2020 (UTC)[reply]
I generally agree most, of all the users, with the comments of Seraphimblade on this page, but not in this section. Of course once the case has been accepted, the unblockables must be treated the same way as everybody gets treated. The problem is usually that the case does not get accepted. For example, in the FRAM 2017 case, all arguments which featured in the 2019 case, were already there, and still the case was declined because the arbs thought there was insufficient prior resolution (indeed, every ANI case was turning into a drama, and when there is excessive drama it is impossible to judge the merits of the case), and they were not interested in investigating the pattern themselves (this is what T&S has finally done). At that point, something must be done, and treating the unblockables as everybody else would just results in declining the case.--Ymblanter (talk) 18:00, 7 June 2020 (UTC)[reply]
  • It's good that ArbCom is considering this, because IMHO this is quite true - some people think they are above the wiki-law, and likely, are. What gets one person blocked won't get another one. I do have a solution for it, but it's likely a dead horse: admins should not be granted their privileges for life, but for a period of 2-5 years, and then they should require re-election. This won't solve everything, but would be a major help. --Piotr Konieczny aka Prokonsul Piotrus| reply here 05:00, 7 June 2020 (UTC)[reply]
    • That will make the problem worse. If an admin blocks an unblockable, then when they come up for confirmation all the unblockable's friends will oppose their RfA because of their outrageous block. Realistically the threat of this will just deter the admin from making the block in the first place. Hut 8.5 12:36, 7 June 2020 (UTC)[reply]
      What you are saying is true, but the other side of the coin is that admin re-election would likely result in the removal of admins who have failed to live up to community expectations. The problem is that this system could easily become politicized (i.e. X opposes Y's re-election because Y oppose Z's re-election and X supported Z). I think community de-adminship is a better solution, but the reality is that such proposals will always be derailed because of the sizable influence of admins who have a vested interest in maintaining the status quo. LEPRICAVARK (talk) 19:46, 7 June 2020 (UTC)[reply]
  • I agree with Pudeo way up-thread, arbcom needs to be prepared to accept cases for investigation based on an extensive history of low-level behavioural issues, even if there is no single egregious thing that can be said to have crossed any particular line. Thryduulf (talk) 13:07, 7 June 2020 (UTC)[reply]
  • Block them. ~ Amory (utc) 15:46, 7 June 2020 (UTC)[reply]
    • A wonderfully simplistic solution - throw out the baby with the bathwater (you can always get another baby). Beyond My Ken (talk) 21:50, 7 June 2020 (UTC)[reply]
      This is an encyclopedia; not a day care. –MJLTalk 22:35, 7 June 2020 (UTC)[reply]
      • That's correct, and prolific content creators write and expand that encyclopedia, which is why you don't throw them away like yesterday's newspaper without determining whether keeping them is worthwhile or not. Just saying "block them" is not productive; it can be, in fact, highly detrimental to the project. We can't on the one hand, worry about editor retention, and on the other hand waste valuable editorial resources like prolific content contributors without due diligence. Beyond My Ken (talk) 01:24, 8 June 2020 (UTC)[reply]
    Why not?
    We've tried crafting case-by-case specialized sanctions for these individuals, and we all know that hasn't worked. We've tried just "letting them be", but they've never once indicated a willingness to follow the rules. What exactly should we do beside block them and finally be done with it?
    We have users here who have wasted years of the community's time and energy trying to get them to change for the better. No amount of "good work" they do will ever be able to justify that to me when we have seen absolutely zero improvement on their end. –MJLTalk 18:29, 8 June 2020 (UTC)[reply]
    Yeah... entirely disagree. If someone is creating a hostile environment for others, they should not be doing so; if they cannot stop themselves, then they should not be here. Someone who routinely act[s] in ways that would get other users blocked has already been determined to not be worthwhile to keep. Maybe my threshold is different then yours, but everything else is deflection. ~ Amory (utc) 09:21, 9 June 2020 (UTC)[reply]
  • @Beyond My Ken: I agree with Amory and MJL above - my immediate response is "why not?" too. Wikipedia is far bigger than any one user. There is no reason, as far as I can see, why any user should be treated differently to any other in respect of harassment, bullying, abuse - or, in fact, breaking any of our rules. Breaking the community's rules arguably does more harm as a more prolific user. The project will be fine with any individual editor going - nobody individually is essential. That's the whole point of Wikipedia - what we produce is greater than the sum of its parts, and far greater than any single editor. Naypta ☺ | ✉ talk page | 16:52, 9 June 2020 (UTC)[reply]
  • The point of Wikipedia is solely, and entirely, to create and improve an online encyclopedia. Anything which does not work toward that end is superfluous. Anything which gets in the way of the end is counter-productive. Evaluating editors based on meritocratic principles helps the encyclopedia. Throwing out prolific content creators hurts it. That's really the bottom line. Beyond My Ken (talk) 18:33, 9 June 2020 (UTC)[reply]
    @Beyond My Ken: Creating and improving an online encyclopedia requires a community of editors who are willing so to do. "Throwing out prolific content creators" for being prolific content creators is not what anyone is suggesting; what's being suggested is that how "prolific" a content creator you are should not have any bearing on whether or not you are allowed to harass, abuse or otherwise be uncivil to other contributors. Having one rule apply to the majority of editors, and another to those who are deemed "too important to remove from the project", is not only actively unhelpful to the encyclopedia, as it drives away editors, but it's also fundamentally contrary to the ethos surrounding Wikipedia. Naypta ☺ | ✉ talk page | 18:37, 9 June 2020 (UTC)[reply]
  • The problem is that the "unblockables" get that tag primarily because they're annoying, not because they're actually "harassing" anyone. "Harassment" is a word that's very easily thrown around without actually fulfilling the requirements of WP:Harassment. We get a lot of "X used a bad word in their edit summary" or "Y called my editing incompetent" or "Z said I don't have good enough English to edit articles" being called "harassment". No one ever guaranteed that Wikipedia was going to be a place where your actions won't be critiqued or criticized, and many people have thinner skins then is needed for participation in a global online community consisting of people from many different backgrounds and cultures. (Anyone who thinks things are rough here should check out Reddit or, going back in time, Usenet.) I'm all in favor of cases of true harassment being appropriately dealt with, and every complaint should be investigated, but not every claim of "harassment" is actually legitimate, and at times the accuser is as much part of the problem as the accused. Beyond My Ken (talk) 18:56, 9 June 2020 (UTC)[reply]
@Beyond My Ken: Personally, I'm of the view that we shouldn't need to compare ourselves to the worst parts of Reddit. We can hold ourselves to a higher standard. There's nothing stopping us from doing so. You can raise an issue with someone without insulting or attacking them - it's really not that hard. There's no reason to attack other editors in edit summaries (or anywhere else) - "X used a bad word in their edit summary" should not be a complaint that's brought up, because "bad words" should not be used directed at other editors, ever. Calling someone's editing "incompetent" oughtn't be generally appropriate, either, although it's not as bad as the former; there are plenty of other ways to convey that same message that use less-charged terms. Saying someone "doesn't have good enough English to edit articles" definitely oughtn't be offensive, and there are plenty of templates that do this in a completely friendly way - consider, for instance, the Welcomeen-lang family.
I said "harass, abuse or otherwise be uncivil", and I meant it. It is my view that civility blocks should be far, far more common than they actually are - an order of magnitude so, perhaps - precisely because of this argument. Whether or not something fits WP:Harassment is not relevant to whether it's civil; personally, I have far more of an axe to grind with uncivil editors than I do even with common-and-garden vandals, because at least we unceremoniously block the latter - the former is an insidious sort of abuse that tears our community apart from the inside out. As I've outlined above, there are always options that don't involve incivility when bringing up issues with editors. Naypta ☺ | ✉ talk page | 19:25, 9 June 2020 (UTC)[reply]
I disagree with much of what you write, but I'll leve it at this: the "incivility crisis" is not what it's cracked up to be. Witness the fact that the editor who campaigned almost entirely on a "incivility is our biggest problem" platform when running for arbitrator was soundly defeated. It ain't. Beyond My Ken (talk) 19:32, 9 June 2020 (UTC)[reply]
(Oh, and BTW, telling an incompetent editor that their edits are incompetent isn't "incivility". WP:CIR, after all. Beyond My Ken (talk) 19:33, 9 June 2020 (UTC))[reply]
@Beyond My Ken: I wasn't active during the 2019 Arbcom elections, so I won't comment on the specific situation, but the statement "a majority of users did not vote for someone who ran on a platform of fighting incivility" is not remotely incompatible with the statement that "civility is an important and significant issue". As to WP:CIR, yes, but the same message can often be communicated in better ways. I think where we diverge here isn't over specifics of policy, so much as it is over communication strategy; the view which I am trying to put across is that we should not ever be using language which could be construed so as to offend where there is any viable alternative whatsoever. Naypta ☺ | ✉ talk page | 19:39, 9 June 2020 (UTC)[reply]
  • You realize that it's not always an editor's choice whether they end up at AN/I, right? That some editors go out of their way to file false or trivial reports just to get back at someone who annoys them. Beyond My Ken (talk) 22:44, 9 June 2020 (UTC)[reply]
  • You appear to be missing my point, many of these complaints have no policy basis whatsoever, they're based on "I am annoyed" or "I feel slighted". The subject has been dragged to AN/I for essentially no reason at all. Beyond My Ken (talk) 05:10, 10 June 2020 (UTC)[reply]
  • I'd happily provide you plenty of examples of AN/I reports that were clearly more than "I am annoyed" or "I feel slighted". However, I clearly understand why doing so would be a garbage idea. Therefore, I will just state, in no uncertain terms: I have seen AN/I reports like that, but 9 times out of 10 its against a relatively low-profile user. An AN/I report against an unblockable generally is more like "This user keeps following me around onwiki and insulting me" which gets closed with a warning or 1 day block (if lucky). –MJLTalk 06:57, 11 June 2020 (UTC)[reply]
If writing articles was really someone's priority then they'd find a way to keep themselves out of the drama boards as not to waste anyone's else's time. This would assume that no articles are sensitive, content disputes don't happen, editors who patrol shouldn't, admins are by extension unnecessary, etc. —PaleoNeonate07:54, 13 June 2020 (UTC)[reply]
(1) Content disputes are solved on article talk pages for the most part; That just isn't what AN/I is for. (2) Writing sensitive articles has nothing to do with frequently being reported for incivility. (3) The last two objections don't even make sense to me. Patrolling articles has nothing to do with writing them. Editors who do anti-vandal work are much more cognizant of their likelihood of being reported to ensure accountability for their actions. Admin actions also don't have anything to do with writing articles 90% of the time. –MJLTalk 16:23, 17 June 2020 (UTC)[reply]
  • This is not necessarily limited to "unblockables", but I have a concrete idea that might help certain kinds of situations. The way things are supposed to work when there is a dispute is that uninvolved editors form the basis of a consensus. A vulnerability of our system is that it's often not hard for someone to bring in enough allies to prevent a proper outcome. I once closed a 20-vs-10 RFC in favor of the 10, where the illegitimacy of the 20 was particularly blatant. It's a lot more difficult to disregard the votecount when a discussion is undermined by a stack of well established editors. When there is a repeating problem, I think the solution is to make sure we get back to a consensus of uninvolved parties. Previous participants would be allowed to comment, but the issue gets resolved by consensus of only previously uninvolved editors. If someone has been undermining the consensus process by calling in paid/COI/ideological-crusader/friends or other allies to vote opposite the norm, they will basically have burned out their ability to do so in the new discussion. Alsee (talk) 18:07, 7 June 2020 (UTC)[reply]
    • That won't work, and is too easily gamed. In a recent case, people claiming to be "uninvolved" showed up who were far more involved than the actual participants, and had involved people claiming their "involved" friends were uninvolved, when the "uninvolved but very involved" friends were there out of pure partisanship. And in complex cases, the uninvolved just might not get it at all. SandyGeorgia (Talk) 18:30, 7 June 2020 (UTC)[reply]
    Alsee, if we can set up and enforce a solid definition of what constitutes "involved", this sounds great. But that's the challenge. {{u|Sdkb}}talk 01:22, 9 June 2020 (UTC)[reply]
    Sdkb, what I had in mind was excluding people who had !voted in previous arguments on the same topic or in relation to the same person. To take a simple case, say there's a big ugly no-consensus AFD - if/when there's another discussion on it the new consensus wouldn't headcount previous participants. Another example: If there are repeated ANI issues with a user, you exclude people who !voted in previous cases involving that user. That way someone can't canvass the same X allies in 5 serial debates to successfully stonewall 5X random-editors who participate one time each. I don't think we would want to make this an always-active rule, and the the part I'm fuzzy on is a good system or standard for invoking it when starting a discussion. The person starting the process might specifically list the related discussion(s), which would indirectly establish a well defined list of who doesn't get to re-!vote in the current process. Alsee (talk) 07:36, 9 June 2020 (UTC)[reply]
  • There are no unblockables. The problem is that some admins react inappropriately and issue three-month blocks for a few expletives on the basis that the perpetrator is not respecting authority because they've done it before. The overreach means the disrespectful editor is unblocked. An editor who generally builds good content but who is abrasive should be blocked for 48 hours when necessary. Each repeat (for a different incident) gets another 48 hours. No one will unblock, and the problem will be solved. Johnuniq (talk) 02:37, 8 June 2020 (UTC)[reply]
  • I suspect some people are tired of hearing me say this, but I'd like to reiterate my position that a lot of these problems could be solved (or at least somewhat ameliorated) by the creation of a proper UserBlind mode, in which users could fully evaluate, comment on, and take action regarding conduct issues without knowing the identities of any of the involved parties. --Yair rand (talk) 04:49, 8 June 2020 (UTC)[reply]
    Yair rand, I like this idea a lot. The main challenge is that, alas, I don't trust that editors will resist the temptation to see who is behind the pseudonyms, and it would be very hard if not impossible to enforce it technically. {{u|Sdkb}}talk 01:17, 9 June 2020 (UTC)[reply]
    @Sdkb: It doesn't need to be enforced, and it doesn't even need to be verifiable. It just needs to become reasonably likely that one doesn't know the identities of the involved parties, and the interpersonal dynamics will shift dramatically. --Yair rand (talk) 02:29, 9 June 2020 (UTC)[reply]
    @Yair rand: I was actually thinking about this just yesterday - I'm so glad to see someone has actually made it! I would fully support this being not only suggested, but required, when dealing with complaints at ANI, for instance. Naypta ☺ | ✉ talk page | 10:24, 10 June 2020 (UTC)[reply]
    I think this is an excellent idea and would be useful for a wide variety of noticeboards. I think it had better be optional at first, but I can envisage a scenario where it's eventually mandatory at ANI. — Bilorv (Black Lives Matter) 20:10, 15 June 2020 (UTC)[reply]
  • I'm inclined to agree with Ballioni and Tryptofish here - this ARBCOM has already demonstrated they're not particularly of that view - and indeed has overegged it already to make it really clear. Additionally, definitions of unblockables is absolutely in the eye of the beholder. If Yair rand's idea can be made to work, sure, though I'm unsure how it could handle cases where people refer to parts of a username, and certain varieties of involved behaviour would be hard to identify. Nosebagbear (talk) 10:37, 8 June 2020 (UTC)[reply]
  • This behaviour is a bad behaviour is just like saying lawless to curb this we just have to take out a policy that will say no user is above blocking,ban and sanction. And the administrator must follow if not they should be removed. We are all equal. Trust is just the gap here. That is my own view. Tbiw (talk) 20:58, 8 June 2020 (UTC)[reply]
  • The problem is a real one, but I'd frame it rather as a question of what to do with users who refuse to modify their behavior despite any number of sanctions and warnings, and who instead rely on their long list of positive contributions to shield them from criticis. The number of users in this category is not large, but it's a very visible group; and our entire discussion about civility, and about behavioral standards for experienced editors has been framed around these editors. It's time we started applying the principle that no single editor is indispensable, and dealing with long-term behavioral issues the same way we would in someone with less experience. In dealing with run-of-the-mill disruption, often the most effective method is the use of TBANs. The framing statement above recognizes that this sort of intermediate sanction is difficult to enforce because of the perception that the "community" cuts these users more slack; and while that may be true, I don't see that a reason to change what is fundamentally the fairest approach. Vanamonde (Talk) 16:01, 9 June 2020 (UTC)[reply]
  • I agree with Amorymeltzer: Block them.
There are also systemic issues with the blocking policy: Because of the "second mover advantage", an admin is always able to unblock a friend of theirs, without an easy way to re-block absent positive consensus at AN/I, which the unblockable's friends will normally be able to obstruct. Discretionary sanctions blocks avoid this, but are not always possible, and in these cases the harassment often just refocuses on the blocking admin. With ArbCom as the blocker, these problems do not exist. Sandstein 17:08, 9 June 2020 (UTC)[reply]

At the root of the problem is that brusque, aggressive behaviour is effective at driving other editors away, and so there is an incentive to behave this way, either knowingly or unknowingly of the end result. Closers interpret rough consensus (linked to from Wikipedia:Closing discussions § Consensus) "by looking at strength of argument, and underlying policy (if any). Arguments that contradict policy, are based on unsubstantiated personal opinion rather than fact, or are logically fallacious, are frequently discounted." In combination with the principle that consensus can change and that closers should be evaluating the consensus reached by the participants, and not interjecting their own views on the strength of the arguments, closers often only make use of guidelines to the extent that the participants do. Accordingly, a small number of dissenters can block progress in any discussion, and guidelines only get followed if sufficient numbers of supporters participate in all related discussions. (It can be different people, but there needs to be enough in each discussion to re-establish consensus.) As a result, more often than not, the deadlock is only broken through attrition of participants.

If we want to curb uncollaborative behaviour, we need to resolve content disputes in a way that doesn't provide a motivation for it. Decisions need to be made in a more timely manner, and be less prone to reversal, so that editors can adapt to a settled consensus, rather than being rewarded for driving editors away. On the flip side, we need to be able to deal with recalcitrant editors effectively, so that others aren't driven to frustration. As noted by other commenters, some people aren't well suited to engage collaboratively on English Wikipedia; we need to identify them and gently ease them towards other pastimes. isaacl (talk) 00:27, 10 June 2020 (UTC)[reply]

  • No editor with systematic bad behaviour is so prolific that they are a net positive, because the more prolific they are, the more people they will have scared away from the website with their bad behaviour. If a user, across the span of 15 years, scares away just two individuals who would have gone on to be as prolific long-term contributors as the user is, then they have been a net negative. Often these two individuals will be two of the hundreds of people that the poorly-behaved user bit, causing them to leave after three edits because the interaction made them feel frustrated and hurt. Sometimes it's more malicious harassment and we can name several medium- or long-term editors who left the site out of frustration due to the user's actions. If a siteban is the only way ArbCom can respond to unblockables then they should issue more sitebans. — Bilorv (Black Lives Matter) 11:55, 10 June 2020 (UTC)[reply]
    Whereas this is very good and perfectly correct in theory, any attempt to realize this in practice meets an obvious obstacle which is that it is impossible to define what "systematic bad behavior" means. Whereas there are some things which are clearly unacceptable, many of them are in a grey area (different for different users) and what is acceptable and what is not depends very much on the background. Let me just give you a couple of personal examples, not in order to define an exhaustive classification, and not to dig up some old stuff, but just to illustrate what I am talking about. (i) For me, if someone says I am lying it is a very serious personal attack. Possibly the most serious one can imagine. I had people here, completely baseless, telling me that I am lying. On some occasions I took this to ANI, an on zero occasions this led to anything except for multiple users telling me that there is nothing wrong in telling me that I am lying, and that this not even a personal attack. (ii) Some time ago, I had a user who, for whatever reason, decided to entertain themselves by making fun of me. Any time I would show up at ANI/AN/AC/RfC they would show up in the same thread making personal comments about me, and sometimes explaining how I am a net negative for Wikipedia. A couple of times they went overboard and were warned, but continued. At some point they said something very offensive and at the same point completely stupid, and I overreacted quoting the last warning they got and putting it, let us say, not in the best possible context. Within minutes, I was warned by an administrator saying that one more time and I would be blocked. She did not tell anything to a user provoking me. That was I believe the only time I was seriously threatened by a block, and I took it seriously and stopped editing. In other circumstances, it might led me to complete retirement. This was on one of the dramaboards, and not a single person tried, if not to defend me, but at least to say that the behavior of my opponent was not perfectly ok. This happens all the time, consistently, and some people get lucky, but some just leave. Yes, it should not happen, but it happens.--Ymblanter (talk) 12:30, 10 June 2020 (UTC)[reply]
    @Ymblanter: That sounds like a really awful experience, and I'm genuinely sorry it's the one you've had. Isn't it an argument precisely for the strengthening and enforcement of these same rules, though? You say it's impossible to define what "systematic bad behavior" means, but I think whatever definition you go by, something like that ought to come under it. We've dealt with harder challenges together as a community than this: hell, we've built something that two decades ago would have been unimaginable. It doesn't seem like it ought to be insurmountable for us to be able to get along together as a community. Naypta ☺ | ✉ talk page | 13:01, 10 June 2020 (UTC)[reply]
    But this is exactly my point. It is ought to come under it, but what is the mechanism? Taking my first example, and not concentrating on details, we have two options: (i) a policy which literally says "Accusing anybody in lying against objections is a blockable offence" and (ii) WP:NPA, and each particular case to determine by consensus what is a personal attack and what is not. I tried to show that (ii) fails, presumably because of different standards. On the other hand, (i) is impractical (should it contain every single way I can swear on other users? Is calling someone a "troglodyte" acceptable?). My understanding this is exactly the idea of UCoC which WMF is promoting, and I am very sceptic about it just because of these considerations. (A few days ago, a user on the Russian Wikivoyage was blocked for a personal attack against me, by another administrator, and on their talk page he promised that once UCoC is out he will complain against us to WMF because we obviously have violated it - this is also a factor to consider, btw).--Ymblanter (talk) 13:17, 10 June 2020 (UTC)[reply]
    @Ymblanter: Naturally, language is subjective, and so decisions have to be subjective too. Compiling a list of "banned words" is pointless, I'm sure I don't have to convince you of that - so what we need, I'd argue, rather than to produce an exhaustive list, would be to have an independent panel of editors who work on civility complaints, preferably blinded to the names of the users involved. What one might be able to do is to produce a UTRS-style system for civility complaints, where recent contributions and relevant diffs can be included, with all involved usernames redacted, and impartial third parties - unaware of who the involved parties are - can make judgement calls on whether what was said was appropriate and civil or not.
    Is such a system perfect? No, of course not. There will always, no matter what we do, be cases in which the wrong call is made - one way or another. We shouldn't be making the perfect the enemy of the good, though; especially considering that the good - pretty well any change - is better than the abysmal way that we currently handle such things (all too often, not at all). Naypta ☺ | ✉ talk page | 10:32, 11 June 2020 (UTC)[reply]
  • The mere fact that we can't agree who the unblockables are, or rather, which currently blocked editors were once considered unblockable, shows one of the difficulties here. It has been suggested that as a meritocracy we should give extra tolerance to those who have done most, and for genuinely isolated incidents of misjudgment that makes sense. But it also makes sense that we have some failsafe, perhaps an ability for Arbcom to take any case where a block has been made and then reversed by another, or where there are issues of concern re an admin account - though presumably they already have that judging from some of the motions that have come out desysopping individual admins ϢereSpielChequers 18:39, 10 June 2020 (UTC)[reply]
    I am going to take a stance that is both parts cynical and charitable given the right interpretation. That is: we totally can agree on who the unblockables are, but many users are reluctant to accept a definition which includes people they like under it. –MJLTalk 23:51, 10 June 2020 (UTC)[reply]
    I can think of at least one person who I like and who some once considered to be unblockable, but who now has been blocked, and blocked more than once. Friendships do sometimes skew our judgment, and I wish there was more questioning about recusal in RFAs and ARBCOM elections. But my point about our inability to agree as to who the unblockables are is unrelated to that. There are some people who consider that the unblockables are admins and other functionaries, presumably on the assumption that seniority brings protection (I know it doesn't, but I know that some people argue that it does). There are others who consider that the unblockables are those who have carefully avoided becoming admins or other functionaries precisely because they don't want to be judged by the higher standards that the community applies to such, or to expose themselves to potential sanction by ARBCOM. A third group of people perceived to be unblockables are the WMF staff. It certainly used to be the case that if the WMF stood by a staffer ARBCOM would treat them as unblockable. In its current phase, where the WMF portrays itself as taking harassment etc more seriously than the community, it seems inconceivable that the WMF would stand behind a staffer who the community would otherwise have banned. But the WMF's attitudes have flipped on this completely in the past, and could presumably do so again, so I would consider this to be a third, and currently empty category of unblockables. That isn't an exhaustive list of all the definitions of unblockables that are "in play", but it is enough to show that we as a community don't agree who the unblockables are. We might all be able to agree that an unblockable person is someone who should have been blocked, but who for various reasons can't be blocked or is swiftly unblocked if they do things that others would expect to serve a block for. But I know that when we drill below that, such as to the three categories that I mentioned earlier, we as a community will not be able to agree on all three of those categories. ϢereSpielChequers 11:14, 11 June 2020 (UTC)[reply]
    I haven't been here that long but the only definition of "unblockable" I've ever heard isn't on that list, and it's an editor who does not get blocked for doing things that other editors routinely get blocked for (usually personal attacks), and who, when blocked, is quickly unblocked. Hence, "unblockable", because they never get blocked and when they do they never stay blocked. Levivich[dubiousdiscuss] 18:50, 12 June 2020 (UTC)[reply]
    People in that group eventually mend their ways or get banned by Arbcom. By the time they are thought of by some as "unblockable" they have long block logs and in reality they are in the last chance saloon. If we didn't have Arbcom then yes we would have that sort of "unblockable". ϢereSpielChequers 22:27, 12 June 2020 (UTC)[reply]
    Some editors spend years at the last chance saloon between "unblockable" status and Arbcom finality. Levivich[dubiousdiscuss] 04:25, 15 June 2020 (UTC)[reply]
    Any user worth their salt would avoid publicly naming someone else an "unblockable" because doing so would only invoke their wrath and ire. Generally, when someone is publicly known as an unblockable, it's because they are already close to being on the way out since the tide of public opinion has turned against them. –MJLTalk 16:17, 17 June 2020 (UTC)[reply]
  • I agree with Sandstein's comments above. The blocking policy is broken and gameable. Fixing that policy to remove the "second mover advantage" (while still providing some mechanism for first mover oversight) would go a long way towards resolving the issue of "unblockables". Kaldari (talk) 01:35, 11 June 2020 (UTC)[reply]
  • I have said in a few case statements that I think ArbCom should more actively take cases involving administrators. I still believe those statements, and believe ArbCom is, generally speaking, on the right track in handling this problem. I think the committee should also be more active in taking long-term, low-level issues especially when the community has shown an inability to deal with the problem. Maybe that's the crux; if it's clear that sending it back to the community will not resolve anything, take it up. Wug·a·po·des 01:26, 13 June 2020 (UTC)[reply]
  • Simple solution - we all know who the unblockables are, so when an unblockable has been filed against, recruit help from WMF T&S. This is what I alluded to earlier about working together to resolve difficult community issues that even arbcom cannot resolve. Work jointly and distribute the responsibility more evenly in an effort to reach a remedy that is less likely to be questioned or criticized as showing favoritism or bias. Atsme Talk 📧 19:34, 15 June 2020 (UTC)[reply]
Except it clearly isn't simple, because we absolutely don't all know who the unblockables are, because a) people define the term very differently, and, b) people view different people as falling into those (different) definitions Nosebagbear (talk) 09:57, 18 June 2020 (UTC)[reply]
  • The unblockables are Ymblanter's Group II - users who have a lot of wiki-friends. We are highly vulnerable to such people. With a sufficiently large number of minions, you can get and your minions elected an admins, and to ArbCom, delete or modify any article, change any policy, ban any user. It's a disaster waiting to happen, and the unblockables are but one symptom of it. We've seen over and over again ArbCom fearing to perform its role, citing adverse community reaction, even admins blocking arbs. We've seen a taste of what can happen in Gamergate. Hawkeye7 (discuss) 01:24, 26 June 2020 (UTC)[reply]
  • As several people have said above, block them. The question, then, is who will bell the cat? In cases where ArbCom has already become involved, the only appropriate answer is ArbCom themselves; any community-based approach will result in interminable debate around the letter and/or spirit of ArbCom's previous decision, whatever it was. If existing sanctions are not working, ArbCom should, by motion, open an ARCA discussion or full case (as needed) and widen, clarify, or escalate the existing restriction or sanction, up to and including a full siteban, if needs be. Waiting until things get so bad that the community opens a second (or in some cases, a third) case about the same user is a colossal waste of everyone's time and energy. In general, if ArbCom observes the community open multiple successive cases in relation to substantially the same pattern of misbehavior, they should consider that a learning experience; next time, take the initiative. It is not OK to simply throw general sanctions over the wall to the community and forget about them. --NYKevin 23:09, 29 June 2020 (UTC)[reply]
  • But we're talking about people who will not hestitate to block ArbCom, and have gotten away with doing so in the past. That's why ArbCom conducts discussions out of view, and why they feat to act. These people have to be dealt with by WMF. Hawkeye7 (discuss) 23:32, 29 June 2020 (UTC)[reply]
      • Blocking an arbitrator because of a disagreement with their decisions is textbook disruptive behavior, and should earn an admonishment at an absolute minimum. Perhaps we need to invent some form of formalized judicial immunity for ArbCom members, but honestly, I would expect this to be perfectly obvious to anyone who has read the blocking policy. Regardless, ArbCom has both the technical ability and the policy-based authority to enforce its decisions, and if there is some sort of widespread community revolt against that authority, they can appeal to the WMF as a last resort. Realistically, it should not come to that. --NYKevin 23:59, 29 June 2020 (UTC)[reply]
  • There's one other aspect about "unblockables" that needs to be mentioned. Consider the following hypothetical case: A criticizes B bluntly, perhaps even rudely. Other people reading this criticism then think, "If I said what A said about B, I'd be sanctioned, maybe blocked. But A gets away with it because A is an established editor, & I'm just a nobody." The label "unblockable" implies the existence of a class structure that can be threatening to those who don't see themselves as a part of Wikipedia because it is undefined. Just as "established editor" is an undefined label. -- llywrch (talk) 05:01, 30 June 2020 (UTC)[reply]
    • To repeat my earlier point, there are no unblockables. The problem is that some admins react inappropriately and issue three-month blocks for a few expletives on the basis that the perpetrator is not respecting authority. The overreach means the disrespectful editor is unblocked. There would be no problem if inappropriate outbursts resulted in a 48-hour block. Johnuniq (talk) 05:15, 30 June 2020 (UTC)[reply]
      • No, there are. Most editors do not expect and will not get the same lenient treatment. And the admin who has the hubris to overturn such a block is another unblockable, because he expects (correctly) that he will get special treatment too. Hawkeye7 (discuss) 06:07, 30 June 2020 (UTC)[reply]
  • We are (supposedly) here to write an encyclopedia - those who contribute much are given leeway because on balance the give more than they take. There are no 'unblockables' as such, just those whose contributions to the encyclopedia are recognized with leniency. Similar to any Western court system, which ArbCom is not but derives its decision-making functions emulate. Carlossuarez46 (talk) 23:01, 1 July 2020 (UTC)[reply]
  • This is actually the easiest question on here to answer: block them indefinitely, and make it clear that until they get with the program they will remain blocked. If they decide to contribute, they are required to do so on parole - arbcom oversees the edits and makes random checks to ensure that the party is staying the program. If they aren't, then they graduate to LTA cases and the admin corps is advised to keep them out by any means necessary. In line with this, a community requirement that ARBCOM must take a case if it reaches them a second time regardless of how it got there should be passed too becuase I'm sick and tired of watchingour weak kneed arbitration committee refuse to do what it was created to do and take on the heavy loads. TomStar81 (Talk) 10:11, 3 July 2020 (UTC)[reply]
    TomStar81, and how do you deal with WP:RANDY? There's a serious risk of losing very long-standing editors who lose their temper when faced with obdurate idiocy. If we don't handle both sides of that problem, then content gets worse. I can think of several dozen incidents where a well-known fire-breathing monster has had to run interference to prevent blocking of one of our best producers of high quality content. Guy (help!) 10:18, 3 July 2020 (UTC)[reply]
    JzG So we lose them. Who cares. One editors sunset is another editor's sunrise, and I rather a group of new editors who can abide by the rules than one extrodinary editor who has a god complex which he refuses to riegn in. If the rules apply to everyone, then god damnit apply them to everyone and stop letting these privalged jackasses run roughshod over the rules and regultions at everyone elses's expense. TomStar81 (Talk) 10:24, 3 July 2020 (UTC)[reply]
    Addendum: It sounds harsh, but blocking them seems to me the best way forward. I've had two cases and one ongoing LTA case that touch on this and dealing with these people who are asses but allegedly contribute meaningfully is frustrated by ARBCOM's apparent inability to do what it was set up to do and fix the problem. Our idiot arbitrators can't even take cares when years worth of evidence shows that they should. At this point, I feel if arbcom can't do it then just block them. If they want to contribute, they adjust their attitude, otherwise IMO we don't need'em. TomStar81 (Talk) 10:40, 3 July 2020 (UTC)[reply]
    @TomStar81: I agree fundamentally with what you're saying, but I definitely don't agree with the idiot arbitrator language, and I think you should strike that. That is definitely uncivil language, and not suitable for this RfC, or anywhere else on Wikipedia. Let's not become that which we are trying to prevent. Naypta ☺ | ✉ talk page | 11:07, 3 July 2020 (UTC)[reply]
    TomStar81, which LTA? Ping me via email if you porefer. Guy (help!) 11:50, 3 July 2020 (UTC)[reply]
    @ TomStar One of the problems of a policy of "requirement that ARBCOM must take a case if it reaches them a second time regardless of how it got there" is that Arbcom could never then decline any case however spurious or mistaken, without opening them up to some troll opening up a new case agaainst the same party. Taken literally, someone could open a case against another editor for "possessing a boring signature in one of the more interesting parts of the pedia", the Arbs would have auto accepted it as a second case against the same party. Presumably they would close it PDQ as "not April the first yet", so I'mm not sure your proposal solves the problem that concerns you. ϢereSpielChequers 17:23, 3 July 2020 (UTC)[reply]
    • @WereSpielChequers: One of the problems with not having arbcom forced to accept a case the second time around is that arbcom will do nothing in perpetuity. Take Carmaker1, Joefromrandb, and Fram as case studies for why we need this: Carmaker1 self outed as an employee of the auto industry, but his case was turned down even after questions were asked about his conflict of interest and his possible paid editing, the latter of which would be a terms of service violation if they refused to disclosed they were being paid to edit. Despite this relevation, the case was declined. Joefromrandb was declined the first time around despite strong evidence that the arbistration committee needed to look into this and they ended up taking it only after further issues of the sort id'd in a years long look back were brought up again - and in that case they didn't get joe's participation because he said what he had to say previously, which made for a kangaroo court since the accused didn't participate it made arbcom look foolish and overbaring. Fram came before the board on multiple occasions and never once did they take it until T&S took it and then they bitched that it should have been their turf, overlooking the fact that if they had taken the cases at any point prior to T&S it would likely have not come to this. Forcing ARBCOM to take a case works for everyone benifit, and we can ammend the when and how as the abuse of the system makes itself felt. TomStar81 (Talk) 18:48, 24 July 2020 (UTC)[reply]
      • @TomStar81 I am not familiar with two of those three cases, but I don't dispute that sometimes Arbcom's lenience the first time is followed by a second time. My concern is that an automatic policy of "ARBCOM must take a case if it reaches them a second time regardless of how it got there" Is that it is too easily gamable by trolls, and the test for that is not whether there have been cases where Arbcom declined the first time and a stronger case followed, but whether there are trolls out there who would take advantage of such a loophole. ϢereSpielChequers 20:28, 24 July 2020 (UTC)[reply]

No editor is unblockable. EllenCT (talk) 20:06, 6 July 2020 (UTC)[reply]

8. Relationship with T&S

edit

On 21 November 2019, the WMF announced that they would "no longer use partial or temporary Office Action bans." In the conversation on meta that led to that announcement, there were several concerns raised involving the relationship between the WMF and ArbCom, including questions of self-governance, appeals, behavioural expectations and standards, and the right of the accused to see the evidence and charges against them.

How can the Wikimedia Foundation and the Arbitration Committee better work together to eliminate duplication and increase transparency and trust, while meeting both the needs of the community and the legal obligations of the foundation?

Discussion
  • I don't know what will work to address this problem, but I can say one thing that won't work-- electing an ArbCom that holds WMF, WMF employees, or WMF board members above the standards of the rest of the community. SandyGeorgia (Talk) 15:12, 6 June 2020 (UTC)[reply]
  • To me, the WMF's hamfisted and tone-deaf handling of that matter suggests that the WMF should go back to providing technical support and server space, and leave the governance and direction of en.wiki to the en.wiki community.—S Marshall T/C 15:13, 6 June 2020 (UTC)[reply]
  • See my user page. Over the years, there has crept in an attitude that WMF is above the community instead of being a partner. That is a systemic attitude problem that the community can't fix, only the leadership of the WMF can, over time. We are not "the little people" and many of us resent the attitude of the WMF when dealing with community members, for good reason. Most of the interactions I've had with WMF have left an incredibly bad taste in my mouth. I haven't had that with Arb, even when I've disagreed with them. Dennis Brown - 15:17, 6 June 2020 (UTC)[reply]
    • re Dennis Brown could you elaborate further here on why you believe that the community cannot address this systemic attitude problem, which is a real problem? Because I would point out that there is one WMF person that I know of who does not fit the mold, and we should be finding ways to have More Like Her. Surely it is somehow doable. SandyGeorgia (Talk) 15:25, 6 June 2020 (UTC)[reply]
      • I refer to WMF as an entity, and there are probably many fine individuals within that entity. The ham-fisted approach to Fram is a good example, which demonstrated they can do whatever they want, and it took a virtual online riot to get them to hand off to Arb instead. They have all the power, they can lock the whole place down, or black out for a day, or do any number of unwise things we simply don't have the power to do or stop, so any change must start with the ones with power, not the community. Again, my user page explains my perspective. Dennis Brown - 15:30, 6 June 2020 (UTC)[reply]
        • Thanks, Dennis Brown. Although your user page states that We are not subjects of the WMF. We do not work for you. We are not beneath you. We owe you nothing; we are volunteers. We are trying to work WITH you to create something special, but you have to give us the same respect you demand for yourselves. this is increasingly less and less true. With limited exceptions, it is WMF employees who have made editing here a rather miserable experience, with increasingly less concern about accuracy in content or adherence to en.Wikipedia policy. And I guess you are right there is little we can do about that, except vote with our feet by retiring, quietly leaving, or stopping editing. SandyGeorgia (Talk) 17:13, 6 June 2020 (UTC)[reply]
          • I think we are more or less on the same page. I had handed in my bit then after a while came back, and honestly, I'm not extremely active, because of the WMF. I'm trying to be open minded and see if they turn it around, stop trying to dictate to the community on things the community can handle, but I'm not filled with hope. I'm happy to be a volunteer partner, but I'm not willing to be an unpaid employee. Dennis Brown - 18:08, 6 June 2020 (UTC)[reply]
          • I appreciate that has been your experience. Personally, the editors that have made my experience a poor one are not Wikimedia Foundation employees. So while I have no issues with trying to improve that aspect, there are plenty of structural issues that are entirely within the purview of the community to manage. isaacl (talk) 23:05, 6 June 2020 (UTC)[reply]
          • WMF have basically made the same "mistake" about five times, over the years. I put "mistake" in quotes, because the thing they want to do is never an unmitigated bad idea, and also not obviously wrong - in some cases it may even have been right.
          • The mistake is - as Dennis says - to be ham-fisted about it. We are all ham-fisted sometimes, sure. But repeating this mistake - is just not good.
          • And I'm sure Issacl is correct to say that editors cause other editors more grief than the WMF does. I think though that they also provide more help and support. Also there is at least the prospect of reigning in bad actor editors. The WMF is, in some respects, and immovable object, or at least as maneuverable as an oil tanker.
          • All the best: Rich Farmbrough 16:35, 15 June 2020 (UTC).[reply]
  • For the most part, I agree with S Marshall. With few exceptions, incidents should be addressed in-house (within the scope of the English Wikipedia), including but not limited to the role of the Arbitration Committee itself. El_C 15:19, 6 June 2020 (UTC)[reply]
  • When someone is calling in bomb threats on Wikipedia, stalking them in real life, or a pedophile, T&S should take over. ArbCom can ban locally as an initial measure, but per the rest of my comments here: they are not a court and have no power or ability to help other than by blocking. Let professionals deal with things that could involve law enforcement. TonyBallioni (talk) 15:29, 6 June 2020 (UTC)[reply]
    That is nice, concise way of putting it. ANI/Arb should handle everything else. Dennis Brown - 15:32, 6 June 2020 (UTC)[reply]
    For sure, I also agree with Tony that T&S has a key role to play in these sort of extreme cases. El_C 15:34, 6 June 2020 (UTC)[reply]
    Well, I have walked this walk twice, and there is NO CHANCE I would call in anyone from T&S, since it is about "trust", and the people I trust are not there. We can deal with the police for the real life aspect, and the arbs for desysopping and blocking, and trust them to handle it expeditiously and confidentially. SandyGeorgia (Talk) 17:15, 6 June 2020 (UTC)[reply]
    That’s fair, and I don’t think anyone would say you should be forced to (or at least I wouldn’t say that.) I know of at least one case where they were helpful connecting a user with the FBI (told to me by the user in question), so I think they can be a useful resource, but I acknowledge that after FRAM many in this community are skeptical of them. That should be an individual choice as to whether to reach out to them as a resource. TonyBallioni (talk) 17:20, 6 June 2020 (UTC)[reply]
    I agree with this as well. T&S should also have a role in dealing with cases involving serial ban evasion. MER-C 18:31, 6 June 2020 (UTC)[reply]
    Per MER-C, more T&S action against serial upe and extreme crosswiki harassment would be useful. Moneytrees🌴Talk🌲Help out at CCI! 02:20, 7 June 2020 (UTC)[reply]
    Indeed, what if WMF was the one to complain to the ISPs in such cases instead of some random volunteer? I suspect that the complaint might get more traction. --Rschen7754 02:22, 7 June 2020 (UTC)[reply]
  • Only one real comment here, relating to WMF/T&S and the Fram case. There are some excellent people on the T&S team, but what that team suffered was a failure of leadership. Bodies like that need leadership that is open to input, to listening, and to discussion. Dictatorial leadership, which is what we saw, is a barrier to any kind of proper interaction between WMF/T&S/ArbCom/Community. So that's something WMF needs to get right. Boing! said Zebedee (talk) 15:44, 6 June 2020 (UTC)[reply]
  • That's what my response above means; when one is threatened, the last people one will call in is people one cannot trust because of repeated demonstrated competence issues. If my only option way back then had been to call in T&S, I would have left Wikipedia instead. I trust people who have demonstrated competence via editing with me in the trenches, to an extent that I know they won't mess up when someone's life is being threatened. SandyGeorgia (Talk) 17:18, 6 June 2020 (UTC)[reply]
  • To answer the question about the relationship between Arbcom & TS: Last year, before the contretemps, Arbcom had the meeting with T&S, some communication over the issue occurred but evidently not enough. What should have happened in that affair is what should continue to happen, regularly meet with each other over matters, communicate, and discuss them. Since this RfC is being run by Arbcom, my message to Arbcom is to do better at those things and get T&S to do better. Alanscottwalker (talk) 17:00, 6 June 2020 (UTC)[reply]
    Several years ago, there was a discussion I think on the Arbcom pages about, inter alia the Terms of Use. One strain of thought, there, if I recall correctly was that despite the fact that we are all bound by the Terms of Use, including every Arbcom member being so bound, it was posited that only the WMF should be enforcing the Terms of Use on English Wikipedia. That does not seem to square with some of the more adamant, no WMF involvement positions. In every post we make on this website, we agree to be bound by the Terms of Use. Alanscottwalker (talk) 10:55, 7 June 2020 (UTC)[reply]
    @Alanscottwalker: Regarding your first point, ArbCom has a video meeting with T&S every month. Without saying too much about the content of those meetings, one regular agenda point is a list of potential upcoming office actions that are relevant to enwiki. Before FRAMBAN, I think most of us took for granted that anything mentioned there must be so severe that the action would be uncontroversial (I certainly did). That was a major failure on our part, and had we been more on the ball we could possibly have avoided the whole thing. Post-Fram, I think we have been better at regularly attending the meetings and applying scrutiny, and I agree that it's very important that that continue. We should also give credit where credit is due to T&S, who have always been diligent in passing on information and giving us the opportunity to comment on office actions before they happen. – Joe (talk) 14:36, 8 June 2020 (UTC)[reply]
  • At the risk of derailing this discussion, it's not just T&S, it's all of WMF. Evidence: superprotect, WP:FRAM, Visual Editor, Universal Code of Conduct, IP Editing: Privacy Enhancement and Abuse Mitigation. Unfortunately, to fix this problem, an entire culture change has to take place at WMF.
    That all being said: arbitrators need to be willing to call out bad behavior on the part of the WMF, in private, but sometimes in public as happened last year. And in converse, if WMF believes that there are issues on a particular wiki, they should feel free to bring them up to the community for discussion. And yes, legal matters should continue to be referred to WMF. --Rschen7754 18:57, 6 June 2020 (UTC)[reply]
  • I strongly agree with Rschen7754, because the WMF culture (and most of its staff/employees/board whatever they are called) is a very serious problem. But, we also need to elect arbs who will stand up to WMF abuses rather than avoid them. They went a good distance this year in finally dealing with some abusive admins, but then seriously punted when real action on a serious case was needed. SandyGeorgia (Talk) 20:25, 6 June 2020 (UTC)[reply]
  • I've recently started to warm to the idea of more active involvement by T&S, which is something that I'm surprised to hear myself say, because I sure did not think that way circa Framgate. I think T&S are correct that en-wiki has become ineffective at dealing with "toxic" behavior. Make a clear-cut personal attack, a la calling someone something that is vulgar, and we usually deal with it. But engage in prolonged dismissiveness, and both ArbCom and the community have become too accepting of that. Looking at the long back-and-forth sections of ANI where nothing gets resolved but a lot of angry words get said, it's reasonable to ask whether there is an institutional failure going on. Not that I believe T&S to be competent to fix that sort of thing on big, established wikis. But I'm in favor of a sort of collaborative approach between ArbCom and T&S. ArbCom should have jurisdiction here for anything that does not have to go to WMF legal, and T&S should back off. But T&S can still give advice to which ArbCom should listen. I'd like ArbCom to consider the kinds of "toxic, incivil" conduct that T&S talks about, and take it more seriously. The old essay at User:Kirill Lokshin/Professionalism is actually more timely than ever. --Tryptofish (talk) 23:16, 6 June 2020 (UTC)[reply]
  • T&S should become only involved where either interaction with law enforcement is likely to be necessary (child predators, threats of harm and violence), where they are literally required to (formal legal processes/court orders), or in the one instance it has widely been considered acceptable for them to do so (severe cross-wiki abuse). Beyond that, the community should handle most problems, with ArbCom there to handle issues which either involve private/off-wiki evidence or where the community has proved unable to resolve the problem. The ArbCom should remain a fully independent body from T&S. Or in short, basically "How we do it today; keep doing that." T&S is, of course, welcome to provide advice to ArbCom or the community (after all, anyone's free to do that), but whether to accept or reject such advice should be solely ArbCom's/the community's decision. Beyond offering input, T&S should never become involved in local governance or local enforcement questions. Seraphimblade Talk to me 03:11, 7 June 2020 (UTC)[reply]
  • I agree with Seraphimblade above me in toto. I do not want to see more T&S involvement here outside the areas which Seraphimblade lists, as I do not trust their judgement, and there is no way for us to have influence over their makeup, as we do with ArbCom. Beyond My Ken (talk) 06:32, 7 June 2020 (UTC)[reply]
  • I strongly agree with Tryptofish. T&S and Arbcom need to work in partnership with each other, with Arbcom referring matters to T&S that it cannot or should not handle for whatever reason (legal issues, child protection issues, cross-wiki abuse, etc) and T&S referring matters to Arbcom that the latter is better placed to handle. Both should refer to the other for non-binding advice and opinion. Thryduulf (talk) 13:16, 7 June 2020 (UTC)[reply]
  • T&S staff members have no place on ENWP and should be banned from any involvement in any aspect of its administration or interaction with its editors or Arbcom. The WMF culture of gravy-train activism are completely at odds with ENWPs culture. The current T&S team contains a number of individuals who have repeatedly shown they lack judgement even in the area in which they are supposedly employed (check their CV's and do some online digging into their past employment if you want the gory details). WMF staff members have a basic conflict of interest in that their personal employment is dependant on doing what their bosses say, regardless of the actual facts in the case. We have seen how that pans out with how they jump when WMF insiders make spurious complaints about ENWP editors justifiably criticising them. There should be no co-operation with the T&S team until they are completely replaced. And we need to elect arbitors who have a spine to stand up to the WMF rather than rolling over and showing their belly. Only in death does duty end (talk) 14:40, 7 June 2020 (UTC)[reply]
  • This is vaguely offtopic, but a real undercurrent is that, given enwiki's outside size and presence among the hundreds of WMF projects, all parties — WMF, ArbCom, our community, other communities — are forced to simultaneously recognize that 1. we're not special and 2. we're special. Any coordination, trust, or structure has to struggle with that. ~ Amory (utc) 15:57, 7 June 2020 (UTC)[reply]
  • I support Only in death on this with the exception of the research into T & S staff, which I am unprepared to check for its validity and therefore do not use as part of my argument. The WMF are not our bosses, but paid editors of a particularly inept kind with a declared policy of forcing bias into the encyclopedia (the rationale behind their new global behavioral guidelines). T & S have mismanaged the one task for which their existence is justifiable, dealing with report-mandated illegalities such as pedophilia, and in the Fram case showed contempt for the community and perpetrated a serious injustice. The WMF deserves nothing but our scorn and any collaboration with them, especially with T & S, is reprehensible. Yngvadottir (talk) 17:12, 7 June 2020 (UTC)[reply]
  • Totally agree with Seraphimblade, Only in death and Yngvadottir above. What they said.Smeat75 (talk) 17:29, 7 June 2020 (UTC)[reply]
  • Those who followed the in-depth research regarding last year's debacle are aware that T&S was utilized as a tool to silence one of our volunteers who had justifiably criticized the content work of an editor that had WMF ties. That's what happened, even if nobody will ever publicly admit it. As you can imagine, those of us who are aware of the details have zero trust in the individuals responsible for what happened. LEPRICAVARK (talk) 17:34, 7 June 2020 (UTC)[reply]
  • This section is pretty close to unanimous. I will agree with and perhaps go beyond the strongest of the comments above. The Foundation ran a "consultation" with the community where it was agreed that Trust&Safety would stay OUT of low level content squabble claims of harassment. The Board announced a reversal on this, T&S is gearing up to wage a CodeOfConduct war against global consensus, and we are talking about how to better trust them?!?
    That is just the tip of the iceberg. The Foundation-rebranding project had over 90% opposition to a 'Wikipedia' based branding, and they just announced that they are going advance three Wikipedia-based proposals to chose from. They are working on a Masking project which would continue to allow IP's to edit but conceal the IP address from us, making it more difficult to deal with IP abuse. The response was clearly an informal consensus against it, but apparently they plan to go forwards with it anyway. In 2011 they published a strategy to deprecate wikitext - the current strategy is still to eliminate our wikitext engine (called Parser migration project) and apparently still hoping to shift us all into VE with VE's interpretation of a "wikitext mode", and of course still trying to move everything to a VE default. After 6 years the Foundation finally admitted Flow was dead, but the new replacement project (Discussion Tools) basically cloned Flow's back end wikitext corruption design. The Foundation's "answer" to the Superprotect incident was to create a Product Guidance document doubling down on the original issue - declaring that the community is generously allowed to request early deployment of a product and that is the end of that. The Global Strategy process was atrocious - the final document is vague and watered down but it appears certain to spawn multiple crises. I fear we are in the calm before many storms break simultaneously. I fear the best way to restore trust in the WMF is that some day we will have to seek new hosting, and pressure the WMF to transfer the Wikipedia trademark. Then we can happily trust the Foundation to do absolutely anything it wants. Alsee (talk) 19:15, 7 June 2020 (UTC)[reply]
    A new hosting is a non-starter. We need to sort this conflict out inside the existing configuration.--Ymblanter (talk) 19:30, 7 June 2020 (UTC)[reply]
    Don't be so sure. Enciclopedia Libre is still around, but just the threat of so many leaving for it made even the idea of allowing advertising toxic and untouchable. If things get bad enough, telling them "Fork off" may be the only viable resolution. If enough volunteers leave for the new project, they'll come around pretty damn quick. The English Wikipedia only has value insofar as its volunteers are willing to support it. If we show we no longer are because of what WMF has done, WMF suddenly has an existential threat on their hands, and they will not be able to ignore it. Seraphimblade Talk to me 20:51, 7 June 2020 (UTC)[reply]
    The issue with a fork is that it would need a critical mass to move. That isn't going to happen because, for however loudly they shout, there is only a small minority of active users who would do so - and more than one of them are not actually very active in improving the encyclopaedia. Thryduulf (talk) 21:21, 7 June 2020 (UTC)[reply]
    And, in addition, Wikipedia 2.0 will need hosting and legal protection from the start. And it will probably would like to collaborate with external parties, so it will need to have people talking to these parties. And all these people will need to be paid. Which means we will need to have finances and fundraising. And a HR. And suddenly we are up with WMF 2.0, and I do not see a single reason to think it would be better than the current WMF.--Ymblanter (talk) 21:27, 7 June 2020 (UTC)[reply]
    Don't be so sure. A lot of these assumptions rest on a preference for complicated bells and whistles. The old editing software, without all the poorly designed Viz Ed crap and counterproductive template madness and Wikidata tie-ins, is free to use. The project does not need to support a very poor software development company and political action group, its featherbedding executives, or its conventions and other junkets. @Thryduulf: If that means me, a comprehensive apology from the WMF would cause me to reconsider the cutback in my activity; as it is, I can't square with my conscience creating new articles that they will claim were created for them and their "movement". The Foundation is the problem, and those of you wishing to appease them are assisting in undermining the project. In particular T & S, but they owe us an apology for much prior to Framgate. Alsee's list is a good start on reminding those who still assume we have to work with them. Yngvadottir (talk) 22:31, 7 June 2020 (UTC)[reply]
    My point is that it is not possible to run a project of this size without any supporting organization at all. And we all remember that this organization, which is now WMF and wants to rebrand into MPF, started with less than a dozen of people and in the beginning it was fully aligned with the community. And then things developed as they developed. Now you would need to start with at least several dozens people which will not necessarily would be aligned with the community from the start. If we could not make sure the previous development going right (and I have never seen a comprehensive analysis on how it could have been done differently), the chance of it going right from the second attempt is zero. And, indeed, in addition most editors do not even know what WMF is.--Ymblanter (talk) 05:51, 8 June 2020 (UTC)[reply]
    Yes a fork requires a critical mass, and also some features of the WMF, lawyers devs etc. But the reason for a fork is not to do without a central organisation, but to have one more aligned with the community and less with silicon valley. Host it at least a thousand miles from silicon valley and it will be a plant growing in very different soil. I doubt that this will be the year for a fork, money is tight and will be for a year or two. However the faultlines seem deep to me, and with the cost of hosting falling and the "respectability" of Wikipedia rising, it is probably just a matter of time before a fork happens. If the fault lines follow languages we may even be able to keep things somewhat cordial. But I don't see the WMF relinquishing the trade name Wikipedia. The fork will need a new name, or a revived one. ϢereSpielChequers 15:56, 11 June 2020 (UTC)[reply]
    There's not going to be a fork coming out of this. Not now, and not in any foreseeable timescale. And we all know it. So why add even more words talking about something with no chance of any relevance to this RfC whatsoever? Boing! said Zebedee (talk) 16:14, 11 June 2020 (UTC)[reply]
    Because the current and future relationship with T&S is one of those areas that could either widen the rift between the WMF and the community or, if we get this right, bind the two more closely together. ϢereSpielChequers 16:42, 11 June 2020 (UTC)[reply]
    Talking about forking is going to do absolutely nothing to help - T&S know the idea is nonsense just as much as everyone here should. Anyone who wants to improve the relationship should try to stick to reality. Boing! said Zebedee (talk) 19:20, 11 June 2020 (UTC)[reply]
    Since the WMF was able to host WikiVoyage when the WikiTravel community decided to fork, I doubt they are as sure this community won't fork as you are. ϢereSpielChequers 16:17, 12 June 2020 (UTC)[reply]
    And this great reading actually strongly suggests that forking would be technically impossible (the conclusions are towards the end).--Ymblanter (talk) 13:18, 25 June 2020 (UTC)[reply]
    It would be more efficient to get rid of the problems than to walk away from them. Start with sorting out the board. Ultimately they are responsible. · · · Peter Southwood (talk): 14:12, 9 July 2020 (UTC)[reply]
    How do you propose the English Wikipedia Community and/or Arbitration Committee should go about "sorting out the board"? Boing! said Zebedee (talk) 14:31, 9 July 2020 (UTC)[reply]
    I'll argue that a fork of en.wikipedia is very possible: we just need to have a plan, & not try to do too much too soon. Pick a name, some quick rules -- or just copy the parts of Wikipedia's policies that we agree on. Recruit 30-40 of the most productive content creators: right now less than 400 people are contributing 90% of the content. Find a donor to contribute hardware & bandwidth: for years Wikipedia had only three servers & maybe a 1 GBit connection. Take one of the lists compiled of vital articles -- say the top 1,000 or 10,000, while excluding any biographies of living people in order to avoid potential defamation lawsuits -- & concentrate on making those articles FA quality. Once the fork is proven to be viable, start expanding coverage & invite people to join. Momentum will follow & Wikipedians will start migrating. -- llywrch (talk) 06:31, 30 June 2020 (UTC)[reply]
    "Find a donor to contribute hardware & bandwidth" - yeah, that'll be the easy bit, and we only need three servers and 1Gb. And then persuade Google to drop the original Wikipedia as its top search result and use the fork instead. Get real. Boing! said Zebedee (talk) 06:37, 30 June 2020 (UTC)[reply]
    I disagree that it would be anywhere near as difficult as you make it out to be, but this RFC is about anti-harassment measures, and I'd rather not take up more space here on this considerable tangent. --Yair rand (talk) 06:58, 30 June 2020 (UTC)[reply]
    It's one of those things where rather than argue about it, if people think it "is very possible" and it's a good thing to do, just do it! We'll see everyone there if it succeeds. isaacl (talk) 09:09, 30 June 2020 (UTC)[reply]
    I didn't mean to hijack this discussion, but I believe it's worth emphasizing a fork of Wikipedia is much easier than it might appear at first glance. Companies like Intel have programs for donating old hardware to non-profits; there are also programs for donating bandwidth. Research it if you don't believe me. The trick is to start small & be focused so not to bite off more than you can chew. There are more resources in the Wikiverse than Wikipedia & Fandom (website). (And as I thought about it afterwards, our generalized survey articles are a weakness, IMHO due to poor morale: there is little interest in developing such articles, which have a higher chance of conflict with other volunteers & require more work for at best disinterest from the Foundation & others, when working on specific & narrow topics are more satisfying.) -- llywrch (talk) 13:56, 30 June 2020 (UTC)[reply]
    Don't you think it's a bit late for a Wikipedia fork to start small? Boing! said Zebedee (talk) 14:25, 30 June 2020 (UTC)[reply]
    Physically forking the server software and database to a small cluster? Sure. But as everyone participating in this discussion knows, the project is more than the hardware and software. Forking the community, or building a completely new one from different editors, is a much harder thing to do, and scaling it to a sustainable level that will subsequently grow (assuming that is a goal) is the trick. But again, there isn't much point in arguing about it in theory. I like it when good things happen (we could really use some this year), so it'd be great if interested parties could work together and deploy a fork, if they fervently believe it is a desirable path to take. Let us know where to go and I'm sure some people will at least take a look! isaacl (talk) 20:55, 30 June 2020 (UTC)[reply]
    I doubt the fork will happen this year, everyone's budget is tight or at least uncertain. But with Wikipedia's reputation on the rise, the opportunity is there for any moderately wealthy ambitious university to host such a fork and give it a "respectable" university based brand. Give it an editing interface that works moderately well on tablets and smartphones, fix a couple of other problems that Wikipedia has, and time the launch for another low in WMF relations with the community and it should have a fair chance of success. ϢereSpielChequers 18:54, 1 July 2020 (UTC)[reply]
    I'm not sure what would be the PR selling point for a university to forgo supporting Wikipedia for another crowd-sourced encyclopedia that would make sense to the general public who doesn't care about internal politics. The new project would have to bring something new to the table for a university to get involved: perhaps better integration across different language versions, or focus on areas not currently well covered? Given the amount of commitment required, I don't think it would be a good idea to base a rollout schedule on any Wikimedia Foundation miscues; it should just launch based on known considerations. But whoever wants to do it can work out the details. isaacl (talk) 20:24, 1 July 2020 (UTC)[reply]
  • Behavioral discipline should be delegated to the communities, not handled centrally and bureaucratically. There are nothing but problems on the horizon if T&S insists upon intervening in day-to-day operations of En-WP. Carrite (talk) 00:07, 8 June 2020 (UTC)[reply]
  • T&S and the Board have just demonstrated that the aggregate consensus of all interested editors was irrelevant and continued on with a UCOC. The same applies for masking, and branding and so-on. The WMF has demonstrated repeatedly flawed judgement, especially on interacting with the community. Engage on more issues, early on, the only time they can be changed. Then engage publically where necessary. Focus on a need for individuals to see all evidence except in harm and similar cases. Nosebagbear (talk) 10:42, 8 June 2020 (UTC)[reply]
  • Just FYI: Youtube, head of the WMF area that includes the T&S department taking questions, including on T&S and the universal code of conduct: Maggie Dennis, VP of Community Resilience & Sustainability last week. Alanscottwalker (talk) 15:04, 9 June 2020 (UTC)[reply]
    It is useful to note that this is the same person as Moonriddengirl--Ymblanter (talk) 15:17, 9 June 2020 (UTC)[reply]
    Notes are available at m:IRC office hours/Office hours 2020-06-04. --Yair rand (talk) 17:33, 9 June 2020 (UTC)[reply]
    One particularly interesting quote: The “bright line” with harassment can sometimes become indistinct both for us and for local governance bodies, who do sometimes directly ask us to take on cases they’ve been asked by their communities to review. For instance, we all agree I believe that where physical harm is involved or illegal actions, the Foundation has a role. What about harassment that leads to suicidal behavior? What about off-wiki doxxing? Blackmail? One of the roles of the community review process we’re setting up, both in the interim and in the long term as part of the UCoC process, is to determine if the Foundation’s involvement in a case was appropriate. It could result in some cases being overturned and remanded to communities, and we are comfortable with that. --Yair rand (talk) 17:46, 9 June 2020 (UTC)[reply]
  • I do not like the WMF. I don't like their size and scale; I don't like their deceptive advertising banners on our website; I don't like that we have no say in how the WMF is run. Nonetheless, out of the Wikipedia community, ArbCom and the WMF, I don't know who I trust the least. Perhaps if ArbCom acted more firmly, and had a sensible process that encourages people to come forwards when they are being harassed, and does not punish those users for doing so, then WMF would not feel the need to overcompensate for ArbCom's incompetence. When the WMF feel they need to step in and act then they are not the root cause of the issue. (As for the technical side of things, I don't see it relevant to a discussion on anti-harassment. But here too the community is incredibly overreactive to any decision taken by the WMF, even when it is a good one.) — Bilorv (Black Lives Matter) 12:13, 10 June 2020 (UTC)[reply]
  • and do "they" like the goony mob? Do they like its/their size and scale(s)? Do they like the ways it/they whale(s)?
When I read stuff like "Your edit was complete trash" on my watchlist said to a long-term contributor, with plenty of other reasonable contributors disagreeing I can't help but think that careful deprecator should at the least be forcefully reminded that this style of talk-page and edit-summary behavior is inappropriate. And I know it won't happen. It was that particular trash-talker who brought me to see the urgency of dealing with Wikipedia though; so, maybe I could argue they're not a net negative (if I had an overly inflated sense of self-importance or thought I wasn't alone in being drawn in like that). One possible "movement" strategy... deploy irascible gatekeepers who insist on their prose, so that people will always be attracted to the fight. Churn & repeat. ^^ -- SashiRolls 🌿 · 🍥 15:10, 10 June 2020 (UTC)[reply]
  • Like many here, I can remember a day about a decade ago when the values at the WMF were very different. A day when the WMF stood by one of their staffers who was rightly desysopped by Arbcom, but not banned because the WMF intervened and the then CEO stood by him. We are now in an era where the WMF wants to be seen to be tougher on harassment than the community, but is not always clear as to how they want to do that, or how to do that fairly and wisely. The old description of the WMF as a bunch of twenty somethings talking to a bunch of greybeards as if we were adolescents has more than a grain of truth in it. There is a theory that nothing has really changed, and that the common factor remains the WMF support for its people against the community. But I for one welcome the WMF of recent years wanting to take harassment and incivility at least as seriously as the community does. However I would want to see a bit more thoughtful behaviour from the WMF before I accept that the leopard has truly changed its spots. As part of that I would like to see from them a commitment to evolutionary change rather than revolutionary change, and by that I mean lots of small changes that we can agree are positive, instead of a preference for major changes and a willingness not to carry the community with them. Two examples I will be watching the WMF for: Replace phabricator with a wiki based system that non techies are comfortable reporting problems on, and recognise that many of the people who report being bitten on Wikipedia were driven away by edit conflicts due to WMF software, rather than editing by volunteers. ϢereSpielChequers 17:52, 10 June 2020 (UTC)[reply]
  • These are good points. One further observation: someone associated with the Berkman Center wrote into one of the five disclaimers (WP:RISK) that Wikipedia was addictive. It might be worthwhile, considering duty of care, that Trust & Safety develop a bot to temporarily block administrators/non-admins who have not taken a half-hour break from editing the site in over 24 hours. (A recent example was 26 hours and 50 minutes without a break; I have no idea if this is a record.) This would protect not only the administrator from themselves, but also those who may have to interact with that admin, due to their misuse of tools during the binge / super-human shift. -- SashiRolls 🌿 · 🍥 03:23, 15 June 2020 (UTC)[reply]
@SashiRolls: that would be an odd change but perfectly reasonable, the effect of being awake for 24 hours is about as disruptive to cognitive ability as 3 cans of beer. Were an admin to announce that they'd been tipsy when they made loggable actions, the Community would legitimately concerned. Nosebagbear (talk) 22:02, 28 June 2020 (UTC)[reply]
SashiRolls, Nosebagbear - no sleep or asleep at the wheel? Atsme Talk 📧 19:41, 29 June 2020 (UTC)[reply]
  • I think Arbcom should take the position that conduct exclusive to the English Wikipedia is its domain, it is ready and able to police serious conduct, and T&S's intervention would be counterproductive.--Wehwalt (talk) 22:32, 10 June 2020 (UTC)[reply]
    • At least for the time being, I do too. But I think it's a two-way street, and that the assumption of responsibility comes with, well, responsibility. ArbCom needs to deliver on decreasing what T&S has (somewhat clumsily) called "toxicity", or WMF will be justified in reassessing the balance of power. Hopefully, this RfC marks a good step in the right direction. --Tryptofish (talk) 22:44, 10 June 2020 (UTC)[reply]
  • T&S needs to stick strictly to bright-line issues - threats against people/property, legal matters (that haven't been telegraphed), anything that's explicitly illegal (presuming it hasn't been dealt with by the community). We have seen what happens when they're given a responsibility that is a shade of grey, and the fact their upstairs men don't really seem to have taken the hint means that we should not be giving them the power to enforce any policy or directive that is open to good-faith misinterpretation.
    Bluntly speaking, it seems like the WMF is unwilling to take community concerns on board except in a "your concerns are noted, get out of this meeting" fashion. We saw a prelude of this with Framgate and the Superprotect fiasco at de.wp, so it isn't an issue with en.wp specifically being ignored. And if the WMF will not listen to the community, then its officers have no right, outside of these bright-line issues that require an elevated responce, to sanction users based on interpretations of a vague, ill defined, culturally-based policy. —A little blue Bori v^_^v 2020's a bust; thanks SARS-CoV-2 02:16, 14 June 2020 (UTC)[reply]
  • I didn't expect this - but as I've said in other sections, utilize WMF T&S in difficult situations such as with the unblockables. What harm is there in working together as needed? I see it as a win-win. They are far enough detached to not have the close alliances we naturally build over the years - it's human nature to make Wikifriends with those editors we share common interests, especially when you collaborate well together. Atsme Talk 📧 19:40, 15 June 2020 (UTC)[reply]
    • The downside to this is that T&S does not do nuance or rationalisation. With unblockables, you almost need a scalpel to do the work; T&S is an atomic bomb. Their toolkit on-wiki is limited to blocking users, generally for indefinite periods of time, and as Framgate showed they're loath to debate their decisions in situations where the situation is not cut-and-dried (which, while helpful in most cases, amounted to a streisanding and the propagation of multiple conspiracy theories in Fram's case and almost certainly would result in much the same if they did this with any other "unblockable"). —A little blue Bori v^_^v 2020's a bust; thanks SARS-CoV-2 02:54, 16 June 2020 (UTC)[reply]
      • All excellent points by Couriano. By outsourcing to T&S you aren't just bringing in a third party to consider it, you're nominating a new set of not clearly laid out rules, rather than the en-wiki community policies. As an organisation they've been reticent to concede fault, which is not generally a positive feature in a group you want to consider whether someone has broken rules. Nosebagbear (talk) 10:05, 18 June 2020 (UTC)[reply]
  • The problem, I think, is a fundamental disagreement as to what are "the needs of the community". The comments I see on this page and elsewhere seem to indicate a spectrum of viewpoints, while the communications I see coming from WMF spokespeople on topics like this seem concentrated to one end. The end result seems similar to trying to get people with very different viewpoints on a controversial topic to work together to write an NPOV article, but without even the normal incentives to compromise and perhaps a little blindness as to how anyone even could reasonably disagree with an "obviously correct" viewpoint.
    As for "legal obligations", I don't think there are any that are really under dispute here. Disputed obligations seem more of the "moral" variety, with a similar situation as with "needs of the community" as to what they are and to what extent there really are obligations. Anomie 17:35, 21 June 2020 (UTC)[reply]
  • I would like to see the WMF provide resources and training for mediation and dispute resolution. In my view, most of our most toxic disputes begin as content disputes, and fester and escalate because we have no effective way of making a binding determination on content. Medicine pricing is a perfect example. I think that case should be a wake-up call for all of us. Guy (help!) 15:29, 23 June 2020 (UTC)[reply]
    • Yes, as I said earlier, we need content dispute resolution procedures that can deal with disputes more effectively, thereby letting people move on from arguing to adapting to what was decided, and removing incentive for poor behaviour that (knowingly or unknowingly) editors use to win arguments by attrition. isaacl (talk) 17:11, 23 June 2020 (UTC)[reply]
    • The problem isn't that we lack an effective way to make binding decisions on content. The problem is that we sometimes only address certain types of content disputes when they have escalated to the point where Arbcom will handle them as conduct disputes. There are types of content dispute where the issue can escalate to the reliable sources noticeboard, the BLP noticeboard, the MOS or the fringe theories noticeboard. For others we have the option of RFCs and have used them for otherwise intractable disputes such as the naming of the island of Ireland. The number of occasions per year where an RFC doesn't resolve a content dispute and it has to go to Arbcom is not that huge, especially considering that our purpose is to create content on every subject, however controversial. I'm not saying that the current system is perfect, it would be good if we could resolve issues such as medicine pricing while the pot was still simmering instead of having some disputes eventually boil over. But this RFC is about Arbcom and T&S, and I doubt that the foundation wants T&S to get into content disputes. Remember when people get really riled up, some will keep escalating until they have taken their dispute as far as they can take it; currently that's Arbcom, adding T&S as a new layer means that some disputes would just go to that extra layer. If we aren't happy with our current dispute resolution mechanisms we should look at improving them, knowing that simply adding T&S as a new more supreme court would make things worse not better. ϢereSpielChequers 07:45, 25 June 2020 (UTC)[reply]
      • I don't think that's what Guy's suggesting. Or if it is, it isn't what I support. What I think could be valuable is if the foundation provided training to us volunteer editors on mediation and dispute resolution. This could empower more of us to hopefully find ways to de-escalate and find consensus before it reaches the top of the conflict resolution pyramid. Best, Barkeep49 (talk) 14:56, 25 June 2020 (UTC)[reply]
        I don't dispute that "resources and training for mediation and dispute resolution" would also be useful. My comment was a response to "we have no effective way of making a binding determination on content." I think that the chapters could be used to commission some training on this - assuming by the time it is ready we are back to in person meetings. Still WMF money, but not one hopes involving T&S. ϢereSpielChequers 20:30, 25 June 2020 (UTC)[reply]
      • There are many discussions that fade out and fail to reach a conclusion due to a lack of sustained participation, and many RFCs that do not achieve a consensus, in many cases because there are equally valid opposing views that are just based on different underlying assumptions or interpretations, which English Wikipedia's decision-making model is not well equipped to resolving. Most editors genuinely look for something approaching a real-world consensus: a decision that most people can live with. So a small number of vociferous editors can easily stalemate progress. Remove the incentive to drive people away with aggressive behaviour, and editors will be less apt to indulge in it. isaacl (talk) 21:54, 25 June 2020 (UTC)[reply]
    • I agree with the idea for WMF training community members in mediation and dispute resolution, and I think it's perhaps the best suggestion for how we should structure the relationship between T&S and the community. I would really welcome a relationship where T&S offers period workshops to editors on dealing with conflict, deescalating disputes, bystander intervention, and mediating compromises (free of charge, or at least pay-what-you-can). The WMF has a substantial budget, and they should start investing in community development so that we are better able to avoid reaching the point where the WMF needs to use a ban hammer. Wug·a·po·des 23:49, 25 June 2020 (UTC)[reply]
      • Wugapodes, do you think that WMF have anyone on staff with the skills to teach those skills? If so, why are they not using them already? Would this be another situation where they would hire in an expensive name-brand consultancy with no prior knowledge of the projects talking down at the people who have been at the coalface for years? Done well, it could have great value, but do you have any confidence in the WMF's ability to do the job properly based on recent performance? · · · Peter Southwood (talk): 13:51, 9 July 2020 (UTC)[reply]
        • I don't know the answer to the first two questions, but presumably they don't. I understand the scepticism regarding the WMF, but it's simply unproductive to assume bad faith at this point. The WMF and the community have an obvious shared goal in ensuring the editorial community is healthy and inclusive, and if we go into negotiations treating them like adversaries, we will lose. Even if we don't see eye to eye on all points, we need to assume that the WMF will engage with these requests in good faith and with a basic level of competence. Do I think the WMF's first attempt at this will be perfect and without issues? No, of course not. Do I think that they will do their best to fulfill a request by the community to help train us to deescalate situations before they get to the level of needing WMF intervention? Yes. Why would they not? That's been their goal this whole time, and I think they would welcome a community requested solution. Wug·a·po·des 19:44, 10 July 2020 (UTC)[reply]
  • I have a practical suggestion: one member of ArbCom in each tranche should be appointed by T&S. Hawkeye7 (discuss) 01:07, 26 June 2020 (UTC)[reply]
    • If they were paid members of the T&S team this would certainly increase links between T&S and Arbcom, but how would you handle the tension on occasions when the T&S people know things that they can't share with Arbcom, or are voting per WMF orders. If T&S were nominating Wikipedians to Arbcom then I'm not sure it would improve the relationship, if anything it could complicate it. Especially if it wasn't clear whether these people were the main communication between T&S and Arbcom or some sort of backdoor communication. It can sometimes work to wear two hats, but often that requires stepping out of scenarios where you'd have to wear both hats at the same time. ϢereSpielChequers 19:16, 28 June 2020 (UTC)[reply]
      • Agreed. This would make the situation worse, not better. —A little blue Bori v^_^v 2020's a bust; thanks SARS-CoV-2 19:19, 28 June 2020 (UTC)[reply]
        • Not to mention the nightmare when ARBCOM were doing things like writing their open letter on the Fram case (an action that received a staggering level of support, by far and away the most supported ARBCOM action I've ever seen) - an appointed arb would be a huge chilling effect on incidents involving T&S, or encourage keeping information from one of the committee. Both hugely negative incidents. Nosebagbear (talk) 22:02, 28 June 2020 (UTC)[reply]
      • We already deal with these sorts of tensions. We have had occasions when some ArbCom members were given information they could not share, and it is even more common for ArbCom to have information it does not share with the community. If the WMF members vote per WMF orders, that is a desirable outcome. Hawkeye7 (discuss) 23:47, 29 June 2020 (UTC)[reply]
        • You have more faith in the WMF than me. I suspect that any such controversial vote would lead to people asking whether the WMF appointee voted the way they did becasue they were convinced or because they were ordered. As for information that Arbcom has that it cannot share, that isn't my concern on your proposal. I have been on committees where some people have pertinant information they can't or won't share with others, not a problem if they recuse, but don't expect consensus and harmony otherwise. ϢereSpielChequers 18:28, 1 July 2020 (UTC)[reply]
  • Given the context is in this RfC is "anti-harassment", I don't see much role WMF should play. WMF should focus on legal issues like copyright, libel, and the like and remove offending material. If someone is a recidivist in these areas, by all means send them to ArbCom. As for anti-harassment, I'm not sure that WMF is well positioned to examine facts and make credibility determinations. Carlossuarez46 (talk) 23:07, 1 July 2020 (UTC)[reply]
  • To begin with the WMF needs to be more active on the wikis of thier choice. If the community is to trust them then we need to see them out here with us. David Petraeus had the right idea in 2007 with his troop surge, part of which was getting the guys out of the bases and into the streets becaue he had some sense that the people your defending need to see you're making the effort. It may also be a good idea to invest in some digital anthropologists to get a better coomuncation between the the various groups on here so we can all start moving in the same direction again. TomStar81 (Talk) 10:35, 3 July 2020 (UTC)[reply]
  • Communication (both ways), transparency, competence, accountability. · · · Peter Southwood (talk): 14:57, 9 July 2020 (UTC)[reply]
  • It looks like some individuals here seriously think that WP:CIVILITY and WP:AGF just do no apply to WMF employees, certainly not if they have "WMF" in thsir signatures. I am afraid before talking about any meaningful relationship with T&S we first need to have some house-cleaning exercise.--Ymblanter (talk) 19:55, 10 July 2020 (UTC)[reply]
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.