::::Thanks {{u|GorillaWarfare}} yes, the Bbb23 one which has fizzled out except for the IP trolling. Hope all's well! [[User:Serial Number 54129|<span style="color:black">'''——'''</span>]][[Special:Contributions/Serial Number 54129|<span style="color:black">''SN''</span>]][[User talk:Serial Number 54129|<span style="color:#8B0000">54129</span>]] 15:56, 14 April 2020 (UTC)
::::Thanks {{u|GorillaWarfare}} yes, the Bbb23 one which has fizzled out except for the IP trolling. Hope all's well! [[User:Serial Number 54129|<span style="color:black">'''——'''</span>]][[Special:Contributions/Serial Number 54129|<span style="color:black">''SN''</span>]][[User talk:Serial Number 54129|<span style="color:#8B0000">54129</span>]] 15:56, 14 April 2020 (UTC)
:::::I'm not sure I agree with you that it's trolling -- it seems like a valid enough opinion. [[User:GorillaWarfare|GorillaWarfare]] <small>[[User talk:GorillaWarfare|(talk)]]</small> 16:17, 14 April 2020 (UTC)
:::::I'm not sure I agree with you that it's trolling -- it seems like a valid enough opinion. [[User:GorillaWarfare|GorillaWarfare]] <small>[[User talk:GorillaWarfare|(talk)]]</small> 16:17, 14 April 2020 (UTC)
:::::It was not intended as trolling. I spent a long time reading that thread and the comment was my attempt at making sense of what happened and what the disagreement was. Whether it contributes anything of value is not for me to decide, of course. But I tried. [[Special:Contributions/67.164.113.165|67.164.113.165]] ([[User talk:67.164.113.165|talk]]) 18:54, 14 April 2020 (UTC)
:::::It was not intended as trolling. I spent a long time reading that thread and the comment was my attempt at making sense of what happened and what the disagreement was. Whether it contributes anything of value is not for me to decide, of course. But I tried. [[Special:Contributions/67.164.113.165|67.164.113.165]] ([[User talk:67.164.113.165|talk]]) 18:, 14 April 2020 (UTC)
This page has archives. Sections older than 10 days may be automatically archived by Lowercase sigmabot III.
Behaviour on this page: This page is for discussing announcements relating to the Arbitration Committee. Editors commenting here are required to act with appropriate decorum. While grievances, complaints, or criticism of arbitration decisions are frequently posted here, you are expected to present them without being rude or hostile. Comments that are uncivil may be removed without warning. Personal attacks against other users, including arbitrators or the clerks, will be met with sanctions.
Comment on Bbb23 and symptoms of an overactive ArbCom
I posted this to CheckUser-l and ArbCom-en, but I think it worth posting publicly:
Just a general comment without being particularly aware of any specific incidents:
I’m sure there were likely checks that can be questioned, but that can be said of literally anyone who has the tool because the use of the tool is discretionary and discretion means that people are going to make errors in judgement calls. An error rate of 2-3% in any line of work requiring discretion is more than reasonable, but Wikipedia/Wikimedia’s system of accountability only looks at individual events.
Given the scale of his work, I’m sure whatever letter ArbCom sent could likely have enough individual events justifying it that would not reflect the scale of his good work. Given that he ran around 2,000 checks a month, even an error rate of 0.5%, which would be more than reasonable, would be 10 checks to point to a month when drafting something. There are of course individual events that might warrant this, but those are egregious ones involving using the tool to manipulate and control or intentional release of data, none of which I have seen here.
Having looked through his log, while he was more liberal than I may have been, but overwhelmingly he was right, and from a statistical standpoint likely was around the norm or better.
This is a symptom of an overactive ArbCom, probably in reaction to last one, but I am quickly losing confidence in the ability of this committee to be fair to anyone who is active in difficult areas, and have lost substantial respect for many of its members. TonyBallioni (talk) 13:54, 2 April 2020 (UTC)[reply]
One of the primary roles of the committee is to audit the use of checkuser and to provide appropriate corrective guidance where necessary. –xenotalk14:10, 2 April 2020 (UTC)[reply]
That response assumes infallibility. Given the last 3 months, this ArbCom has proven itself to be the most unfair and overactive that we have seen in recent memory. We have a desysop based on blatantly false FoFs (BHG), an ArbCom case over an IBAN that should have been kicked to AN and resolved in 72 hours, and now this. Sure, it’s within your authority to do all these things. My point is that you all have rushed to get involved in matters faster than any ArbCom in recent memory, and have a remarkably high public error rate because of this haste. This was an unforced error. TonyBallioni (talk) 14:22, 2 April 2020 (UTC)[reply]
No. I’m saying that the response of “this is within our remit” when the criticism is that they’ve been rushing into things all year and making mistakes along the way is an argument from authority that assumes because they can do it that they get it right. TonyBallioni (talk) 16:27, 2 April 2020 (UTC)[reply]
Okay, I can see the reasoning behind that even if I don't agree. Personally, I've been very pleased with this committee's work and in light of the Framgate debacle from last year I don't think anyone should be surprised if ArbCom is more inclined to take initiative than they have been in the past. LEPRICAVARK (talk) 16:34, 2 April 2020 (UTC)[reply]
(edit conflict) Unfortunately the argument that the more work one does inevitably leads to a higher error-rate in doing so was found inadequate in RHaworth's arbcom, and I assume the same princile applies. Also—and I don't think I'm known as a cheerleader of this arbcom, probably far from it—but although we're only painting pictures here, if there were policy violations involved (IF—I'm not sayng their were) then they would specifically not be mitigated by volume of activity.I can well understand Bbb23's refusal to be micromanaged, if that was the case—it's an absolutely untenable position to put anyone, and absolutely impossible for anyone to accept. It would be odd, indeed, if anyone thought it would have another outcome than it did. ——SN5412914:10, 2 April 2020 (UTC)[reply]
To summarize my arguments on CU-l:
Volume does not and should not negate intentional policy violations.
Enforcing established standards should be done even if it makes other admins/functionaries who also violate those standards uncomfortable and scared to continue violating those standards. Relevant for this case, and all the recent desysops.
Don't worry too much, he'll be back. The average enwiki retirement lasts about two weeks. But not before tens of users come out to lend emotional support, which becomes even more comical for the users on their 5th+ "retirement" who still get the same "oh the wiki will crumble without you" comments every time.
As to the specific case, I have enjoyed working with Bbb23 in the past, and hope he returns and continues with the 97% of his work that was excellent. -- Ajraddatz (talk) 14:50, 2 April 2020 (UTC)[reply]
The problem with that argument is there isn’t any evidence objective standards have been violated. Subjective ones possibly, sure, but very few things here are objective. The standard allow for discretion and are fairly vague. I could probably find checks in any individual CU log I wouldn’t have run, including arbs, stewards, and ombuds. That’s normal because there’s judgement involved and different people will have different judgement. The argument here is essentially that CUs are expected to make no errors, and that those overseeing the issue also make no errors. That’s not possible when dealing with matters of judgement. TonyBallioni (talk) 15:15, 2 April 2020 (UTC)[reply]
CheckUser does require judgment calls, but those calls still need to be made within the framework of the global and local policy. Nobody expects CUs to be perfect and error-free all the time, but problems arise when there is a trend of improper use of the tools, the user has previously been cautioned or made aware of what the problem is, and the behaviour continues regardless. Bonus points for sanctimonious dismissal of concerns and no self-reflection or attempt to understand what the problem is (Not necessarily saying that is the case here, but certainly has been with a lot of the recent desysops. And the same "judgement and vague policies" argument can be made with respect to those cases, in a similar irresponsible manner IMO). I think it is entirely acceptable for ArbCom to issue a warning to a user in that situation, if for no other reason than to reinforce what the standard is. As the local group responsible for oversight of functionary tools, that should be not only their right but their responsibility. -- Ajraddatz (talk) 15:30, 2 April 2020 (UTC)[reply]
No way to know for sure, but I expect it has far more to do with addiction to the website than the predictable outpouring of support every time an admin/long-term editor "retires". -- Ajraddatz (talk) 16:00, 2 April 2020 (UTC)[reply]
@TonyBallioni: just to be clear have checkusers been given more insight into what has happened here or are you speculating nearly as much (obviously you can see Bbb's checks in a way the rest of us can't so you have some more information) as the rest of us? Because from what Bbb has posted, I can speculate about two possible ideas behind ArbCom's actions. Option 1 is the error rate idea that seems to have been the focus here. If that's the case it appears like there are differing views among checkusers (sample size 2 in this thread but obviously also the checkusers of ArbCom deciding to warn but not revoke) about Bbb's performance. Alternatively, Option 2 is that there were certain categories (triggering events?) of checks that Bbb was performing that ArbCom wanted him to stop performing. I don't know which of these is more likely, but perhaps you do.Regardless, I don't blame Bbb for being really upset. Not in the least. But my hope is that in a few days or few weeks or even few months when he's in a better place that he can restart communication with ArbCom so he can find a way to not feel micromanaged and return to the valuable work he has done onwiki. Best, Barkeep49 (talk) 15:23, 2 April 2020 (UTC)[reply]
I recognize that CU is a very privacy-sensitive area, but in the interest of transparency, would ArbCom be willing to comment generally on what CheckUser policies Bbb23 violated? creffett (talk) 15:27, 2 April 2020 (UTC)[reply]
(edit conflict) @Barkeep49 and RoySmith: Roy, see User talk:Bbb23: Bbb23 has retired after a warning from the committee.Barkeep, we haven’t had specifics but have had discussions, but from what some arbs have said and what Bbb23 has publicly said, they have disagreements over interpreting the CU policy (particularly wikipedia:CheckUser#Grounds_for_checking.) That’s fine, but I also think if ArbCom were to propose a stricter interpretation of that in an RfC the community would overwhelmingly reject it, and so what we have amounts to ArbCom attempting to change policy by enforcement rather than its members trying to change the actual policy. I do not think the community as a whole supports reading the policy to effectively eliminate the fourth ground for checking, which places substantial discretion to the individual CU. It needs to be justifiable, but I don’t think Bbb23 was going around checking his enemies and I think he likely had a good idea why he was looking when he looked. TonyBallioni (talk) 15:37, 2 April 2020 (UTC)[reply]
@TonyBallioni: regardless of Bbb23 (right now), perhaps such a general RfC is warranted, vis a vis finding out/confirming exactly what the community wants/expects from CU discretion. A positive enforcement of the current wording—which I agree would be the most lkely outcome by a landslide—would presumably prevent this happening in the future. (Assuming, of course, that that's what's happened this time.) ——SN5412915:57, 2 April 2020 (UTC)[reply]
What is this? Like I think the community would support giving CU discretion. I don't think the community would support checking on every first time RfA voter (as Wikipedia Online speculates happens but I've always understood doesn't happen). So using an RfC to rebuke this ArbCom based on facts that we don't know (and which it seems even CUs don't fully know seems unwise). Using the RfC, in a thoughtful manner to evaluate to what extent the community is and isn't comfortable with the use of checkusers using their powers, all within the confines of global policy, does seem like this could be helpful. So as with so much on wiki, the devil's in the details. Best, Barkeep49 (talk) 16:11, 2 April 2020 (UTC)[reply]
Serial Number 54129, RfA because I wanted to give an example of something that critics have speculated about and which I think if true (and I don't believe it is) the community wouldn't support. Rebuke because Tony has been critical about this decision and you used the phrase "prevent this happening in the future". Now maybe that's a misreading. That's why I started with the question of what the "this" that you used in your last two sentences would be. Best, Barkeep49 (talk) 16:26, 2 April 2020 (UTC)[reply]
This is a great loss & I hope Bbb23 returns. But many members of the "community", as I have read it, have been sensitive about some checkuser blocks which, by their nature, are often unexplained or are not supported by SPI reports which have been filed. The community has to trust (or at least accept) checkusers and their judgment to act within the existing guidelines. I don't think an RfC proposing a general extension of permission to checkusers to be more liberal in their checks of editor accounts would be supported by all. LizRead!Talk!16:32, 2 April 2020 (UTC)[reply]
Speaking as one of the community members who share the sentiments you described, I certainly wouldn't be supportive of such an extension. When a CU says they have evidence of socking, the common folk are expected to believe them. ArbCom says they have evidence that Bbb23 violated policy. Well, I believe them. LEPRICAVARK (talk) 16:37, 2 April 2020 (UTC)[reply]
I wouldn’t support expansion either. I also am pretty sure that an RfC to tighten the policy would be defeated in a landslide. This committee’s views on the policy I do not think are in line with the community view on it and sock puppetry. The issue isn’t only Bbb23. The issue is a committee doing via enforcement what individual members couldn’t get passed via RfC: namely tightening use of the tool beyond what is currently authorized by policy. There’s a fine line between interpreting policy and changing it, and I think they’re trying to cross that line. There’s a happy middle ground, and I think the existing policy is actually works quite well. TonyBallioni (talk) 16:48, 2 April 2020 (UTC)[reply]
Who the checkusers are checking is always a black box to the rest of us, and has to be so. But my interpretation of Bbb23's statement is that ArbCom suddenly told him they didn't trust his judgement: "They have ordered me to limit my checking to 'non-discretionary' cases", and the "warning" was that they could take away his CU bit at any time. That's a serious attack on his integrity without possibility of rebuttal; I sympathise with his reaction and unless the committee—or the other CUs, who I know properly consult with each other frequently behind the scenes—had previously expressed concerns, I do think this was overly harsh. ArbCom is charged with disciplining and judging all of us, and has the power to destroy people's on-wiki reputations, but those reputations are the product of years of work here, and those with higher permissions have them because they are trusted and have the responsibility of exercising judgement in difficult situations. ArbCom telling a trusted editor they are unworthy of that level of trust is not a small thing, and their career can and should be mourned. Yngvadottir (talk) 16:42, 2 April 2020 (UTC)[reply]
I also think there was a conversation to be had between the Committee and Bbb23 first, before escalating to an outright directive.El_C16:49, 2 April 2020 (UTC)[reply]
According WP:AUDIT/STATS from January to December, Bbb23 performed 20,905 checks last year. Either he's CU'ing everything that moves, or then socking is a such a widespread issue that you can't really have proper grounds for checking and there's no way to audit that kind of amount. Of course this wouldn't be a problem if IP information was masked (T227733)... --Pudeo (talk) 17:00, 2 April 2020 (UTC)[reply]
Considering that a given sockpuppet case will have up to 20 or so socks, that's only maybe 1,000 or so cases. And considering Bbb23 does several cases per day, and is our most active and trusted CU, that's not all that many CUs. Softlavender (talk) 17:11, 2 April 2020 (UTC)[reply]
Pudeo, handling the kinds of SPIs that Bbb handled frequently indeed takes a lot of checks. If an SPI has 20 socks, it may well be that you have to run CU not just on a lot of the socks but also on a lot of the individual IPs used by the socks you're checking. And handling such big cases can get enormously complicated, with a ton of windows open and comparisons drawn constantly between socks and IPs and whatnot, especially if the sockmaster isn't just using one machine to make the same kinds of edits in the same pattern. Guess who I always called on when it got too complicated for me--that's right, I called on Bbb23. Drmies (talk) 01:57, 3 April 2020 (UTC)[reply]
Wikipedia shouldn't need to rely on one anonymous volunteer to look at private data more than 20,000 times a year. That's a systemic failure. Mr Ernie (talk) 07:28, 3 April 2020 (UTC)[reply]
This current ArbCom seems determined to rid us of our best admins. So far they are succeeding admirably, and very few people seem to care or to realize what a ridiculous and insidious form of authoritarianism this is, that a mere handful of people can peremptorily drive the best, hardest working, and most trusted admins off the project with complete impunity. And apparently there's nothing we can do about it, and too few people that care or realize the danger to put a stop to it. Softlavender (talk) 17:08, 2 April 2020 (UTC)[reply]
I prefer the interpretation in which the current ArbCom is committed to holding even the most prominent, well-connected admins responsible for their conduct. Which is very much a good thing. LEPRICAVARK (talk) 17:16, 2 April 2020 (UTC)[reply]
It's highly unlikely anybody has been driven off the project. Bbb23 chose to take what is far more likely a wikibreak than a retirement – and chose to announce it as a probable retirement on his widely-watched talk page – knowing full well that it would ignite the next overblown superuser drama, this time with him at the center – knowing full well that many editors would be quick to pick up the flag as yet another loss of a highly-experienced and highly-competent editor resulting from the deplorable, outrageous overreach of a group that has more power than they do. Oh, the humanity! Retirement is not a political tool. In all likelihood, Bbb23 has not left the project, and, if he has, that was entirely his choice and his decision. That sort of hyperbolic talk (and hyperbolic thinking) has no place in rational discussion, but we see it repeated in each and every case like this. That's the more important story in my estimation. ―Mandruss☎19:17, 4 April 2020 (UTC)[reply]
I'm thinking this "acceptable error rate" argument is not a particularly good one. The acceptable error rate for an admin, AFAIK, is zero, with occasional, independent exceptions that happen in good faith and are easily and rapidly resolvable. You never hear about this supposed "acceptable error rate" defense for normal users. I don't get to make a bad block and then tell the community "I have an error rate of only 1%, so just look the other way". I mean, sure, we all screw up sometimes, and that's okay, but those of us who block vandals aren't running a 3% percentage of bad blocks. Those of us who close AfDs aren't running a 3% percentage of closes where we get overturned at Deletion Review. Those of us who delete CSDs don't get a 3% allocation of out-of-policy deletions that we can perform before someone does something about it. In any of these areas, if we screw up we get called out on it, and we're expected to not do it again. If we're continually misusing the admin bit, we get desysopped. It does not matter if it's 3% of the time, whether you make 1000 actions in a year or 50,000. Yet ironically it's only the user who makes 50,000 who will be granted this defense that their 250 bad actions in the past year is within some sort of acceptable "error rate". Nah. That's not the standard we're held to. The standard is that if you're making so many edits or actions that you're running an "error rate" that is having any sort of negative impact, you're working too fast, and your priority needs to be to slow down and eliminate your errors. I've been a strong critic of the overreaches of this Arbcom but also of Bbb. I don't have absolute trust that Bbb deserved this, nor that this is just another overreach by Arbcom. We don't have the facts and likely never will. What I do know is that Bbb consistently seemed completely unaccountable, both when he was in the right and in the wrong. There have been no shortage of on and off-wiki discussions about this, I'd go so far to call it a well-known fact. When he was in the wrong, and he sometimes very obviously was, even in spite of the confidentiality he was protected by, there was never anything anybody could do about it. He got away with it purely because of his activity levels; he was the undisputed king of the CUs. I for one assumed that he would remain untouchable no matter how in the wrong he was. Even knowing the facts, how deserved this measure was would probably still be up for heated debate. But I have to say, I appreciate Arbcom's willingness to hold him accountable. We're primarily responsible for enforcing accountability amongst our individual selves. When that fails, the community is supposed to step in. Bbb consistently failed in this, and due to his "confidentiality", the community could not realistically do anything about it. Whether or not their specific sanction was justified (to me it comes across as lenient, in lieu of revoking CU, an extremely restricted privilege), this is exactly the situation where they should have become involved. I applaud them. ~Swarm~{sting}17:20, 2 April 2020 (UTC)[reply]
I'm not sure that's right. If someone's contributions is orders of magnitude greater than someone else, it does stand to reason that their errors would be greater by virtue of sheer volume. El_C17:27, 2 April 2020 (UTC)[reply]
I think the argument Swarm was making in that regard was if they are making so many contributions that errors are getting introduced, maybe they should slow down to eliminate such errors instead of just excusing them. I do not wish to speak for Swarm though, so I could be mistaken. PackMecEng (talk) 17:31, 2 April 2020 (UTC)[reply]
That notion is exactly what I'm disputing. Like I said, that argument only holds up if an "acceptable error rate" is a thing, which it isn't. It's an unacceptable excuse that only gets invoked with power users with a high number of errors. No one else is afforded this "free pass" for any number of errors. ~Swarm~{sting}17:33, 2 April 2020 (UTC)[reply]
El C, this thread opened with a vehement argument that an "error rate of 2-3% ... is more than reasonable". Swarm seems to be countering that no it isn't, and that this would seem to be ArbComm's position. While we all screw up on occasion I cannot but agree with Swarm that in this sort of area "The standard is that if you're making so many edits or actions that you're running an "error rate" that is having any sort of negative impact, you're working too fast, and your priority needs to be to slow down and eliminate your errors." And if you don't and won't it is appropriate for the stick to be pried out of your hand while you are shoo'ed away from the horse. Gog the Mild (talk) 17:35, 2 April 2020 (UTC)[reply]
Gog the Mild, I'm not even sure that error rate is a meaningful concept here. It is expected that CU checks will turn up negative and positive results. Do they mean 2-3% of checks fall outside of policy, or 2-3% of positive checks are successfully demonstrated by an appealing user to be false positives? Either of those would be rather high, and I doubt that either is actually the case here. As usual, we can't know the details. That's the really frustrating part. Guy (help!) 17:41, 2 April 2020 (UTC)[reply]
Gog the Mild, I was speaking imprecisely. See below for clarity: I suppose a better way of saying it is that for any judgemental area, 2-3% room for disagreement between people who take a more liberal view and people who take a more stringent view is allowable, and that people are likely going to make mistakes in their judgement on occasion in addition to these disagreements. There’s a difference between that, which I think in unavoidable in ever area, and a forced error like checking an opposer in an RfA where you were the nominator. TonyBallioni (talk) 17:51, 2 April 2020 (UTC)[reply]
I was recently warned for saying something was Nazi-esque. I've made 14,290 edits, and only one of them broke Godwin's law, putting my Godwin Error Rate at 0.007%, but exactly none of our colleagues would find that line of argument convincing. Same with admin tools. The commonality between RHarworth and Bbb is that the high volume is, itself, the problem. There are other admin who are similarly high-volume-outliers, and they would similarly be well-advised to slow down. One might change the oil on 10 cars a day without spilling any, but not 100 or 1,000. Some actions on WP need to be done slowly and carefully, and CUing and blocking users should be top of that list. Levivich[dubious – discuss]17:38, 2 April 2020 (UTC)[reply]
I agree with Swarm, but also want to point out that we aren't really talking about an error rate here. Bbb23 isn't making widgets and occasionally making honest mistakes that result in the widgets being unusable. Bbb23 is making judgment calls as to whether there is sufficient evidence of abuse to justify the invasion of a user's privacy, and he is routinely not making those calls correctly. This has been going on for a long time, and is another one of these funny Wikipedia memes that everyone knows is true but nobody has done anything about. I'm glad that these cases are being dug up and actually addressed. I've never found myself fanboying over the Arbitration Committee before, but the leadership demonstrated by this Committee has been refreshing. -- Ajraddatz (talk) 17:44, 2 April 2020 (UTC)[reply]
Well, if consensus is against my view, that's something I take to heart. Speaking for myself, I've been more active lately with the world-wide pandemic crisis, which has resulted in more errors than normal by virtue of sheer volume. But maybe I should slow down, then. I'm just not sure it'd be to the benefit of the project to do so at this time... El_C17:54, 2 April 2020 (UTC)[reply]
I want to speak generally and not specific to Bbb. I think we can talk about errors and uncorrected errors. If I make an occasional mistake in most contexts and fix it that makes me a good Wikipedian. If I make the occasional mistake as an administrator and someone points that out to me and I fix it, well that makes me a good sysop. Certain kinds of errors would never be acceptable - 1 instance of a checkuser releasing private information would fall in that category. But for the most part I think it behoves us to think differently about mistakes that are fixed and mistakes (as defined by the community) where the person making the mistake digs their heels in. So if you increase volume but don't increase uncorrected mistakes I don't think you (El C) or anyone else is doing anything troubling. At least that's my take Best, Barkeep49 (talk) 18:10, 2 April 2020 (UTC)[reply]
Ironically the pandemic strikes me as an apt invocation. Some people, in my region at least, are touting the seemingly small death percentage as an argument that it's "not a big deal", and that the societal shutdowns are draconian and irrational. Why should we destroy the economy by shutting down business when the death rate is not even 2%? 98 percent of people who get it live through it and are fine! Well, it doesn't sound like a big deal when you put it that way! But 2% of the American population is 6,544,000 fucking people! It doesn't matter what your ultimate contribution is if the damage you're doing is significantly harmful. Your damage percentage, or euphemized, your "error rate", is irrelevant. Your contributing the same level of damage either way. ~Swarm~{sting}18:25, 2 April 2020 (UTC)[reply]
From what I can see that Bbb23 wrote, it is not his error rate that ArbCom is concerned with, but that they claim he was using CU in ways "that violate policy". That's a completely different thing from making mistakes. Black Kite (talk)17:39, 2 April 2020 (UTC)[reply]
Let me clarify: there is no objective standard as to what constitutes a breach of WP:NOFISHING, and the text allows a lot of discretion. My 2-3% rate was arguing that for any judgemental area, that’s about the best you can expect someone auditing judgement calls to disagree with even if you’re doing your absolute best. There’s a difference between disagreeing on what constitutes reasonable suspicion and blatantly violating the rules we are explicitly not allowed to follow. One mistake in the latter isn’t allowable. In the former, I’m willing to give everyone in every position some room for disagreement. TonyBallioni (talk) 17:46, 2 April 2020 (UTC)[reply]
Hey all, sorry for the sort of ad-hoc response you're getting here. We had intended this to be a private advisory note to Bbb23 and not a public rebuke, and so we did not prepare a public statement about our inquiry. I want to be clear that we did not issue this notice to Bbb23 based off
of one or two checks in which a checkuser on the ArbCom would have made a judgment call different to the one Bbb23 made. As TonyBallioni and others have already mentioned, the CU policy (like most policies) is open to interpretation, and reasonable people will interpret the same policy differently. We absolutely expect and understand that checkusers will not agree 100% of the time on whether a specific check should be run. However, given the private information that is exposed to checkusers when a check is run, we do expect checkusers to be able to justify the checks that they are making are supported by policy, and are not "fishing".
Following multiple complaints to the Committee, as well as the results of checkuser block appeals considered by the Committee, we performed an audit of Bbb23's checkuser tool use. I will note that we did not calculate a threshold of acceptable errors, as has been discussed here, but instead reviewed whether or not there was a pattern of running checks on users that did not appear to be supported by policy. We also reviewed whether Bbb23's checks were in line with the kinds of checks being run by the other checkusers on the checkuser team—again, we understand that policy is open to interpretation, and wanted to form a view of whether Bbb23's interpretation was in line with that of the rest of the enwiki CU team. We contacted Bbb23 to ask about the standard he applies when deciding to check users, as well as request more detailed justifications of some individual checks. Following this conversation, the Committee sent Bbb23 a notice that reiterated checkuser policy, especially against fishing, and asked him to avoid performing checks that fall in the "grey area" around what is considered fishing. We also encouraged him to consult with other members of the checkuser team when in doubt of the validity of a check, and warned that further misuse of the checkuser tool would result in removal. I hope this provides a little more clarity. GorillaWarfare(talk)18:20, 2 April 2020 (UTC)[reply]
"We had intended this to be a private advisory note to Bbb23 and not a public rebuke, and so we did not prepare a public statement about our inquiry." When will you learn? --~Swarm~{sting}18:28, 2 April 2020 (UTC)[reply]
Thanks, GorillaWarfare, I really do appreciate your public response here. I think it’s fine to say to Rschen7754 that the dispute is functionally what the policy WP:NOFISHING means for new accounts behaving like socks where a master is not known. We all make these checks and there is no clear line. I wouldn’t want a hardline interpretation of it, but others might. TonyBallioni (talk) 18:40, 2 April 2020 (UTC)[reply]
@TonyBallioni: I disagree with you that this is about the ArbCom taking a hardline stance on WP:NOFISHING. We are not making an example of Bbb23 for making the same kinds of checks that "we all make", if that's how it appears to you—again, we did examine whether Bbb23's checks were in line with the kinds of checks being made by the rest of the enwiki CU team. GorillaWarfare(talk)18:44, 2 April 2020 (UTC)[reply]
GorillaWarfare, two members of the committee have privately made claims that directly contradict that policy: namely that outcome has any impact on if the initial check is valid. Yes, this is about the committee taking a hardline and individual committee members thinking they have the authority to disregard the community’s established norms and replace them with their own. As I’ve said on list: if this is the line some arbs want to take; they should try to amend the policy. No such consensus exists for that view, which is why that RfC will never be started. TonyBallioni (talk) 18:50, 2 April 2020 (UTC)[reply]
Just curious...what does it hurt? It seems to me that the biggest risk is to the IPs who have something to hide. Are we at the point we can't trust our most trusted users in the effort to protect anonymity? Perhaps incidents such as this one will help make a case for registration where only a select few highly trusted CUs are allowed access to the info - maybe WMF could consider obtaining a surety bond to cover them as a precautionary measure. I have far more trust in our CUs than I do in, say, members of WP:COIN. AtsmeTalk📧12:05, 3 April 2020 (UTC)[reply]
GorillaWarfare, thank you for commenting on this, whether or not people agree with ArbCom's action I hope they at least appreciate the statement - it's clear that you all were caught off-guard by this becoming a public issue. creffett (talk) 18:29, 2 April 2020 (UTC)[reply]
I don't doubt that Bbb23 ran some checks that fell in the "grey area", and I dare say that any active CU will have at least a few such checks in their log, but I trust that they were done out of a desire to protect the project, not of out any malice. At any rate, you've lost one of your best, most experienced CUs. —DoRD (talk) 18:35, 2 April 2020 (UTC)[reply]
To be clear, I don't believe Bbb23's motivations were bad. But this is not about "a few" grey area checks. Checkusers have an obligation to abide by the CU policy, and the ArbCom has an obligation here as well to investigate complaints and ensure the tools are used in accordance with the applicable policies. GorillaWarfare(talk)18:40, 2 April 2020 (UTC)[reply]
TonyBallioni, have you been reading a completely different set of comments from GW to the ones on my screen? I was a little twitchy before, and am now much reassured. Although Swarm's comment was to the point. Gog the Mild (talk) 18:58, 2 April 2020 (UTC)[reply]
I wasn’t referencing GW, who I appreciate. I’m more concerned that some arbs have privately expressed that they’re looking at outcomes in determining this. That’s a direct contradiction of policy. TonyBallioni (talk) 19:06, 2 April 2020 (UTC)[reply]
@TB, if you and I mean the same thing by "looking at outcomes", then it seems perfectly reasonable to me regardless of what policy says. A CU who runs 100 checks, 99 of them turn out to be socks, is less likely to be acting inappropriately than a CU who runs 100 checks all of whom turn out to be innocent new users. There's always been some leeway given for "copper's instinct"—as formalized in WP:DUCK—but when the instincts are consistently off, that's a problem. (Disclaimer that I no longer have access to the CU logs and have no idea if this is what's been happening here.) ‑ Iridescent19:10, 2 April 2020 (UTC)[reply]
I think that’s a fair point that could be raised in a discussion. I’m concerned that it’s being used in way to claim one of our best CUs is violating policy. There’s a valid discussion to be had around if one person should be doing so many and what lines are to be drawn concerning fishing, but there should be an assumption this person was trying to help the project and was acting within his understanding of policy. I’m not sure that’s the case here. TonyBallioni (talk) 19:23, 2 April 2020 (UTC)[reply]
Well with that response from GW I am fairly confident that the committee acted appropriately and clearly within their authority. Seems more like a nothing to see here kind of situation. So lets not add to the drama and just move on. PackMecEng (talk) 19:46, 2 April 2020 (UTC)[reply]
If the purported standard is "discretion" than we have to have a group of uninvolved CU's (which Arbcom is) make the determination -- leaving it to the one CU to be the sole judge in his/her own actions, is not a standard, it's a licence to complete unaccountability. Alanscottwalker (talk) 20:22, 2 April 2020 (UTC)[reply]
I know Bbb23 has been a workhorse, and has had the best interests of WP at heart; I've had a few disagreements with him about how he interacts with the hoi polloi on-wiki, but I've never doubted his integrity, especially regarding the use of checkuser. But knowing that and saying it doesn't address the actual issue, because we obviously have no idea whether Bbb23 was running too many fishing-type checks or not, and have no way to ever find that out. For me, at least, there are enough sane individuals I'm familiar with on ArbCom, that I'm confident that the decision to send a cautionary email of some kind was not just plain crazy or out of left field; there was probably some kind of legitimate underlying problem there. However, if I can be forgiven for making wild-ass guesses based on observations of past behavior, I would not be at all surprised if (a) the email was worded (in retrospect) somewhat poorly and/or too aggressively, and (b) Bbb23 over-reacted to it. It's almost as if the members of ArbCom and Bbb23 were all imperfect human beings or something. I remain deeply impressed and appreciative of the tireless unpaid work that both ArbCom and Bbb23 perform, imperfect or not, and wish this had gone down differently. --Floquenbeam (talk) 22:34, 2 April 2020 (UTC)[reply]
Floquenbeam, the email has been shared with the checkuser team. It was firm as it needed to be, but I do not believe it to be aggressive.
This was not out of the blue, we had sent a general reminder on similar issues last year, to the entire functionaries team
When I was relatively new, and all I knew about CU was the policy summarized at WP:SPI in stern words, I believed CU to be a very rarely used tool. It seemed almost impossible for me to provide sufficient reasons for a check to be done, in almost all cases. When I gained experience, over time, I noticed that this impression seems to be wrong. The wording of the policy isn't as strict as WP:SPI makes it look like, and checks are performed more easily and regularly than I ever expected. I'm not even saying this is a bad thing, but it surprised me. There seems to be a difference between a naive policy reader's perception of CU practice and the actual requirements for a check. The discussion seems to confirm this view: The wording of the policy looks strict to the naive reader, but is being interpreted pretty loosely by those bound to it. ~ ToBeFree (talk) 02:01, 3 April 2020 (UTC)[reply]
Would it be possible to publish some (anonymous and aggregated) statistics about CU usage? ~ ToBeFree raises a valid question. Things like:
Number of checks performed per day
Percentage of check requests that are rejected
Percentage of checks that are done without a third-party request
Percentage of checks that are negative
would go a long way towards helping the rest of us understand the extent of this tool's use, and don't seem like they would be in conflict with our privacy policies. -- RoySmith(talk)15:03, 3 April 2020 (UTC)[reply]
I anticipate protest about the last bullet point, claiming that this is an irrelevant metric. I disagree with this protest in advance. ~ ToBeFree (talk) 15:05, 3 April 2020 (UTC)[reply]
ToBeFree, Why is it irrelevant? I certainly agree that it's an inappropriate metric to use to judge any particular check, and I'm not asking that it be broken down by individual CU. But, we're entrusting a group of people with a sensitive tool. It would be good to know how carefully they're wielding it. I'm also aware that the set of CU's is small, so the anonymizing power of aggregation is limited. Particularly so if the distribution of checks performed per CU is highly skewed, as it appears to be. Still, it should be feasible to give us some idea of how the tool is being used without impacting the confidentiality of the checkers or the privacy of the chekees. -- RoySmith(talk)15:17, 3 April 2020 (UTC)[reply]
I think it's relevant, for the same reasons. There has been some discussion above whether ArbCom used this metric to admonish Bbb23, and whether that would have been in line with the policy. However, I think that a high number of negative results can indicate a problem with the policy, and a low number of negative results can confirm that the policy has been well-designed. ~ ToBeFree (talk) 15:27, 3 April 2020 (UTC)[reply]
ToBeFree, Ah, I mis-parsed what you wrote earlier. You said you disagreed with the protest. I mis-read that as disagreeing with the desire for the statistic. My apologies. -- RoySmith(talk)15:36, 3 April 2020 (UTC)[reply]
User:RoySmith, User:ToBeFree: I'm not sure there is really a concept of a negative check. CU is not like a coronavirus test. It's evidence gathering. Evidence is any info that changes the user's subjective probability of some hypothesis. Sometimes it nails a suspicion established some other way, while other times it only helps notice a pattern. I can accept the term "positive" for a conclusive finding from a CU, but that doesn't make an inconclusive finding negative. 2601:648:8202:96B0:E0CB:579B:1F5:84ED (talk) 07:35, 13 April 2020 (UTC)[reply]
In case you're not already aware, we do publish monthly checkuser and oversight statistics at WP:AUDIT/STATS. I don't see any reason we couldn't publish a daily version, though I'm not sure it would provide insight that the monthly numbers do not. Your other requests are a little trickier. While it would potentially be feasible to gather statistics on how many requests for CU at SPI are rejected, checkusers receive requests for CU from all sorts of venues, including private messages, which we don't have access to. The same goes for your third request -- we don't necessarily know at a glance which checks are based off third-party requests or which are from the CU's own suspicions, so gathering this data would require each CU to go back through their checks and report this to us. Again, not impossible, but certainly difficult. As for the last question, this is perhaps the hardest to elucidate. Checks often do not result in a binary "positive" or "negative" result, but rather provide an additional data point into an investigation that also takes into account behavior and other inputs. Sometimes checks do not immediately result in anything, but later in combination with other evidence result in action. GorillaWarfare(talk)15:16, 3 April 2020 (UTC)[reply]
I'm surprised to read we don't necessarily know at a glance which checks are based off third-party requests or which are from the CU's own suspicions, so gathering this data would require each CU to go back through their checks and report this to us. I imagine there is some sort of log of CUs (on the CU wiki?). Checkusers should be recording somewhere the reason each check (or batch of checks) was performed, even if it's a multiple-choice "SPI/talk page req/mailing list/private email/on own initiative" selection. That would make it easier to gather that aggregate data. If this isn't already done, I hope some changes can be made, because we should have a record of why every CU was run. Levivich[dubious – discuss]16:00, 3 April 2020 (UTC)[reply]
Checkusers do enter an explanation for checks they run (much like an edit summary). However we mostly use this to record the justification for the check and not necessarily the origin of the check -- checks requested at SPI do autofill into the CU log so that's easy, but not all checks originate from an SPI. GorillaWarfare(talk)16:04, 3 April 2020 (UTC)[reply]
Thanks for the quick response, GW. Maybe it would be helpful to use like a four-letter code in the "edit summary" explanation that denotes the origin of the request, assuming that would be searchable, we'd know what portion of CUs are done based on public complaints, private complaints, or no complaints. Levivich[dubious – discuss]19:31, 3 April 2020 (UTC)[reply]
@ToBeFree: I deduce two things from your applange.i) that a phenomenally small number of CU-checks are run (still in single figures this year), and thatii) the logs, such as they are, are publically available (with the unspokeniii) being, so why aren't ours). ——SN5412916:39, 3 April 2020 (UTC)[reply]
Ajraddatz does that table represent the total number of "incidents" where CU was used on dewiki? Because if so I think the difference between enwiki and dewiki is radically different - we have about as many SPIs with a CU action (which we know is not the only circumstance which triggers a CU) as they have total. Best, Barkeep49 (talk) 16:46, 3 April 2020 (UTC) Fixing ping of AjraddatzBarkeep49 (talk) 16:47, 3 April 2020 (UTC)[reply]
Points to note:
Enwiki logs are per check, Dewiki logs are per case
Dewiki activity is about 1/7 (personal estimation based on various statistics) of Enwiki activity
Stewards' checks on global account creations do not appear in either statistic
Dewiki CU policy is extremely strict; checkuser is limited to requests from third parties; requests must be public except in rare circumstances; all non-public requests lead to public log entries. The log is comprehensive, without exceptions. See de:WP:CU:
("Die Checkuser-Berechtigten werden ausschließlich bei Anfragen Dritter aktiv."; "Eine nicht öffentliche (per E-Mail oder im IRC gestellte) Anfrage ist beispielsweise möglich, wenn unerlaubt eine Privatadresse veröffentlicht wurde und eine offene Checkuser-Anfrage dem Betroffenen durch die damit verbundene erhöhte Aufmerksamkeit nur zum weiteren Nachteil gereichen würde.")
"nein" ("no") entries are declined requests for CU action.
Indeed: while the enwiki policy is quite liberal compared to some projects, from my understanding the dewiki CU policy is one of the strictest. Though I will say that if people here are interested in either changing the enwiki CU policy around when checks may be run, or requiring CU investigations to be publicly logged like they are at dewiki, that's probably a better conversation for a place like WT:CHK or WP:VPP. GorillaWarfare(talk)17:07, 3 April 2020 (UTC)[reply]
Not sure if you meant to ping me, but I can confirm that dewiki has probably the strictest CU requirements out of the Wikimedia family. They take the two-part test for deciding whether a checkuser should be done (1 - is there evidence of disruption and 2 - is there evidence of misuse of multiple accounts) to its most restrictive bounds and end up only investigating around 5 cases per month. -- Ajraddatz (talk) 17:32, 3 April 2020 (UTC)[reply]
As someone told me, it's almost a public vote on whether a CU should be performed - everyone winds up complaining about the privacy of someone being violated and in the end the CU is almost never performed. --Rschen775418:06, 3 April 2020 (UTC)[reply]
That's good, because there's no evidence to suggest Bbb23 would have failed to comply with Committee directive. He disagreed with it, so he quit. Simple as that. El_C18:12, 3 April 2020 (UTC)[reply]
Maybe, with last year's big issue being a bit of a wakeup, we're transitioning from old days when a little critique / guidance to an admin/CU/major figure was so rare (and non-existent to the pillar people) that it was interpreted as a sledge hammer; transitioning to where it can viewed as a more routine healthy thing. Nobody is so perfect that such input is never needed. If the wording was diplomatic & appropriate, perhaps Bbb23 could re-interpret it that way. If it wasn't, perhaps Arbcom can make it so and then Bbb23 take it in that way. North8000 (talk) 19:43, 3 April 2020 (UTC)[reply]
For the record, the language of the email has been shared on the CU mailing list. Some of our CUs think we are changing policy, but they don't seem to think the language was unduly harsh. I'm sure they'll correct me if I'm wrong about that. Actually I'm not sure why we can't or shouldn't just post the entire thing on-wiki but I'd want to clear that with the rest of the committee first. Beeblebrox (talk) 20:00, 3 April 2020 (UTC)[reply]
I haven't commented here but I was one of the ones most vocally critical on the mailing list. Beeblebrox's summary is accurate: the email we saw was in the diplomatic language we all expect from ArbCom; we disagreed with some of its content but I don't think any of us thought it was harsh. I would say the CUs who spoke up see it as a novel interpretation of policy which also conficts with a previous interpretation of policy, and more importantly what we understand to be community expectation for use of the tool, and as such we suggested the Committee should test their guidance with a public RfC before pursuing enforcement of it. Nonetheless, I don't see the benefit of publishing the email here, it was intended to be private correspondence and making it public is unlikely to change anything. Had Bbb23 wanted it to be published he could have done so himself; it would be a violation of privacy for no productive reason to publish it, which ironically is what the policy is all about in the first place. Ivanvector (Talk/Edits) 12:37, 4 April 2020 (UTC)[reply]
It's really hard for the community to weigh in without the ability to see the CU log unfortunately. I don't know that publishing the email would really help. --Rschen775418:10, 4 April 2020 (UTC)[reply]
Meh. We're all volunteers here. Anyone can withdraw their services for any reason or no reason at all. If Bbb23 thinks the new requirements are so insulting and meddlesome that he doesn't want to do the work anymore, then he has every right to walk away. You don't need to agree with his interpretation of things, but you do need to respect his decision to leave. It's not like WP has a claim on his time. ReykYO!13:29, 4 April 2020 (UTC)[reply]
My take on this is there should be an RfC. We've already heard from three CUs who clearly disagree with ArbCom's interpretation. It's reasonable to expect there's others who agree but aren't as vocal.
So, what we've got are two groups of the most experienced, trusted, and empowered people in the project at odds with each other over interpretation of policy, with nobody else privy to the details. Neither of those groups makes policy. The community makes policy. Functionaries implement and enforce it. So, let the community decide if the guidance ArbCom is giving to the CUs falls within policy. One of two things will happen. Either the community will agree with ArbCom's interpretation, in which case the CUs will have clear direction. Or they will disagree, and again the CU's will have clear direction.
Much of what ArbCom and CUs do is confidential, and rightly so. Policy decisions, however, need to be done in the open, and by the community. In theory, the community has the final say because we can vote in a new ArbCom next year. But given the current opacity around the details, we don't even know which arbs voted which way, so people who disagree with the decision have no information on which to base an informed decision comes election time. -- RoySmith(talk)16:00, 4 April 2020 (UTC)[reply]
Can you clarify what you're talking about? What exactly is this "difference in interpretation" of CU policy that is being discussed? From my reading of GW's comment, the actual issue was WP:NOTFISHING, the simple principle that CU checks must be motivated by evidence to begin with. What constitutes "evidence" is subjective and open to interpretation, sure, but you seem to be saying that Arbcom has taken up some sort of novel, radical interpretation of the policy that is egregiously out of line with the traditional views by the CU team and by the community, to the extent that they're "rewriting policy". So what exactly is this controversial new stance that they have taken? ~Swarm~{sting}16:21, 4 April 2020 (UTC)[reply]
Swarm, No, I'm not saying that at all. What I'm saying is that we have no information about what the disagreement is, and that's a bad thing. I wish I could clarify what I'm talking about, but I can't, because not enough information is being made public.— Preceding unsigned comment added by RoySmith (talk • contribs) 16:36, 4 April 2020 (UTC)[reply]
Well you say that they have made a "policy decision" that CUs are disagreeing with. GW says they're being accused of "rewriting policy". So presumably they've stated some sort of contentious policy position, yes? Sorry, I assumed you were one of the CUs who disagreed. I am in agreement with you that more information is needed, but I see now that you cannot be the one to provide it. So, if any Arbs or CUs would like to provide this clarification, I agree that we should hear it. ~Swarm~{sting}16:45, 4 April 2020 (UTC)[reply]
@Swarm and RoySmith:, I’ll reply since I’ve been one of the more critical ones: the issue some of us had was the phrasing surrounding what justified a reason to check an account that is new. “Obvious not new” or some variation of it has been one of the more common reasons to run a check for a while. The thing is that policy requires reasonable suspicion of abuse of multiple accounts. The most common reason to check an account that is clearly not new to Wikipedia but where a specific master is not known is involvement in project space discussions, which is prohibited under WP:SOCK. Another common reason to check would be use of an undeclared alt to edit in areas where community review of actions is normal, such as areas under discretionary sanctions or pages that are highly controversial as not using your main account there may constitute an effort to evade scrutiny. Brand new accounts creating spam pages perfectly with no errors, perfectly formatted citations, and link spam disguised to look like citations is another common reason.About a year ago, ArbCom sent guidance to the functionaries list cautioning that simply not being a new account is not a valid grounds for a check without an indication of abuse if the account was in fact not new. The three reasons I listed above were included as valid reasons to run a check if there were enough indications to suggest that the account was not in fact new, because those constitute violations of the sockpuppetry policy. The email ArbCom forwarded us that they sent to Bbb23 appeared to several of us on the list to say that these checks were no longer valid, and several of us objected saying that it was inconsistent with last guidance and would amount to a change in policy. Worm That Turned replied to an email I sent clarifying that what the committee was saying was that simply being a new user in project space, a controversial area, or creating a spam page was not on its own a reason for a check, but that they may be valid reasons for a check if other indicators existed suggesting that there was abuse of multiple accounts. This statement is something I think the CU team as a whole agrees with, I personally agree with, and I read as being consistent with the guidance issued last year, and I think everyone is on the same page with that. I’ll ping Risker, Ivanvector, and GorillaWarfare since they’ve all commented, to see if it’s a fair summation of the discussions.Tl;dr: I think after dramah surrounding this, there’s a general sense of people finally being on the same page. TonyBallioni (talk) 05:09, 5 April 2020 (UTC)[reply]
TonyBallioni, So are you now withdrawing your complaint? And do you agree that there is not "anything in the message to Bbb23 that was in any way a change in the interpretation of policy" ? (see Risker's comment below). Or if you don't agree with that, would you agree that Risker's comment is at least a reasonable way to see the message to Bbb23, and concede that Arbcom's message was reasonable? If checkuser is a difficult area to work in, is it that you don't "respect" that Arbcom is a difficult area to work in? And to begin with, why would you put your criticism in such personalized terms, at all? -- Alanscottwalker (talk) 12:15, 5 April 2020 (UTC)[reply]
I agree with Risker that the individual checks pointed out were not checks that I would run and that pointing this out would be consistent with policy. I would disagree with her that the entirety of the email that was sent read in a way that was consistent with past guidance and policy, and others have commented similarly. When multiple CUs, including past arbs, are reading an email you send as a policy change, at best it is poorly worded. I do think that it was clarified after the fact to something everyone could agree with, but yeah, I’d have been pretty mad too if I had received an email where reasonable people could read it and think policy was being changed and I was being negatively impacted by it. My overall criticism that this ArbCom has been rushing into things resulting in making mistakes along the way stands. TonyBallioni (talk) 14:46, 5 April 2020 (UTC)[reply]
Without knowing the applicable history, or how long we deliberated about the email notice, are you really well-placed to say we were rushing? –xenotalk15:23, 5 April 2020 (UTC)[reply]
Let me clarify: by rushing into things I mean this committee has been substantially more active than any committee in recent memory and looks from the outside to just be trying to get stuff done to say it’s doing stuff. In the process you all desysoped someone based on dubious findings of fact, accepted a case that could have been resolved in 72 hours at AN (Motorsports) that probably wouldn’t have been accepted as a full case by past committees, and lost one of our best CUs by sending an email that multiple others on the team thought contained language that changed policy and wasn’t consistent with prior guidance. That just looks like a committee that wants to get stuff done and doesn’t particularly care about the collateral along the way. TonyBallioni (talk) 15:45, 5 April 2020 (UTC)[reply]
Months of work, and in this matter work over two successive Arbcoms seems hardly rushing anything (time is not infinite for any of us), especially if privacy violations go on in the meantime. As for mistakes even when not rushing (are checkusers rushing around doing privacy violating checks, and are they in too much of a rush to listen to Arbcom and eachother about bad checks), and then defending checkusers for being mistaken, but turning around to insist that Arbcom must be perfect is at best ungenerous. And even though you can't seem to consider you were wrong in not understanding the e-mail, Arbcom is rightly (even in their human imperfection) willing to talk it out in the checkuser forum, even if you begin with accusations (and personalization) against them. -- Alanscottwalker (talk) 16:47, 5 April 2020 (UTC)[reply]
I do not insist in perfection by anyone. I am not perfect, I don't expect ArbCom to be perfect, and I certainly don't expect any other editor on this site to be perfect. I do think this committee has been more active than others and making mistakes along the way. I consider the way this was handled a mistake, but I'm grateful to those committee members like Worm That Turned, Maxim, and GorillaWarfare who were willing to engage and listen to the concerns raised. That doesn't negate my view that this committee needs to reconsider when it is best to be involved. TonyBallioni (talk) 16:54, 5 April 2020 (UTC)[reply]
@TonyBallioni: You have said that you would not have run the examples of Bbb23's checks we presented to the CU team, and you have clarified below that you agree with our interpretation of policy now that the misunderstanding has been cleared up. So why do you still believe the Arbitration Committee should not have become involved here? The Arbitration Committee has a duty to investigate complaints of checkuser misuse, and to act when those complaints are valid. In my view we acted about as leniently as we could have, by sending a private note to Bbb23. But if I'm understanding you correctly, you believe we should not have involved ourselves at all? GorillaWarfare(talk)17:05, 5 April 2020 (UTC)[reply]
GorillaWarfare, no, I'm saying in my view the current committee has trying to be more active and take more on it's plate, and when that happens you get things like emails where you upset people because there's unclear clauses that can be read as a change to a commonly understood norm and past guidance. I'm basically saying "Let's do something" is less ideal when a lot is going on and that more mistakes, even ones of understanding, are likely to be made by a committee that is getting involved with most things sent to it these days. I hope that makes sense. TonyBallioni (talk) 17:13, 5 April 2020 (UTC)[reply]
I can assure you we are not getting involved with most things sent to us. But I do think I understand what you were saying now, thank you for explaining. GorillaWarfare(talk)17:25, 5 April 2020 (UTC)[reply]
I hope this committee doesn't reconsider when it is best to be involved. Motorsports was a bit puzzling to me too but then I think of this perspective. I think at least one of the things that really has you upset - the desysop of BHG - is because the previous ArbCom did the "let's wait" approach that you seem to be advocating. If they had accepted the first case I don't think that outcome happens. What you characterize as "rushing" I characterize as "not willing after FRAM to kick cans down the road". Best, Barkeep49 (talk) 17:07, 5 April 2020 (UTC)[reply]
Not averse to saying the occasional 'the less Arbcom does the better', but seeing how difficult and long their work is, and how little in the scheme of this project they structurally can do and do (to begin with, is there a greater break on human activity, than doing something by committee -- (talk about compounding human imperfection)), I just can't see the problem is so-called rushing. -- Alanscottwalker (talk) 17:49, 5 April 2020 (UTC)[reply]
Yeah, that's fair. Probably a better way of saying it is I'm generally of the view that this ArbCom has tried to take on too much and achieve big results for the things it does take on, and in doing so has made mistakes in either outcomes or communications that in my view resulted in the project being worse off. I think that's a fair critique, and obviously people are free to disagree with it, but it's what I've been trying to get across. TonyBallioni (talk) 17:59, 5 April 2020 (UTC)[reply]
The question is what other approaches are available for long-simmering problems? The arbitration committee could just hand back issues to the community to resolve, but as you said yourself, it sucks even for experienced, respected editors to intervene. I've been trying to think of ways to encourage fruitful intervention, but it's hard to know who can do it effectively. I suggested that editors could pre-identify who are their trusted editors, but I recognize that only the smallest fraction of editors would ever bother, and of that, only some of them will identify trusted editors who truly feel comfortable in raising negative behaviour. I thought of a volunteer group of experienced editors who could take requests to try to figure out who would be a good trusted editor for a given person, but that basically creates a clique of editors that isn't necessarily representative of the community. At least the arbitration committee is elected. Without a hierarchy to make decisions, the only approach most people will take is the same as in real life: voluntarily separate themselves from those they are unable to collaborate with. Unfortunately, this provides an incentive for editors to be more aggressive towards others to drive away those with different opinions (either unknowingly or not). isaacl (talk) 21:46, 5 April 2020 (UTC)[reply]
I think it is most unlikely anyone is motivated by big results (indeed that looks like an assumption of bad faith). If I recall correctly two of this years cases were accepted by last years committee (reinforceing the long-tail nature of Arbcom and suggesting this years committee rather continues the plodding work, like in this CU matter), and all committees only have a very few remedies to begin with, further diminishing doing anything. Thus, the much stronger incentive and systematic bias for Arbcom will always be to do nothing and when they reluctantly get around to it, take half-measures (those almost invariably short of banning, or of removing permissions forever, and a few-times, you maybe have a half-way probation). Read the 1000s of words of case requests statements in total, and regardless of the import of the matter, they are uniformly mind-numbing or soul-crushing further incentivizing doing as little-as-possible. And on top of all that, how these people get to the committee is by an invasive process of community support, entrusting them, usually after years of service and community trust, and then the majority of these trusted and independent thinkers have to be convinced to maybe do anything, at all, against the eternal human allure of passing it off as someone else's work or problem, against the eternal human fear of mistake, and amidst the flood of e-mails. At any rate, in this CU matter, the silver lining of it becoming public, is the non-CU community (99.999 etc percent of us) gets to see even opaquely, that the people Wikipedia has entrusted actually does do their assigned and entrusted review (at least, occasionally.) -- Alanscottwalker (talk) 14:06, 6 April 2020 (UTC)[reply]
From reading the above discussion, it looks like the disagreement is centered around: 1) Whether CU's have unreviewable discretion, and instead should be entirely free of all review, audit, critique, and communication regarding their actions; and 2) assuming auditing is generally valid, whether WP:NOFISHING is meant to completely shield whole classes of CU actions from review and audit, by allegedly prohibiting any review and audit of a CU check where the check has turned out negative.
There's been a long discussion on the checkuser-L mailing list about this. As it turns out, when some examples of "concerning" checks were presented to the enwiki checkusers, the discussion participants seemed to pretty uniformly agree that they were inappropriate checks. Thus, the rationalization that the comments to Bbb23 was in any way a change in policy is not really right. As one of the few people who's held the CU permission for 10+ years, and who spent a lot of time on the now-retired Audit Subcommittee, I don't see anything in the message to Bbb23 that was in any way a change in the interpretation of policy. I will point out to everyone here that, while Bbb23 has stated that he is retiring, he has not resigned as a checkuser, has not turned in his bits, remains subscribed to the checkuser-L mailing list, and nobody on Arbcom (as far as I can tell) has even suggested that his bits should be removed involuntarily at this time. Nobody rewrote policy, and nothing that happened here is new interpretation of that policy. It's the same interpretation that's been in place since I first got the bits in 2009. Risker (talk) 19:48, 4 April 2020 (UTC)[reply]
Risker, thank you. Were comments made to Bbb23 prior to the letter? This seems to have come as a shock to him, a formal warning when one might have expected some more gentle steering first. Guy (help!) 19:58, 4 April 2020 (UTC)[reply]
That's also what I alluded to above (see stricken comment), but Bradv did clarify that this was part of an ongoing discussion. There's no way for us to tell, one way or the other, as far as shock goes. Of course, the loss of Bb23 is saddening, to say the least. El_C20:02, 4 April 2020 (UTC)[reply]
Levivich, respectfully, I don't think that (talk page curation) is indicative of anything. I, at least, am not taking Bbb23's announcement lightly. El_C20:38, 4 April 2020 (UTC)[reply]
El C, not that I'm taking it lightly, but I do kind of feel differently about it. Imagine if I had posted a retirement notice in response to the warning you recently posted on my talk page. If I had done that, that would have been really a dickish thing of me to do to you. It would "pin" my retirement on you. It wouldn't be fair to you. It's really the same thing when an admin pins their retirement on Arbcom as a result of receiving a warning. It's kind of a dickish thing to do to Arbcom. Now we're all (or some of us are) questioning whether Arbcom's warning was justified and asking Arbcom to "defend itself", which the arbs are doing. Of course it's never good when we lose longstanding dedicated contributors, but at the same time, I think this kind of dramatic retirement is not really cool. On top of it, when the retiring editor continues editing to remove posts from their talk page that are in any way slightly critical, I do think it's indicative of the suggestion that the editor is overly sensitive to criticism, as is retiring in response to a warning. I don't know, I don't want to be "gravedancing", but I feel like that needs to be said. Bbb certainly isn't the first or the last editor to retire and take parting shots on the way out, but I feel like we shouldn't condone that kind of retirement. Levivich[dubious – discuss]21:13, 4 April 2020 (UTC)[reply]
Levivich, I understand your point, but I think Bbb23 has earned the right to leave with a protest, whether we consider it due or not. El_C21:22, 4 April 2020 (UTC)[reply]
leave with a protest - Assumes facts not in evidence. The number of longtime editors who have "left with a protest" and stayed gone is small compared to the number who have "left with a protest" and returned after their "leaving" has leveraged maximum effect. Real protest retirements more often occur following a cool-down wikibreak. I could be wrong in this case, but the objective reality favors a skeptical view. ―Mandruss☎23:39, 4 April 2020 (UTC)[reply]
Anyway, for me, the level of one's contributions to the project plays a key part in that determination. El_C23:53, 4 April 2020 (UTC)[reply]
I support the ArbCom decision and hope that we see more investigations like these from ArbCom in the future. Also, Bbb23 always doesn't like to be criticised. I have noticed this since I came to Wikipedia. Bbb23 isn't the only admin like this, many admins hate to be criticised which is why admins are driving away editors from this project.
I am only here to help grow this project and to make it more reliable, I don't want unknown checkusers to see my private information. I wish I can see how many times a CU editor asked about my private information. Can we make this real? Maybe I should propose this. Every editor has the right to know if his private information was exposed. If I see a CU editor violating the checkuser policy, I am going to report them to the ArbCom.--SharʿabSalam▼ (talk) 03:57, 5 April 2020 (UTC)[reply]
The CU log isn’t public because if it was it would expose your private information to the public. You could require notification anytime an account is checked, but that’s unlikely to gain consensus because there are cases where the reasons behind the checks are sensitive and you do not want the person being checked to know: example, an administrator is running accounts with tens of thousand of edits like happened in the last ArbCom election.Also, there’s no requirement that my posting a comment on ACN saying the 2020 ArbCom has been rushing into things headlong and making mistakes this year because of it have a neutral header. This isn’t an RfC. As Ivanvector pointed out above, there are several CUs who have criticized the email ArbCom sent as being inconsistent with prior guidance and the actual policy. I think that wouldn’t have happened if this ArbCom took a more cautious approach than it has been. Since the email was sent and this thread was started, I think everyone has come to an agreement on what they view as the requirements of policy, myself included. This was intended as public criticism of an elected body who made an unforced error here. Requiring a neutral header for that makes no sense at all and there’s no policy requiring it. TonyBallioni (talk) 04:32, 5 April 2020 (UTC)[reply]
Except it's not clear at all that they made an error - at least it's not clear to me as a noncheckuser that they did especially in light of comments by people like Risker. Acted imperfectly? Of course it's made up of humans but it's not at all clear that the underlying action - asking Bbb to act in-line with established policy and doing so privately in the hopes that drama could be avoided - is an error let alone an unforced one (as xeno mentions above we have no idea what triggered this discussion so who knows what the inciting (forcing) incident was). Best, Barkeep49 (talk) 15:46, 5 April 2020 (UTC)[reply]
Regardless of whether ArbComm's cautioning of Bbb23 was correct, his reaction reveals his unsuitability for high responsibility as he appears to completely reject any notion of accountability for his actions. Once someone puts themselves above the project, and sees themselves as not being answerable to anyone, and not needing to explain themselves, be it to the community, to ArbComm or to other CU's, then it is time for them to go. SolTrek (talk) 04:14, 5 April 2020 (UTC)[reply]
Bbb23 is allowed to disagree with Committee directive and resign in protest. They have contributed much to the project, but that seems to be overlooked by some here, which is a shame. El_C04:32, 5 April 2020 (UTC)[reply]
Oh you mean when Bbb23 blocked you for socking because you edited while logged out, and you said that wasn't socking, and then you discovered that WP:LOGOUT lionks to WP:ILLEGIT, so you admitted that you were technically socking? [1] And this was after you were re-instated to editing and admitting to having had a previous sock account? [2] You mean that kind of "abuse"? I'd call that being a good CU. Beyond My Ken (talk) 08:35, 7 April 2020 (UTC)[reply]
The purpose of Wikipedia's policies and guidelines is to provide a framework for the smooth operation of the project rather than to provide admins with technicalities on which to block productive editors for sport. Yes, technically, I was socking. You got me! Or did you? Continue reading to find out.
First, since you've asked, this completely unrelated AN notice you linked to is from 6 years ago, when Bbb23 "welcomed" me to Wikipedia with an absurd indefinite block, one that got promptly (I wish) vacated resulting in what can only be described as Bbb23 grasping at straws to get his way and have his bad block reinstated. This included him demonstrably (please, do ask me to demonstrate it) lying about my edits as well as opposing my username change request when it became clear the suboptimal username was his/her one last hope of making a reblock happen (spoilers: a reblock didn't happen). I did create another account at the time, after the block, unfamiliar with Wikipedia's policy, making no attempt whatsoever to conceal the connection between the two, but that's neither here nor there as you can't expect a new user to be familiar with the way things are done. That second account was blocked upon my request after I had the multiple accounts situation explained to me. Technically, sure, "socking" did occur but let's be real here: if we go by the spirit of the policy rather than robotically following its letter, there was none. Also, Bbb23 didn't have CU access at the time, and the block itself had nothing to do with socking or CU checks.
Now let's discuss the abusive CU block, the one that actually matters here as opposed to the 2014 abusive non-CU block that I have no idea why you even brought up. To make matters clear, there were two concurrent CU blocks, a three-month block placed on my IP range and a one-week block placed on my registered account. I will refer to them simply as "the block" to make my explanation easier to follow. Here's what was wrong with it. 1) The block was placed on an IP range with nothing but good edits from it. 2) The block was placed on a registered account with also nothing but good edits from it. Don't ask me why a block that was supposed to prevent me from IP editing was extended to my registered account because I have no answer to give you. Bbb23 works in mysterious ways I suppose. 3) The block was based on a CU check with no basis in the global CU policy which states in no uncertain terms that the CheckUser tool "must be used only to prevent damage to any of the Wikimedia projects. 4) It was what I call a sudden block. Not a soul complained about any of my edits (unsurprising, considering there was nothing wrong with them) which, from that particular IP range that was blocked, spanned over 12 months if memory serves. Bottom line: the CU check was frivolous, and the blocks were not preventative of anything that needed to be prevented. In fact, it is my strong belief the blocks were dished out as a means of making the frivolous CU check appear legitimate (after all, if a check results in blocks, it must've been good, right?), which is a very unpleasant thought indeed but to this day I can come up with no other plausible explanation. Another, even more unpleasant, theory that I have only shared privately up to now is that the ridiculously high volume of Bbb23's checks is to a large extent the result of him/her going through histories of various project pages and indiscriminately using the CU tool against every single IP in there in hope of "catching" logged-out editors so he can then block them for what technically constitutes "socking" without even looking at the actual contributions. Since I have no access to the CU logs, it's mere conjecture that almost reads like a personal attack, but is it really just that? Based on what I've experienced first-hand, I don't think so. It's a good thing the Committee has finally decided to wake up and clean house. A timid warning issued in private is nowhere near good enough though. Bbb23's overreaction to it demonstrates that point better than anything else ever could. Here's a revolutionary thought: if an admin or a functionary finds themselves unable to operate within the norms of the Wikipedia community, perhaps they shouldn't be policing its members. Iaritmioawp (talk) 13:47, 7 April 2020 (UTC)[reply]
Break
To hopefully help with clarity and transparency, particularly with respect to the allegations made above that we are somehow changing policy, here is the relevant part of the email that ArbCom sent to Bbb23:
Invalid reasons for checking accounts include the following:
Suspicious new users.
There must be clear evidence of misuse of multiple accounts. Just being a new account is not enough.
Creating a new article.
If the article is spam, treat it as such. The CU tool should not be used without evidence that multiple accounts are being misused.
Editing a contentious topic.
Many of our new editors get involved because they see an error and want to fix it. This is encouraged, even when the topic is considered contentious.
Commenting at ANI, ANEW, the TEAHOUSE, RfA, or noticeboard.
Editing project-space must not trigger an automatic check; new users are allowed to ask for help or report problems.
An editor returning after a hiatus, with no evidence of disruption or sockpuppetry.
There must be onwiki evidence that an account has been compromised. This should not trigger an automatic check.
A clueless newbie making newbie mistakes.
Help them if they need help, don't violate their privacy or try to find reasons to block them. We must AGF, especially for new editors.
This letter was voted on by the committee, which maintains that these types of checks, particularly when performed routinely, violate the the CheckUser policy. The same policy also states that the use of the CheckUser tool on the English Wikipedia is monitored and controlled by the Arbitration Committee, and CheckUsers may have their permissions revoked by the Arbitration Committee for misuse or abuse of the CheckUser permission.
I can understand and appreciate that Bbb23's sudden retirement came as a shock to the checkuser team, and it is entirely reasonable for people to question ArbCom about how we got to this point and whether our actions were appropriate. While we are not prepared to share the details of the investigation or our deliberations, I will contest the accusations that we are "overactive" or "rushing into things". This was a long time coming. – bradv🍁15:54, 5 April 2020 (UTC)[reply]
That seems pretty reasonable, and to be merely restating policy as it is broadly understood and expected. And since Bbb23 has "left" of his own accord, I guess this can be closed and we can all GoAndRightAnArticleOrSomethingTM :) ——SN5412916:04, 5 April 2020 (UTC)[reply]
Well spotted, Thryduulf, although for once it was deliberate: "...WriteAnArticle" would exclude colleagues who may not write, but do maintain the encyclopedia in other ways. The play on words is intended to include editors who "right" (i.e.undo damage to) the thing as well as "write" the thing :) ——SN5412918:02, 5 April 2020 (UTC)[reply]
Yes, this is the portion of the letter that generated controversy: as I said above, being an editor who is clearly not a new user where there is reasonable suspicion of it being a sockpuppet but where the original master cannot be identified has been seen as grounds for running a check so long as there is reasonable grounds to believe there was a policy violation. Many items in the list above were listed as policy violations that along with an account showing signs of not being a new user would be valid grounds for a check in the email sent by the committee in April 2019.Since it was sent and several of us objected to it saying that being a suspicious account indicating signs of sockpuppetry in one of these areas has been seen as a traditional grounds for a check for years, which is why several of us objected it it (Ivanvector and I probably the loudest, but there were a few others), it has been clarified that the meaning was that these on their own were not grounds, and there needed to be suspicion of use of multiple accounts, but that if there were reasonable grounds to believe that a new user wasn't new and was editing in a way that suggested reasonable concern about misuse of multiple accounts (which each of the bullet points is) a check would be valid, which is where we left it and I think everyone agreed. TonyBallioni (talk) 16:16, 5 April 2020 (UTC)[reply]
Tony, you are acting as though we have said that "suspicious new users" may never be checked, no matter what. If there is evidence that a suspicious new user is engaging abusive sockpuppetry, that is justification to run a check, and both our email and the WP:NOFISHING section of the checkuser policy are clear on that. But simply being a new user and acting "suspiciously" are not valid reasons alone. As you know, new accounts are generally considered "suspicious" when they know too much (finding noticeboards easily, using intricate syntax) and there are plenty of valid reasons new users can know about Wikipedia: they were active on other projects, they have WP:CLEANSTARTed, they were editing as an IP for a while before creating an account, etc. That is what we are trying to clarify. GorillaWarfare(talk)16:59, 5 April 2020 (UTC)[reply]
Yes, I think you have since clarified that: as I said, I think we are all on the same page now, but like I said on the list: when you have several people saying to you that this email reads as being a difference from policy and previously communicated guidance, that shows it probably could have been worded better. Why people are angry is that under a strict reading of the above, you could probably find issues with any active CU since those are a listing of the most common policy violations of WP:SOCK. That's the concern people had.What we do agree on is that simply being a new editor in those areas is not enough unless there is an indicator that their behaviour there would constitute a violation of the sockpuppetry policy. I don't think anyone, myself included, has disagreed with that, and like I've said multiple times here, I think everyone is on the same page. Also on a personal note, thank you for being so willing to engage. It is noticed and appreciated, GorillaWarfare, especially when I'm being prickly :). TonyBallioni (talk) 17:13, 5 April 2020 (UTC)[reply]
I think we may have crossed edits a bit—I deleted my comment (the 16:59 one) with the summary "the linebreak threw me off, did not see the second portion of your comment. Appears we agree on this point." I think you may have started replying before I did so, though, and the comment returned as a result. GorillaWarfare(talk)17:22, 5 April 2020 (UTC)[reply]
Yes, many edit conflicts and I screwed up the formatting :) Like I said a few times here, I think everyone is finally in agreement. TonyBallioni (talk) 17:27, 5 April 2020 (UTC)[reply]
On the subject of edit conflicts, I have to admit I was an edit conflict away from posting something earlier in this thread which, upon further reflection, I probably would have regretted. So, hurrah for edit conflicts :-) -- RoySmith(talk)17:45, 5 April 2020 (UTC)[reply]
The vague wikispeak "editing in a way that suggested reasonable concern about misuse of multiple accounts" could be used to justify any check at any time.
Let's cut to the chase. As an outsider, I think this is about: if a CU sees one account that they think is "suspicious", for whatever reason, can they run a checkuser on that one account in order to try and find other accounts that may be socks, OR does a CU need to have at least two accounts in mind, suspected of being controlled by the same person, in order to run the check to see if those two accounts match? In other words, can you run a check on one account to see if it matches any other accounts, or can you only run a check to see if two or more specified accounts match with each other. This is what I understand the core of the NOFISHING policy to be. To me, in order to have "suspicion of use of multiple accounts", one must have identified at least two accounts that are suspicious. One cannot have a reasonable suspicion of the use of multiple accounts if you can't name what you think the second account is. It seems some checkusers want to CU a single account if they think it's suspicious to look for a second account, even if their suspicion doesn't tell them what the second account is. In other words, can you check a "sock-with-no-master". Do I have that right? Is that what this is about?
As an editor who was checkusered as a new user simply because I wandered into the back room too early (and who has had to deal with socking accusations as a result ever since, even a year later), this issue hits close to home for me. Levivich[dubious – discuss]17:10, 5 April 2020 (UTC)[reply]
This is actually addressed directly in the WP:NOFISHING portion of policy: For example, it is not fishing to check an account where the alleged sockmaster is unknown, but there is reasonable suspicion of sockpuppetry, and a suspected sock-puppet's operator is sometimes unknown until a CheckUser investigation is concluded.GorillaWarfare(talk)17:23, 5 April 2020 (UTC)[reply]
Yeah, what GW said. This is explicitly allowed by policy so long as the other requirements are met, so you would need a change to get rid of it, which is why I and a few others were originally concerned. We get these requests in private fairly frequently from members of the community and it would be a pretty drastic change. TonyBallioni (talk) 17:38, 5 April 2020 (UTC)[reply]
(edit conflict) All "fishing" means is "to check an account where there is no credible evidence to suspect sockpuppetry". It is an incredibly serious accusation because it means a checkuser has gone willy-nilly and checked users without reason, hoping to "fish" out sock puppetry. The policy does not require a CU to have at least two accounts to compare; sometimes a single account exhibits behavior that amounts to "credible evidence to suspect sockpuppetry". Mz7 (talk) 17:43, 5 April 2020 (UTC)[reply]
Mz7, I think you're right, "credible" is probably a better word than "clear", as we used it in criterion #1. By emphasizing the point we may overstated it, and that is duly noted. – bradv🍁17:49, 5 April 2020 (UTC)[reply]
I think Mz7 has said what everyone has been trying to say on this thread and multiple emails in one paragraph. Someone should buy him alcohol of his choice while maintaining social distance :) TonyBallioni (talk) 18:05, 5 April 2020 (UTC)[reply]
Thank you to everybody for their comments. I just want to clarify that my earlier request for an RfC should not be interpreted to mean I had taken a side. I was just looking for a mechanism which would bring more information to light. At this point, there's been enough said, by both arbs and CUs, that I'm satisfied I know as much as I can reasonably expect, or need, to know. -- RoySmith(talk)16:52, 5 April 2020 (UTC)[reply]
Thanks. That does indeed seem like a fairly uncontentious assessment of expectations in line with general SPI practice. I do see how it can be controversial if the CUs have traditionally been granted a good deal of discretionary leeway in terms of checking obviously non-new accounts based on their own judgment, and they're worried their discretion is being curtailed, and that they made a chilling example of the #1 CU, but at the same time GW says the standard of acceptable discretionary judgment was measured against the general, mainstream practices of the CU team, which does not sound particularly unreasonable. ~Swarm~{sting}17:09, 5 April 2020 (UTC)[reply]
It seems much too much, that this is over the exceedingly short elliptical phrase "on their own", especially in a long message, but good to know Arbcom is taking its duties seriously and will talk it out. Alanscottwalker (talk) 17:21, 5 April 2020 (UTC)[reply]
If Bbb23 is retired, does he need the CU - or admin - privilege any more? It seems unnecessary for him to have it if he won’t be using it. 69.238.208.194 (talk) 19:26, 5 April 2020 (UTC)[reply]
Except for this: "A user who resigns from a role but does not then avoid the concerns of their peers, does not cease engaging in discussions about their behavior, and does not seek to evade dialog or prevent resolution of the concern, might or might not be sanctioned in the end, but they would not usually be described as resigning "under a cloud" since the matter was allowed to resolve and a conclusion reached at the time. (This might be the case for someone who resigns early on in a matter because of respect for community concerns.)" - the problem is that they are avoiding the concerns of their pears and refuses to engage in discussions about their behaviour and is evading dialogue to resolve the concern. There is a behavioural issue here in regards of lack of accountability and refusal to discuss their behaviour. This shows a temperamental unsuitability for the position and it is my understanding that this lack of accountability and refusal to discuss or allow critique of their actions is a longstanding and chronic issue. SolTrek (talk) 01:17, 6 April 2020 (UTC)[reply]
The point remains that Bbb23 has not resigned, and so "resigning under a cloud" does not apply at all. If Bbb23 returns to activity and does not follow policy, we will act accordingly. He has read our message to him, which is all we needed—if he wished it to be a continued dialogue we certainly would participate, but we are not requiring it of him beyond the dialogue we already had before issuing the warning. GorillaWarfare(talk)01:25, 6 April 2020 (UTC)[reply]
If Bbb23 does not return within the next ~3-4 months his permissions will be removed for inactivity. If at some point after that he wishes to request the tools back the committee will need to decide whether to (a) grant them as they would to a returning user who became inactive while in unambiguously good standing; (b) require further discussion about the current situation; or (c) simply refuse to regrant them. I don't know currently what my vote would be if I had one, but the CU policy would seem to allow any of the options and so its probably worth the committee members thinking about it now while things are fresh - an extended period of time thinking about it (while possibly simultaneously bringing a future committee up to speed) at the time of request wouldn't be fair to him imo. Thryduulf (talk) 01:41, 6 April 2020 (UTC)[reply]
I am not prone to using this kind of language, but this move by ArbCom is a true outrage. To hobble the war against sockpuppetry by restricting one of the very best CUs that we have is cutting off your nose to spite your face. I would like to see a clear and comprehensive accodunting of who voted for this very stupic move, and who voted against it, but I do warn you that every single arbitrator who voted for it will be opposed by me at the next election. I suggest that before you have another rebellion on the order of the Fram debacle, that ArbCom rescind the warning, apologize profusely to Bbb23, and ask him to please return to his post. This is just madness. Beyond My Ken (talk) 06:55, 6 April 2020 (UTC)[reply]
And I will note that there are commenters above who have good reason to be happy about Bbb23's being provoked into resigning, since, sooner or later, they will inevitably slip up and reveal what banned or blocked account they are a sockpuppet of. Beyond My Ken (talk) 07:09, 6 April 2020 (UTC)[reply]
Most of us aren't privy to the information ArbCom used to make this decision, we don't know what the dialogue between them and Bbb were leading up to the warning, and as can be seen in the conversation above, they've been open and upfront in answering the questions that have been asked, as far as possible while maintaining privacy. Answers that seem to have been accepted, even if not fully agreed with, by even those critical about what's happened. You're free to vote for whomever you wish in next year's ArbCom election, but I don't agree with you that the committee is out of line with community expectations at this time and a "rebellion" won't help anybody. Personally I'm sorry to see Bbb23 go, and I certainly hope they'll come back and resume their excellent work, but if the committee felt there was a need to act then I believe them. — Amakuru (talk) 07:17, 6 April 2020 (UTC)[reply]
Beyond My Ken, as explained elsewhere, this was a warning to Bbb23 to act within our guidelines. We had given a general reminder to all CUs last year, which he did not heed.
I do not expect that we will be releasing the voting, suffice it to say that it got a majority of arbs. I will say that I voted for this email to be sent. I have since seen further statistics that have surprised and concerned me and I do not regret the actions taken by the committee.
Now, Beyond My Ken - if that means you don't vote for me in 2 years (assuming I'm even running), I can live with that. indeed if you want to make a song and dance about it, you can use my recall system on my user age as an arbitrator recall - send a message to the committee that the community isn't happy. It's set up to do that, if you feel strongly enough.
There are privacy issues, sure - I wouldn't expect that ArbCom would release the names of those editors who complained (although I'd love to see it and evaluate who they are), but there ia absolutely no reason why our elected representatives shouldn't release the names of those Arbitrators who voted in favor of the restriction and those who did not. That information should be made public immediately, so we know who's responsible for this terrible, ridiculous and harmful decision. There is absolutely no justification for keeping that information secret. Beyond My Ken (talk) 19:25, 6 April 2020 (UTC)[reply]
Beyond My Ken, this wasn't a public decision. The committee does not need to release it's voting on every point. In this case, a pure vote tally tells a very different story to the statements made. For example, what if an arb voted oppose as they believed the email was not strong enough? There is no way to release this tally accurately without releasing additional information, and these deliberations were necessarily private.WormTT(talk) 19:49, 6 April 2020 (UTC)[reply]
It may not have been intended to be a public decision, but the decision itself is now quite obviously a de facto public matter, so there is no longer any justification for hiding behind privacy issues. I think the community has a right to know who supported this decision, and who did not. If it was unanimous, say so. If ArbCom as an institution is too afraid of releasing the information, then I call on each Arbitrator individually to say, here in this forum, whether they supported the decision or not, or abstained. We trust ArbCom to be well aware of privacy concerns and heed them, which is why I strongly supported ArbCom throughout the T&S debacle, but this debacle is of your own making, and there is no provacy issue involved, since Bbb23 has revealed the gist of the restriction warning. You no longer have a leg to stand on in not releasing how our elected representatives voted on this matter. If you believe your decision was right, then there should be no reason to not stand behind that decision, other than your own personal discomfiture. You folks were on the right side of the Fram mess, but you're on the wrong side of this one. Beyond My Ken (talk) 22:43, 6 April 2020 (UTC)[reply]
Also, if there are nuances such as you posit, i.e. that an Arb voted against it because the warning wasn't strong enough, the obvious answer is not to release a simple "yes/no" list, but to have each Arb say why they voted yes or no. I'm not asking for a release of the transcript of your discussions, as telling as that might be, but there simply cannot be a privacy issue when elected representatives are voting on sending a warning in the name of the community. You've taken an action, supposedly as our proxy, which was secret because of inherent privacy issues, but those don't exist anymore, so telling us how you voted is simply a no-brainer. I think it's called "transparency". Beyond My Ken (talk) 22:50, 6 April 2020 (UTC)[reply]
I don't think you've read the room accurately. Several people were upset when this news first broke, but most of them have settled down and moved on, even if not all are fully pleased with the outcome. The arbs have done an excellent job of responding to input here. There has been no debacle and it hasn't been helpful for you to storm in here guns blazing after the dust had settled. LEPRICAVARK (talk) 23:39, 6 April 2020 (UTC)[reply]
I disagree. I think that you're quite happy to "read the room" the way you wish it to be. I came here when I found out what happened. I would have come earlier if I had heard earlier. The mistake that ArbCom made hasn't become less of a mistake because a mere couple of days has passed, and their obligation to the community to have as much transparency in their decision as possible hasn't evaporated either. There are fundamental issues here that need to be addressed by the Committee which, so far, is simply waving its hands and telling us we're not looking for these droids. No one has addressed the bottom line that an action which was intended to be private is not longer private, and therefore the changed circumstances requires them to behave differently in their obligation to the community, or the fact that their actions caused significant damage to the community's ability to keep sockpuppets at bay. I'm sure that wasn't their intention, but that was the ultimate result -- and, if we listen to the other CUs who have commented here, that standard which is being upheld by the Committee is not in line with the standard as understood by the community or by the CU corps. If that's the case, there are definitely issues to be discussed, so let's start with the most immediate and easiest one, which is who voted how, and how they justify their votes. Beyond My Ken (talk) 01:28, 7 April 2020 (UTC)[reply]
Beyond My Ken: what has been made public are the types and nature of checks that we asked Bbb23 to refrain from performing. I don’t see anyone arguing that these types of checks are permitted by local or global policy, and the committee - in its entirety - is bound to ensure that these policies are followed. –xenotalk01:37, 7 April 2020 (UTC)[reply]
I'll comment here since I started this thread: while I'm not entirely convinced of the method ArbCom used here to look at this in terms of assessing what the risks the policy is trying to prevent are and the scope of the problem combined with the use of judgement, as Risker has pointed out there's general agreement amongst the CUs who have commented, myself included that we would not have run individual checks pointed out by the committee. When combined with the scale of his activities, I think because of the gross amount of potentially similar checks, sending an email probably made sense.I think the email was poorly worded: even though everyone is in agreement now, when the email was first forwarded to us, there was disagreement between members of the team and the committee as to the language posted above, that's a fairly good indication it could have been improved, just as the fact that I've had to explain myself to reasonable people multiple times here in some of the things I have said is an indication that I was not clear as I should have been in my communication. Mistakes happen in communication.BMK, while I'm probably much closer to you and Softlavender on this than I am to xeno's position, I think the important thing here in the big picture is that there is a general understanding between the CU team and arbs who have commented as to what the policy means and what has been accepted previously as practice. There will be disagreements in judgement going forward, that's the nature of judgement, but an understanding of principles is the most important thing.While I certainly hope Bbb23 returns, and I do think there are lessons to be learned here (and I'm not exempting myself from the group who could learn from this), ultimately I don't think it does the project, Bbb23, or the team any good to make this particular discussion any more heated, and think now is the time to deescalate. There will be ArbCom elections in November. People will have the opportunity to vote and write guides then. I'm not trying to get anyone out of accountability here, but I think for what was essentially a private warning to someone that they decided to make public, the arbs have been as transparent and engaged as probably could be expected, which I thank them for, even the ones who I've had harsh words for. I probably should also apologize for any words that were unnecessarily harsh while I'm on the topic. People of good will can and do disagree in good faith, and sometimes the best way to handle that is moving on once common ground has been found. Sorry if I waxed poetic there. TonyBallioni (talk) 02:05, 7 April 2020 (UTC)[reply]
There was no restriction issued. Arbcom presumably checked some of Bbb’s 50+ looks per day into private user data, found a few potentially or directly inappropriate checks, and reminded Bbb privately about the guidelines. As a result Bbb diva quit and the whole thing turned public. Bbb is free to resume checking thousands of accounts per month anytime they want. Arbcom is completely justified in maintaining the standards of the CU right. Mr Ernie (talk) 07:18, 6 April 2020 (UTC)[reply]
I'm confident that there is not going to be a rebellion similar to what happened during FRAMGATE. A large part of the reason that so much anger was directed at T&S was that they stubbornly refused to explain their actions. Whether or not one agrees with ArbCom's actions here, it's plain to see that the arbs have been prompt and responsible in answering questions and explaining why the email was sent. How you choose to vote is your own business, but I sincerely hope our arbs aren't making decisions with the aim of helping themselves get reelected. LEPRICAVARK (talk) 14:38, 6 April 2020 (UTC)[reply]
It sounds like Arbcom did their job and gave a mild warning. Bbb23 should accept it as guidance (which we can all use) and come back. North8000 (talk) 11:50, 6 April 2020 (UTC)[reply]
I agree with Tony Ballioni. ArbCom has turned into a witchhunt against longterm effective admins, apparently believing it has a remit from T&S to rid us of all the most productive admins on the slimmest grounds possible. I am taking careful note of all the current arbs who are enabling and supporting these actions regarding the four admins (so far since January) in question, which (ArbCom actions) are in my mind clearly doing immeasurable harm to the project, and I will vote against them in the next ArbCom election. I also hope that all of the current arbitrators are going to spend several hours of each day to take up the slack at SPI that Bbb23's absence causes. He took a wiki-break a few years ago and SPI got so backed up it took weeks to get any SPI report looked at. Softlavender (talk) 00:18, 7 April 2020 (UTC)[reply]
Softlavender, it's comments such as yours in this thread that help foster an environment in which well-connected individuals are able to get away with harmful behavior simply because they have enough influential friends. The current ArbCom has earned my respect by taking on tough cases and refusing to let prominent editors off lightly for poor behavior. While these actions may anger those who put personalities above principles, the current arbs are to be commended, not condemned, for their courage and commitment to doing their job in an equitable fashion. LEPRICAVARK (talk) 01:53, 7 April 2020 (UTC)[reply]
Softlavender has every bit as much right to voice her opinion as anyone else here. And remember - the door swings both ways. Who is connected to who perhaps is the question. I haven't followed any of the recent proceedings, as least not any since Framgate - so I'm not taking any sides here; but it does trouble me when one user tries to shut down another user from voicing their views. — Ched (talk) 02:19, 7 April 2020 (UTC)[reply]
Ched, I am not seeking to shut anyone down. I am very much aware that Softlavender is entitled to express her opinion, which is that ArbCom has adopted the goal of deliberately ridding the community of productive admins. I am merely voicing my opinion that this ongoing vilification of the arbs is harmful to our community. It contributes to an atmosphere in which those who are well-connected are not held responsible for their behavior because those who attempt to do so face an onslaught of anger. LEPRICAVARK (talk) 03:06, 7 April 2020 (UTC)[reply]
Indeed. Bottom line here is that he was doing something he should not have been doing. And then got caught. And then had the chutzpah to complain about it. A private warning that he made public. Makes you wonder how long the mis-use of tools had been going on for, and the wider implications of such actions. LugnutsFire Walk with Me06:52, 7 April 2020 (UTC)[reply]
You want to talk about chutzpah? How about an editor who's been an admin-basher from way back, almost never has a good thing to say about an admin -- unless, of course, they want something from an admin, in which case they become a silver-tongued sycophant blowing smoke up the admin's ass. Now that'schutzpah, among other things. Beyond My Ken (talk) 08:26, 7 April 2020 (UTC)[reply]
Only bash the bad 'uns and the ones that break the trust of the community. Shame your buddy can't help you out now each time some goes to to 3RR about your conduct. LugnutsFire Walk with Me09:18, 7 April 2020 (UTC)[reply]
You bash 'em all, actually, but that doesn't matter, because Bbb23 is not, and never has been a "bad 'un', "buddy" = and that's precisely the problem about what's happened here. ArbCom's actions provoked the loss of by far our best CU (no offense meant to our other fine CUs), and that harms the community, but ArbCom is unwilling to take responsibility for what they did by revealing the voting on sending out the warning. We've been told that there was a "consensus", but -- as we all know -- a consensus does not require unanimity, so it's likely that some were for it, and some were against it. Considering the damage their action resulted in, and the transparency that the Committee owes the community now that a private matter has been made public, asking for the voting to be revealed to the community is a reasonable request. Every Arbitrator ought to be willing to stand by their decision and, if necessary, make arguments justifying it. That doesn't appear to be the case. What happened to Bbb23 made me mad, but what ArbCom is now doing makes me sad: I thought these were people we could count on to behave ethically when the chips were down. Beyond My Ken (talk) 13:27, 7 April 2020 (UTC)[reply]
Lugnuts specified both 'bad 'uns' and those who break our trust. As far as I am concerned, few actions warrant the loss of trust more than the misuse of CU privileges. In fairness to Bbb23, I admittedly don't know how serious this misuse was. Evidently it was not too extreme since ArbCom merely issued a warning, but it is nevertheless troubling to know that our most prolific checkuser was violating policy, possibly because they did not fully understand the policy under which they were operating. Perhaps if Bbb23 had stuck around to answer community questions about their actions, we might have more information. LEPRICAVARK (talk) 13:44, 7 April 2020 (UTC)[reply]
Beyond My Ken, you seem to be ignoring why arbcom sent a policy reminder. It was because Bbb wasn’t adhering to the CU policy, which deals with sensitive and private user data. What exactly is unethical about that? Mr Ernie (talk) 13:51, 7 April 2020 (UTC)[reply]
One thing that will never help is the conflict between ideals (privacy) and technical realities. I'm not convinced that it was literaly "mis-use of tools", even if it went beyond common interpretation of policy. CUs are already trusted to be able to look at some information the general public has no access to. Because of the way processes work with the project's ideals, they must also justify each access. This is an incredibly unefficient system to deal with the persistent problem of socking and there also appears to be a gray area (if I understood the above, CUs were somewhat divided on the interpretation of policy or on technical issues). I understand that on the mailing list a consensus eventually formed before the warning, however. If mis-use of tools was so blatant, CheckUser right would likely have been removed (and could be via a case, of course, just like the admin rights)... —PaleoNeonate – 07:52, 7 April 2020 (UTC)[reply]
Funny that BMK is selective on the crux of this matter - that Bbb23 got caught doing something he should not have been doing, and goes down the personal attack route instead. I guess that's Ken's modus operandi. An admin who doesn't like to be told he's in the wrong and doesn't take constructive feedback about their actions. That's a bad combination. LugnutsFire Walk with Me14:47, 7 April 2020 (UTC)[reply]
Giving bad behavior a free pass because the offender is "longterm" and "effective" is such a well-known cancer on Wikipedia that it's humorous that you would tout it as some sort of immunity unironically. I don't care how "longterm" or "established" you are, if you're routinely violating policy someone has to draw a line in the sand eventually. I've criticized this Arbcom's draconian measures myself but this is not even a draconian measure. If anything it seems like they've listened to feedback. Bbb was warned, continued violating policy, and was *gasp* warned again. ~Swarm~{sting}19:33, 7 April 2020 (UTC)[reply]
I've had a look through my email archives to see if I complained about Bbb23 - I haven't, though I see references to a few edits such as this and this and this which generated off-wiki discussions about their behaviour with various users, and all of which ended concluding that there wasn't much point in trying to take any action over it.
The principal problem with Bbb23, as least from my viewpoint, is nothing to do with sockpuppetry or checkusers. I've said a couple of times I really don't care how the encyclopedia gets written as long as it gets done, and discussions about who did what where and when and playing amateur detective are of no interest to me - if somebody evades a block and is disruptive in exactly the same way as before, we can just block them for the same thing regardless. I don't really mind what people working in that area of the project do and leave them to their own devices. Rather, my problem is with unacceptable communication. If you take action against someone, explain why you've done it using words they understand, or get a completely uninvolved admin to explain it to them if you're getting nowhere. If you can't back up and defend your actions, you shouldn't be making them, and that's policy.
As far as I can tell, all Arbcom have done is have a quiet word to get Bbb23 back on track and be a productive admin who gets on well with everyone, not just the corpus of people who do the same sort of work as them. The resulting flounce suggests that Arbcom's action was justifiable in this case. I hope that the criticism over the conduct is taken with the right spirit, and they learn how to become a productive administrator. If they can't do that, then I wish them well in whatever else they decide to do in life. I have to say I am scepitcal of this, as they have scrubbed a bunch of critical messages over their conduct since "retiring", which leaves me to think the opportunity for remorse is slight. Ritchie333(talk)(cont)12:23, 7 April 2020 (UTC)[reply]
I agree, belatedly, that the discussion on the mailing list led to a consensus, eventually, that Arbcom's actions were warranted here. Speaking only for myself here: this was another incidence of the Committee communicating a decision to affected community members poorly, and then reacting poorly to criticism, leading to unnecessary drama. The way this was communicated and subsequently handled reminded me of the 2FA guidance from about this time last year, where the Committee produced a notification to admins which appeared to require admins to enable 2FA, and when questioned initially insisted they had policy backing to make such a requirement (they did not). In this instance, the very first that any of us heard that there was any investigation in progress was Bbb23's note about it. The cu-l thread started with a post from a checkuser asking if anyone else on the list had heard anything about it and/or knew any details; we had not and did not. When other list members also pressed for details, a few members of Arbcom responded somewhere between vaguely and aggressively that they had investigated a complaint and that their interpretation of the policies violated were correct (in a way which implied that their interpretations are above criticism) but not the nature of the complaint or any details about the investigation. When pressed further, arbitrators responded with investigative metrics which are clearly against policy, such as that a high rate of checks not resulting in a finding of sockpuppetry was proof in and of itself that the checks were inappropriate. At least one arbitrator chastised the checkusers as a group for having not detected the issues with Bbb23's checks on our own, still without having said what those issues were. So, yeah, there was animosity.
It was only after a different arbitrator provided specific examples of checks they considered problematic that we really began to have a constructive discussion, and after that Risker has already accurately described how things went afterwards. We were shown four specific checks which depending on which arb was commenting were from a day of a few dozen or "hundreds" of checks. Every CU who commented afterwards agreed that those checks appeared to be problematic, although we were not privy to Bbb23's justification for them so we're still commenting at least partially blind.
My only real point here is that there was no reason for the Committee to have been so opaque with the checkusers on our own private mailing list. I don't know what they were trying to accomplish by shutting us out but it only led to drama. I'm purposely not naming names but I'm sure the arbs and CUs know who I'm referring to, and so a belated thank you to those arbs who treated us as trusted peers rather than petulant subordinates. And I hope the others will take this in the spirit of constructive criticism from which it's intended, though your actions in recent memory do not inspire confidence. Ivanvector (Talk/Edits) 14:01, 7 April 2020 (UTC)[reply]
Ivanvector, Unfair. You already know there was reason for the committee to do it privately. When a committee has already decided to have a private communication, it is unreasonable to expect individuals to blab all about it on a dime. I am sure if you consider, there might actually be good faith reasons for not dragging out another CU in-front of you. -- Alanscottwalker (talk) 15:03, 7 April 2020 (UTC)[reply]
No, this was a measured warning issued in private to a member of the CU team after the problem was discussed with them. The warning was necessary because after outlining several concerns they doubled down and dismissed them which left the committee with few other options. Unfortunately, that individual chose to go public with it. The committee was accused of gross misconduct and attacked well before the committee had the opportunity to properly explain it to the rest of the CU team. I was not part of this decision or subsequent discussion but I watched it go down and I am disappointed and turned off by how it went down. The assertion that the committee poorly communicated the issue and caused unnecessary drama seems very unforgiving to the fact that no public statement was planned in advance; members of the committee were accused of terrible gross negligence before it could be comprehensively and adequately explained. Many inflammatory accusations were made early on and later backtracked after the initial "criticism" turned out to mostly be misinterpretation. Even within this ACN discussion, a considerable number of accusations have been backtracked. Constructive criticism and trust must be built upon mutual respect and I did not see that here. I found the whole thing incredibly disrespectful. Mkdwtalk01:54, 8 April 2020 (UTC)[reply]
CU's are not assigned by the community, they are assigned by a small committee. There is no actual basis for the community-at-large to trust CU's, what the community has done and does is place its trust in a committee to assign and review them. Then too, the community has insulated admins by assigning their review to this same small committee (a committee of other admins and CU's, no less). It's hard to imagine a system more protective of admins and functionaries. Even apart from Softlavender's battlegroundy 'note taking' which then makes even more absurdist their 'witch-hunt' stuff, we can be certain that the community trust is in the committee, not Softlavender and not me. As for the overwrought 'harm to the project', you know who else makes that slippery (shibboleth) argument: the socks, who are just here to improve the pedia, we hear. The community neither accepts that from the socks, nor from special pleaders for favorite admins or CU's, who just haven't managed to be accountable, or even receive communications or listen about issues. The community has decided, it places the responsibility for bad checks and poor conduct by admins squarely in the committee. Think about that, the committee is responsible for each bad check. -- Alanscottwalker (talk) 13:49, 7 April 2020 (UTC)[reply]
Looks like this is about violations of users privacy when their edits are not, have not necessarily caused detrimental effects on the project, we should all be happy that Arbcom is focussed on this type of personal intrusion into what in many many cases are just good faith users. Sock hunters might think they are doing the right thing but actually it is content and contributors that are king and queen here. Govindaharihari (talk) 13:59, 7 April 2020 (UTC)[reply]
Now that you and MrErnie has argued for the viewpoint in opposition to my own, I now know that I am correct in my thinking, as it's been my observation that the two of you are consistently wrong in almost every major Wikipedia debate. 11:09, 8 April 2020 (UTC)
Look BMK, we get that you are upset but you really should dial back the personal attacks. Ad hominems aren't going to accomplish anything constructive. LEPRICAVARK (talk) 12:15, 8 April 2020 (UTC)[reply]
What seems obvious to you may not be perceived that way by others. I'm not interested in agreeing or disagreeing with your assessment. This thread is for the discussion of issues relating to Bbb23 and ArbCom. Whether you think certain participants have been right or wrong on other issues is immaterial. LEPRICAVARK (talk) 14:01, 8 April 2020 (UTC)[reply]
Despite your repeated efforts to personally discredit individual participants in this thread, this discussion is still very much about Bbb23 and ArbCom and not about whether you have agreed or disagreed with specific people in the past. LEPRICAVARK (talk) 14:29, 8 April 2020 (UTC)[reply]
Beyond My Ken: could you clarify your position? Is it that Bbb23 should not be subject to the global and local checkuser policies and should not be subject to auditing by the committee simply because they do (did) the lion’s share of CU work? –xenotalk12:04, 8 April 2020 (UTC)[reply]
No, I have never said that, and that is not my position. ArbCom has the task of monitoring and auditing the behavior of functionaries, but they need to do so in a reasonable and responsible way that does not harm the community. My position is that ArbCom took an action which resulted in the retirement from CU activity -- hopefully short -- of one of the best CheckUsers we have, who has worked tirelessly to keep our community safer. I take it for granted that ArbCom did not intend to cause Bbb23 to retire, but that is nonetheless the result. What was intended as a private matter has become a de facto public matter, and since it has, ArbCom is under an obligation to be transparent with the community and reveal how the members of the committee voted in regard to sending the memo, and for what reason. Further, if ArbCom was serious about protecting the community, they should be acting as vigorously as possible to heal the rift with Bbb23 and convince him to return -- perhaps they are. I have not asked for anything private to be revealed -- who the complaining editors were, for instance, however much that might shed some light on ArbCom's actions (cf. the farcical comments by Iaritmioawp just above the break), but simply that the Arbitrators own up to their action now that it has blown up in their faces. I cannot personally evaluate whether ArbCom's position regarding Bbb23's checks was justified or in line with policy as understood by the community, because that data is not public, but I do know that hiding behind "privacy" concerns at this point is unwarranted, and a failure on ArbCom's part to be responsive to the community which elected it to serve us. Beyond My Ken (talk) 13:22, 8 April 2020 (UTC)[reply]
To boil that down a little: you had a job to do, and you did it, but you apparently did it badly, judging by the results, and now you won't own up to your responsibility for the mess. Beyond My Ken (talk) 13:27, 8 April 2020 (UTC)[reply]
At any point was there a dialogue between abitrators and Bbb23, in which you traded your interpretation of policy with the interpretation of a valued and experienced functionary (much more experienced than any of you), or did you just lay down the ArbCom interpretation as a fiat about which there can be allowed no disagreement? Beyond My Ken (talk) 13:42, 8 April 2020 (UTC)[reply]
You are, of course, completely wrong about that. I trust Bbb23 implicitly, as I do a number of other admins and functionaries -- including some members of the Committee as individuals. In the past, I have trusted ArbCom, and supported them against what I thought was unwarranted criticism, but that trust is wavering now, as they refuse to do what is obviously the right thing. Beyond My Ken (talk) 13:46, 8 April 2020 (UTC)[reply]
SN: I'm not sure who you're replying to, but if it's me, my use of "you" above (i.e. "you had a job to do...") was plural, and therefore collective. It was not addressed to Xeno individually. Beyond My Ken (talk) 13:49, 8 April 2020 (UTC)[reply]
Yes, I was replying to you. My point, really, was that in a collective decision making process, the identities of those who came to the decision should be irrelevant. See Chatham House Rules: making committee members publicly responsible as individuals, rather than as a collective, would stymy the process for the future. To put it another way, you would be hard pressed to find a member willing to express an opinion on anything—however innoccuous—if they thought it would do their legs down the road.My points shoud, I need not remind all, be taken as being from someone who has probably earned the enmity of most of the committee (collectively and individually!) over the years: I am not in the business of cheerleading them. ——SN5412914:07, 8 April 2020 (UTC)[reply]
Really? Then why is the voting on all ArbCom public cases completely open and transparent? They don't vote in the back room and announce the results, we see how each Arbitrator voted. The same goes for motions -- all public. So you are incorrect in saying the ArbCom decisions are "collective". Beyond My Ken (talk) 14:14, 8 April 2020 (UTC)[reply]
Yes, really. The decision, once made, is collective, and private decisions stay private. Incidentally, if you could tun the noise down, it'd be appreciated. Bbb23 doesn't need you cheering them: he needs t come back and abide by policy. It's not difficult; most of us manage it every day. ——SN5412914:20, 8 April 2020 (UTC)[reply]
No evidence has been presented that Bbb23 broke policy, only ArbCom's word, which I'm rapidly beginning to trust less and less, given their unwillingness to admit to their errors. In the meantime, Bbb23 remains innocent until proven guilty. Other CUs have said that they wouldn't have made the checks in question, but they haven't said that the checks went against policy, only ArbCom has said that. Who on ArbCom? We don't know, because they won't say. Beyond My Ken (talk) 14:27, 8 April 2020 (UTC)[reply]
@Beyond My Ken: There was an advisory email sent last year to all checkusers, which Bbb23 received. Bbb23 had other (imo, better) options apart from a public objection and retirement: he could have responded to the committee with a rebuttal; he could have sought additional clarification; he could have asked us to bring the rest of the checkuser or functionary team into the conversation; and of course he could have heeded the request to be more circumspect, seeking advice from the committee or other checkusers if he felt the need to perform a check that would be considered invalid per the above list. He expressed his feeling that we were "unappreciative of [his] CU work": not so - the very first line of our email expressed gratitude. He also wrote that we advised if he performed a "check in a manner that [we] deem violates policy, [his] CU privileges will be removed without further warning." - this is the only part of the email that I feel was ill-considered, and I take responsibility for not noticing that before it was sent and have advised Bbb23 on the checkuser-l thread that I would ask my colleagues to look into softening that line should he choose to re-engage with the project and return to CU work (since it would certainly be not a comfortable position to know that he was potentially only one check, one discretionary disagreement, away from being removed). –xenotalk13:58, 8 April 2020 (UTC)[reply]
That line, threatening to cut off his CU privileges without warning, was a very nasty threat, and I don't blame Bbb23 for doing what he has done -- if I were in his position, I too would feel that my years of hard work protecting the community were being taken for granted and not appreciated (no matter how much boilerplate language there was in the note). This is exactly what I meant about you -- the Committee -- doing your job badly, and by doing so causing harm to the community. It's unconscionable that such language was used on that particular person: you have treated him like dirt. Such an action doesn't warrant "I'm going to try to fix it", it warrants a full and abject apology from the Committee to Bbb23, and for you all to get down on your knees and beg him to return. You obviously find fault with the language used, but somebody put it in there, and I want to know who. Beyond My Ken (talk) 14:09, 8 April 2020 (UTC)[reply]
"it warrants a full and abject apology from the Committee to Bbb23, and for you all to get down on your knees and beg him to return" - Haha, probably the funniest thing I've ever read on here! Remind me on who was calling who a sycophant?! Bbb23 was hoisted by his own petard. Deal with it. LugnutsFire Walk with Me14:15, 8 April 2020 (UTC)[reply]
Glad you found it amusing. Some people will laugh at any old thing. Me? I find hypocrisy absolutely a stitch, which is great because there's so much of it around these days, especially in RL. Beyond My Ken (talk) 14:19, 8 April 2020 (UTC)[reply]
The email was reviewed/voted on by the entire committee except those that were recused or inactive on the motion; as SN has noted, we bear a collective responsibility. –xenotalk14:58, 8 April 2020 (UTC)[reply]
Baloney. If you reviewed/voted on it, then those votes need to be released to the community now that the matter is obviously no longer private. You have an absolute obligation to do that, and that the Committee continues to hide behind the fig leafs of "privacy" and collective responsibility is a serious ethical failing on its part. My regard for ArbCom has fallen significantly. I know you have a hard job, but you must all be willing to take individual responsibility for your actions when you do it badly. Beyond My Ken (talk) 16:39, 8 April 2020 (UTC)[reply]
Really. We lose a valuable service from a respected community member, and as a result we're a little less safer, and they did nothing wrong? No, just, no. Beyond My Ken (talk) 19:13, 8 April 2020 (UTC)[reply]
@Beyond My Ken: There is no obligation to release the votes as you claim, nor is it true that "the matter is obviously no longer private". Bbb23 decided to publish that he was warned, but the issue still centers around non-public data, and our votes were based off of information that we can't and won't make public. Per WP:ARBPOL, private deliberations are a standard occurrence on the Arbitration Committee, and the votes from those deliberations are not published. The obligation is actually in the other direction: The Committee treats as private all communications sent to it, or sent by a Committee member in the performance of their duties. While I have no personal objection to my own vote being published, publishing the votes without the rationales and alternatives will make very little sense, and publishing the additional context presents privacy concerns given that the discussion centers around non-public information. If it is useful to you to know, I voted in support of the warning, so feel free to oppose me if and when I run for ArbCom again.
As for your claim that the ArbCom "treat[ed] Bbb23 like dirt", I disagree—while I think "without further warning" should have been omitted from our letter (since I don't think we would remove the CU tool without at least asking for justification of the checks in question) it is also true that we will not tolerate continued misuse of the tool by continuing to issue warning after warning. It is a part of ArbCom's responsibility to audit checkusers, to advise a checkuser that they need to be adhering more closely to policy when there have been issues, and, if necessary, to remove access to the tool. If adhering to my responsibility means you won't vote for me at the next ArbCom elections, so be it.
I have had concerns for some time around the enormous number of checks Bbb23 has been running in proportion to the rest of the team—for burnout reasons, bus factor reasons, and because it's such a high number of checks to justify. It does not seem healthy for a team to have one member performing thousands of checks a month when the standard checkuser performs maybe a couple dozen. My worries are reinforced by some of the comments we're seeing here, including your demand for the ArbCom to "get down on [our] knees and beg him to return". These comments seem to me to be saying that Bbb23 performs such an important service that he cannot be reminded of policy even if he's not adhering to it. While I certainly would be delighted for Bbb23 to decide not to retire, the warning was justified and our request to him remains in force should he desire to return to activity. If it turns out to be true that the entire English Wikipedia will collapse or be overrun by sockpuppets or what have you if Bbb23 does not return to activity, the problem is not that he has left, but that we let it get to a point where the project was so dependent on one individual. GorillaWarfare(talk)20:25, 8 April 2020 (UTC)[reply]
Is ArbCom's "Check User for Idiots" (see above, just after the Break) the only valid interpretation of En.wiki CheckUser policy? If so, how did this interpretation come about, and under what remit did ArbCom decide that this interpretation is singular and binding on all CheckUsers? Beyond My Ken (talk) 00:49, 9 April 2020 (UTC)[reply]
The bullet points we provided are just a handful of examples of circumstances that, alone, are not valid reasons to check a user. They are based on both the enwiki checkuser policy and our own knowledge of how the checkuser team interprets policy (after all, members of the ArbCom are also for the most part checkusers). It's not new policy, and it's not not novel interpretation. Given that it's just a few examples, no it is not the "only valid interpretation" of CU policy. But if you're asking if someone could reasonably interpret the enwiki CU policy to say that any of those bullet points are allowed as a singular reason to check a user, my answer is also no. As for your question about remit, the Arbitration Committee is the group that oversees the enwiki checkuser team, per policy at Wikipedia:Arbitration Committee/CheckUser and Oversight. It is also made clear in the checkuser policy: The use of the CheckUser tool on the English Wikipedia is monitored and controlled by the Arbitration Committee, and CheckUsers may have their permissions revoked by the Arbitration Committee for misuse or abuse of the CheckUser permission.GorillaWarfare(talk)01:16, 9 April 2020 (UTC)[reply]
Wikipedia:Arbitration/Policy#Scope and responsibilities specifies that among ArbCom's responsibilities is "To approve and remove access to (i) CheckUser and Oversight tools". Wikipedia:Arbitration Committee/CheckUser and Oversight says "Please note that CheckUser and Oversight permissions are subject to periodic review". WP:CheckUser says "The use of the CheckUser tool on the English Wikipedia is monitored and controlled by the Arbitration Committee, and CheckUsers may have their permissions revoked by the Arbitration Committee for misuse or abuse of the CheckUser permission."
I do not see in any of these policies something to the effect that "The Arbitration Committee is the sole valid authority for interpretation of how the CheckUser tool must be used, and this interpretation is binding on all CheckUsers. Individual CheckUsers have no lattitude for interpretation of the CheckUser policies and must adhere to ArbCom's interpretation at all times." Where do I find that? If it's not spelled out in policy, then if ArbCom wants to be the only valid interpreter of CheckUser policy, that should be specifically proposed by the Committee and approved by the community.
As far as I am aware, ArbCom is only authorized to interpret policy when there is a dispute within the community as to how the policy should be interpreted. Where is such a dispute?
@Beyond My Ken: Following your logic, the Arbitration Committee would never be able to remove access to the tool, because any decision that a checkuser had misused or abused checkuser would be "interpreting policy". But that is of course specifically authorized in policy. In this case we decided to issue a private warning rather than revoke access to the tool. Furthermore, as Risker and others have pointed out here, the examples we provided are not controversial among the checkuser team. Is there a specific item on that list you're taking issue with? GorillaWarfare(talk)02:17, 9 April 2020 (UTC)[reply]
Finding that abuse has occured does not require the institution of a single overarching interpretation of policy, which is what seems to have occured here. Beyond My Ken (talk) 02:45, 9 April 2020 (UTC)[reply]
We told Bbb23 that we believed he had been running some checks that were not in line with policy. We indicated which kinds of checks were problematic, which we shared here, and we shared more specifics with him and the CU team (which we obviously cannot share here). What you seem to be proposing is that we tell him that we think has misused the tool, but not actually tell him which kinds of checks were the problem. I'm not willing to provide feedback to someone that doesn't give them constructive information so they can change their behavior, and that is what we did here. We did not institute a "single overarching interpretation of policy", as I have already said—we provided a small number of examples of invalid reasons to run a check. If you have questions about any of them or disagree with them, I'm happy to discuss further. I see that you are dropping the matter and so I will do so as well, but I hope you can at least see where I am coming from here. GorillaWarfare(talk)02:52, 9 April 2020 (UTC)[reply]
Look, you folks are clearly not going to fulfill your ethical obligation to be transparent to the community, so I'm obviously tilting at windmills. I have no way to force you to do anything except by appealing to your consciences, and that appears not to be working, so I'll back off with these final words:
I sit here in my living room waiting to see if I'm going to get sick and possibly die, while all around me, my country and the world at large moves closer and closer to authoritarianism. What I don't need, and I don't think the rest of the community wants, is for the leadership of a project dear to my heart to start taking steps in that direction as well, but that seems to be what has happened when you promulgate what you seem to believe is the only valid interpretation of a policy.
You took certain steps that you thought were justified, and it blew up in your fact, and you're not willing to take the responsibility for that, you hide behind the collective. Your explanations amount to "We know what's best", which I might be more inclined to accept if you were simply more honest with the community about how it happened, and if you seemed to have any willingness to take steps to ameliorate the damage. Of that I see almost nothing, bar Xeno's taking personal responsibility for not seeing the threatening language in the warning. If there's anything we lerned from the fall of Communism, it's that you can't change human nature, so when you threaten a hard-working community servant like that, it is to be expected that they would react unfavorably. That you all didn't see that coming is pure incompetence on your part. That some of you intrepret Bbb23's large number of checks as being inherenty problematic is indicative of your disconnection with the community: the sockpuppetry problem is a very serious one, the most serious problem confronting the project, and needs to be controlled, but you have taken steps to loosen our safeguards against it. This is, again, incompetence.
To me, you appear to be living in a fool's paradise in which it's more important for the authority of ArbCom to be followed blindly then that the community be protected from sockpuppetry. That's very unfortunate. In my opinion, this Committee, even though it has in its members ship quite a few people I have come to appreciate and respect, is an abject failure. Personally, I would recommend that, since you seem determined not to take individual responsibility for the actions taken, you collectively resign, and allow the community to elect a new committee immediately -- and, yes, your failure and incompetence does go that deeply.
Sorry BMK, I have no intention of resigning for doing my part to uphold the community checkuser policy. That is an important part of the job we were elected to do. – bradv🍁02:52, 9 April 2020 (UTC)[reply]
Of course not, I didn't expect anyone to. I did (and do) have hopes that perhaps their ethical failings and awareness of their botched communication might finally get through to one or two of the better members of the Committee, even though nothing in any Arbitartor's comments here would lead me to expect it. Still, there are those who haven't commented, so... I am, obviously, extremely disappointed - I thought better of (some of) you folks. Beyond My Ken (talk) 03:35, 9 April 2020 (UTC)[reply]
I understand that stresses are high right now, and that the world is a scary place. In every country people are terrified of this pandemic, and the extreme measures we're all taking to distance ourselves from others are leaving us feeling bored, frustrated, and alone. This has had a bit of weird effect on Wikipedia. I have noticed editors turn what would be minor disagreements into major conflicts, complete with accusations of bad faith between people who would otherwise consider each other friends. I have noticed people who are accustomed to spending a bit of their free time on Wikipedia instead pour themselves into their editing as an escape from the world, even to the point of losing all perspective of what we're doing here. And in the midst of this new editors are showing up in droves — to help curate the best possible collection of information on the current pandemic, to delve into the science behind viruses and tell us what they know, or even just to help with our vision of documenting the sum of all human knowledge. Editors remain our biggest asset in pursuing the goals of this project, and we need to value them, especially those who are just joining us. We need to give people a chance to participate, even if they've had a chance before and screwed it up. If we're turning over every rock and fishing in every pond to try to uncover bad behaviour and drive people away, the project will suffer. I believe our sockpuppetry policy strikes an appropriate balance between giving people the tools to limit disruption, and assuming good faith on the part of new editors. Now, more than ever, it's important for us to look for the best in people, to be patient with each other, and to remember that we're all in this together. I'm sorry you feel that I have let you down; I cannot show you the evidence behind my vote, but I just hope that you would be willing to take my word for it that this warning was necessary, assuming that I have the best interests of the project at heart. I know you do too. – bradv🍁03:44, 9 April 2020 (UTC)[reply]
Beyond My Ken, Even if everything you said was accurate (and I disagree with the much of it), on Wikipedia consensus is king. So, you would need other people to ascribe to your view. I have already offered you my resignation. You simply need to have more than 10 editors agree with you and less than that number disagree - if you think that's possible, please, do start a numbered petition on my talk page. In reality, I strongly believe the community is behind us in this matter, and you are a lone voice who is simply upset that we warned someone you consider a friend. — Preceding unsigned comment added by Worm That Turned (talk • contribs) 08:15, 9 April 2020 (UTC)[reply]
I would politely suggest that latecomers to the conversation make sure they actually understand the situation, I realize this discussion has gotten very long, but it isn't helpful to come in late and without full possession of the facts, at least the ones that are publicly available. To that end:
Kind of surprised to see anyone still saying this was "out of the blue" when we have repeatedly explained that our email was merely the last part of a months-long conversation with Bbb23.
As to this being a witch hunt, I would note that this, like most things the committee does, was the result of community members coming to us with their concerns. We generally have more than enough on our agenda without going looking for more.
I would also note that everyone who still seems outraged at this has not seen the actual evidence. We may have done a poor job communicating initially why we took this action (since this caught us flat-footed as up until then it had been a private conversation) but at this point there is a consensus amongst those who are allowed to see such things that there was indeed a problem. Very often we see accusations of users of advanced permissions "circling the wagons" to protect one of their own. In this case the opposite has happened, eventually, and there is general agreement as to the inappropriate nature of some of Bbb23's checks.
If you think any of us actually like doing this sort of thing, think again. Most of the time there is sufficient discussion and feedback between functionaries that any outliers that aren't acting within policy basically are self-corrected without the need for the committee to step in. In the rare cases where that does not happen, it is our responsibility to take appropriate action to correct the issue, whether we like it or not.
I also concur. There is no similarity between this and Framgate, and just because ArbCom dropped the ball badly on the BHG case doesn't mean there is some conspiracy to get rid of high-profile admins - that's simply tinfoil-hattery. Black Kite (talk)12:14, 8 April 2020 (UTC)[reply]
Beeblebrox, Yes, the issue appears to have been around messaging, and Bbb23 having been the first to bring it up, necessarily with his own take on it. The only takeaway would be to have a suitably depersonalised explanation ready whenever a formal warning is issued to a functionary. But that's not a very common thing, right? Guy (help!) 17:04, 8 April 2020 (UTC)[reply]
It looks like Wikipedia privacy policy would make that exceedingly difficult (impossible?): "The Committee treats as private all communications sent to it, or sent by a Committee member in the performance of their duties." Trying to draft a public communication individually and as a committee that can't reveal the substance, context, or identities would be a nightmare. -- Alanscottwalker (talk) 19:33, 8 April 2020 (UTC)[reply]
Alanscottwalker, true, but this is about having a response ready as and when someone else pulls the curtain aside. "The committee reviewed a number of uses of checkuser and identified a shortfall in compliance with policy; corrective action was taken" or some such. Guy (help!) 09:15, 9 April 2020 (UTC)[reply]
The committee deals with a large number of private matters. I don't think it needs to be ready with a public statement for all of them. However, this was an admittedly difficult situation with a widely respected and even more widely known editor. Being ready with a statement should that editor choose to make things public does seem sensible and would be something I hope this arbcom considers doing differently in the future. So I basically agree with Guy. However, in the grand scheme of things being caught a little flatfooted on messaging (and I need to emphasize a little because overall I think their response has been timely, thoughtful, and complete) when being right on the merits does not seem to be commensurate with a weeks worth of (sometimes quite hurtful to arbitrators who are editors too) comments. Best, Barkeep49 (talk) 12:42, 9 April 2020 (UTC)[reply]
My attempt to understand this:
User:Beyond My Ken mentions a "war against sockpuppetry"[3] to which Bbb23 was a good contributor.
CU is a sock-hunting tool whose usage is constrained by local (en.wp) wiki policy. de.wp's CU policy is even stricter than ours, so maybe they are more vulnerable to sock abuse than we are.
Some editors like Bbb23 and apparently BMK place very high importance on sock hunting, thus BMK's mention of a "war". So (ISTM) they want more CU'ing, and if that runs against the current CU policy, then the policy is getting in the way.
Bbb23, apparently, some of the time, resolved the conflict by ignoring the policy (WP:IAR) and running more checks than the CU policy authorizes.
BMK says he trusts Bbb23 about the checks. At the level of trusting Bbb23 to not misuse private info found in the checks, that is fine. I have no reason to think Bbb23 abused the info, so in that sense I also trust Bbb23.
OTOH, arbcom is pretty good at seeing what someone did and comparing those actions to site policy, so I trust arbcom when they say Bbb23 ran afoul of policy, even if in a good way.
So it sounds to me that the underlying issue from Bbb23 and BMK's perspective is that the policy itself is too strict. Their solution (my interpretation!) is that arbcom should have let the IAR keep going, instead of intervening as it did. Another approach would be to revise the CU policy to allow more checking, bringing it into sync with Bbb23's judgment of best practice in this area, which is grounded in his experience and deserves to be listened to.
Calling the policy too strict is of course in tension with editors who want the policy to stay where it is, or become even stricter on privacy grounds, or who see sock-obsessed admins as battleground editors peddling a moral panic du jour as a vehicle to act out their aggressive impulses. Socking is a malum prohibitum on Wikipedia after all: some other sites allow using multiple accounts freely, or don't even have user accounts.[4] One can put the problems socking causes into perspective with Wikipedia's infinite numbers of other problems, and conclude that some editors do tend to go overboard about the subject.
What we're left with is a philosophical divide about how to balance editor privacy interests against sock protection. BMK and Bbb23 apparently want more sock protection, Swarm(?) and others seem to want more privacy. It is a clash of values and that's why there's such intensity in the opinions. Revising the CU policy would take opening a discussion, and I don't know what the result would be. It seems to me that Arbcom has done a decent neutral job of observing existing policy, independently of the politically subjective question of whether the current policy is the best one possible. 2601:648:8202:96B0:E0CB:579B:1F5:84ED (talk) 06:38, 14 April 2020 (UTC)[reply]
Thank you very much to everyone who has taken the time to say something nice. Serving our project as a committee member was mostly a pleasure and rarely a chore. To everyone who has thought about volunteering for the role, I encourage you to self-nominate at the next election. AGK ■13:05, 14 April 2020 (UTC)[reply]
I commend the Arbcom for their timely deliberations and careful consideration; it was nice to see a proposal put up on time and closed without unnecessary delay, in spite of the last-minute concerns about a potential "quiet" return. I trust that this arb committee will have their eyes on the ball, with the best interest of Wikipedia, content, and editors in mind. SandyGeorgia (Talk) 18:03, 13 April 2020 (UTC)[reply]
Kudos to ArbCom for its deft handling of this case. I hope it signals that bullying behavior will no longer be tolerated on Wikipedia. Xxanthippe (talk) 01:04, 14 April 2020 (UTC).[reply]
I agree with both SandyGeorgia and Xxanthippe. In combination with the other recent decisions this is a good signal that the English Wikipedia is beginning to get its house in order regarding bullying and gross incivility. Thryduulf (talk) 08:26, 14 April 2020 (UTC)[reply]
@Thryduulf: I think that discussion above can probably be closed now; it's had no substantive commentary for three days now, and has begun attracting anonymous trolling. ——SN5412910:21, 14 April 2020 (UTC)[reply]
It was not intended as trolling. I spent a long time reading that thread and the comment was my attempt at making sense of what happened and what the disagreement was. Whether it contributes anything of value is not for me to decide, of course. But I tried. 67.164.113.165 (talk) 18:55, 14 April 2020 (UTC)[reply]
Just curious...what does it hurt? It seems to me that the biggest risk is to the IPs who have something to hide. Are we at the point we can't trust our most trusted users in the effort to protect anonymity? Perhaps incidents such as this one will help make a case for registration where only a select few highly trusted CUs are allowed access to the info - maybe WMF could consider obtaining a surety bond to cover them as a precautionary measure. I have far more trust in our CUs than I do in, say, members of WP:COIN. Atsme Talk 📧 12:05, 3 April 2020 (UTC)
