Wikipedia talk:Arbitration Committee/Noticeboard

Motion: India-Pakistan

Original announcement

Level 1 desysop of Nv8200pa

Original announcement

For discussion reference I've put together the initial timeline as follows feel free to make corrections (— xaosflux ^Talk 00:15, 23 May 2019 (UTC)) :

• 20190522T0208 Suspect Activity begins
• 20190522T1900 VPT Report
• 20190522T1959 Community Response/Escalate to stewards/arbcom (BLOCKED ACCOUNT)
• 20190522T2006 Steward action (LOCKED ACCOUNT)
• 20190522T2216 WMF Action (UNLOCKED ACCOUNT)
• 20190522T2305 ArbCom Public Response
• 20190522T2313 Crat action (DESYSOP)
• 20190523T0110 Community response (UNBLOCK)
• 20190523T1101 Password reset completion claimed
• 20190523T1236 Community response (CU Confirmation)
• 0000000000000 Possible ArbCom future public request
• 0000000000000 Possible future crat action

• I think it is important to note that the WMF activity was done prior to desysoping and any enwiki internal investigation concluding, reinforcing why applying blocks and access adjustments still make sense even when a steward is quickly responsive. — xaosflux ^Talk 00:15, 23 May 2019 (UTC)
  • Why did the WMF office unlock the account? ST47 (talk) 00:49, 23 May 2019 (UTC)
    • I'm wondering the same thing; with the information currently public, the unlocking seems quite bizarre. Vanamonde (Talk) 00:52, 23 May 2019 (UTC)
  • @JSutherland (WMF):. * Pppery * _survives 00:55, 23 May 2019 (UTC)
    • The only unlock reason listed is "Situation resolved", perhaps WMF can elaborate. — xaosflux ^Talk 00:58, 23 May 2019 (UTC)
      • Joe reset the password earlier and then unlocked. He pinged me on IRC while I was on my way out the door to get dinner to let me know he had done so. I told him I thought it’d be better to let ArbCom know for our local stuff and that I was heading to dinner (I was logged out at the time.) I’m not sure whether ArbCom was made aware, but I’m assuming he did. I think the block should be lifted, which I thought an arb would do, but since I’m back I can go ahead and unblock. TonyBallioni (talk) 01:08, 23 May 2019 (UTC)
        • @TonyBallioni: I endorse the unblock assuming that you expect that the correct person is now in control of the account, however it makes sense that the unblock would not occur until after the desysop. — xaosflux ^Talk 01:47, 23 May 2019 (UTC)
      • I emailed ArbCom to explain what I did with the account, that is correct. Tony's explanation is pretty much on point here. Joe Sutherland (WMF) (talk) 01:49, 23 May 2019 (UTC)

        JSutherland (WMF), I read above "Joe reset the password earlier and then unblocked". Did you get in touch with the original account owner, or just reset and unlock (that sounds a little snarky - but it might make sense to restore the original email address, reset the password, and unlock. I've seen worse in a lot of companies I've worked with)? I know in the past, we've been provided with details as to the likely source of the compromise (last time, we were told the password was successfully used on the nth try) - do you have any insight in this case? SQL^{Query me!} 02:23, 23 May 2019 (UTC)

        @SQL: To answer your questions out of order a little - as I told ArbCom via email, I am not sure how much I can divulge about the technicalities of this case for purely privacy-related reasons. I know the lack of transparency that causes is frustrating but I would obviously like to be sure I'm not breaching any internal protocol by revealing that sort of detail. In this case, the compromise was brought to our attention by the Stewards after it happened, but before the true owner of the Nv8200pa account reported it to us (honestly, likely before they even realised anything was wrong). In such instances we rely more on technical data than working directly with the target of the attack. There are no logs of what I did to make sure the attacker no longer had control (since it was writing directly to the database) but it really does boil down to "reset email, scramble password". Hopefully that makes sense. Joe Sutherland (WMF) (talk) 21:43, 23 May 2019 (UTC)

        Thanks for the update @JSutherland (WMF):. I don't think there is much else to do on this specific incident, but if you are open to feedback I'd suggest adding a step (that may exist but not be exposed here) of: validating that the recovery agent appears consistent with a last known good agent, else - relocking the global account. The reason why, attackers could compromise email accounts as well as editor accounts - especially with poor credential usage such as password reuse. We may include this as our own project recommendation for unblocking in such future cases after we have a further change to review this occurrence. Thanks again, — xaosflux ^Talk 22:06, 23 May 2019 (UTC)

        @Xaosflux: (sorry, missed this.) I'm not sure I follow. What exactly do you mean by "agent" in this context? Joe Sutherland (WMF) (talk) 22:11, 23 May 2019 (UTC)

        JSutherland (WMF), Thank you very much for answering!

        I think what Xaosflux is getting at (and what I was thinking as well) - is the concern that if the WP account is already compromised - it might be a decent assumption to make that the email account is likely compromised too. I know at least one recent compromise was likely due to password reuse - and it's extraordinarily common for users to use literally the same password everywhere. Therefore, if the WP account is compromised - there is a very good chance that unlocking after a simple reset of pw and email may simply hand the the account back to the attacker. I am sure that technical measure is taken to make sure that the attacker cannot re-use the proxy IP that was originally used - but it's not exactly hard to find another working proxy.

        There's another technical concern with merely resetting the email and password - but I don't think it's appropriate to go into here. SQL^{Query me!} 23:24, 23 May 2019 (UTC)

        @JSutherland (WMF): Without communicating with the user, how did you determine that they were in control of their email, when you unlocked? For all anyone knew at the time (?), the same password could have been used on both the WP account and whatever email address you reset to, making it trivial to compromise the WP account again after the password reset. Suffusion of Yellow (talk) 21:56, 23 May 2019 (UTC)

        @Suffusion of Yellow: A valid concern, but since this did not appear to have anything to do with the email address involved I believe such a scenario is unlikely. (See also the CheckUser confirmation below that proves the account is back in its owner's hands.) Joe Sutherland (WMF) (talk) 22:07, 23 May 2019 (UTC)

        @JSutherland (WMF): by agent I mostly meant "person" I think the CU action done by our local CU was great here - but with accounts and their related locks being global, my suggestion is that this type of check should be routine - and failure of the check should result in re-locking. As the password recovery, and account management is beyond the scope of any single project I think this should be included in whatever workflow T&S is doing (if you are using a ticketing system for example, don't close the support ticket out until this confirmation is complete). This is certainly a tricky situation when the only "authentication" being performed is "can you access your email account". — xaosflux ^Talk 22:57, 23 May 2019 (UTC)

• We have not yet verified the correct individual is back in control of the account and the account has not since been used. Nv8200pa has been asked to contact the Arbitration Committee as well as the WMF. As far as we know, Nv8200pa has not contacted either us yet. Until that occurs and we have some type verification the account is accessible by the original account holder, we felt it was necessary to remove the administrative flags from the account out of an abundance of caution as the original compromise involved use of the administrative tool set. Mkdw ^talk 03:03, 23 May 2019 (UTC)

  @Mkdw: If both a WMF T&S rep AND a local CU aren't enough to determine that the right person is in control - there is a big problem (and the account shouldn't be unlocked or unblocked). That being said, having the right person in control is not the same as meeting the satisfactory explanation ... provided or the issues are satisfactorily resolved standard access restoration provision. — xaosflux ^Talk 03:10, 23 May 2019 (UTC)

  Just for the record, I didn’t CU this. I relied on Joe’s determination that an unlock was acceptable because of the email and password reset, and since it was desysoped locally, I thought that was good to undo my block. If ArbCom wishes to reimpose it, I have no objections, and am sorry if I stepped on toes. I just thought the WMF unlock and desysop rendered the block not necessary. TonyBallioni (talk) 03:18, 23 May 2019 (UTC)

@Xaosflux: I believe WMF T&S and a local CU would more than capable of determining whether the right person was in control of an account. The only missing element is the actual account holder in order to do so... Any technical information used to determine this would require the individual to log back into the account and leave a digital log. Again, as far as I am aware the individual has not redeemed their reset password, logged into the account, made an edit, nor contacted either the WMF or ArbCom to complete the process. Likewise, in order to receive any type of confirmation or satisfactory explanation, we would like to speak with the original account holder. Mkdw ^talk 03:20, 23 May 2019 (UTC)

Would be useful if WMF could reply if the password recovery was completed as asked by SQL above (which itself would not create an on-wiki log). — xaosflux ^Talk 03:25, 23 May 2019 (UTC)

Xaosflux, see my response to SQL above, hopefully that answers your question. Joe Sutherland (WMF) (talk) 21:43, 23 May 2019 (UTC)

• Xaosflux, The right person being in control isn't enough IMO. The right person in control - whom will, going forth, resolve to maintain better password hygene (appropriate password, used only on WP - same goes for any linked recovery email, and ideally 2FA being used) would be the bar I hope that this passes. SQL^{Query me!} 03:28, 23 May 2019 (UTC)

  @SQL: I wholly agree with that, which is why I'm in support of ensuring that we had a local block and access removal - since WMF will unlock the account based solely on identification, not on an evaluation of due care. As fast as unblocking a de-privileged account, I'd say the bar is the same as the unlock - though there are question if that was most securely performed as well now. — xaosflux ^Talk 03:43, 23 May 2019 (UTC)

  The WMF is typically in charge of determining if unlocks occur. Stewards do not handle this anymore globally. I believe I’ve unblocked non-sysop accounts for Joe before on en.wiki based on password/email resets, and my view is that if they believe there is enough evidence to unlock globally, that determination by a WMF staffer should be enough to unblock locally. I had been talking to Joe about some of the CU data before, knew that he had been looking at technical details beyond what CU can show, and had experience working on this sort of thing with him before, so once the desysop happened, it looked sufficient to me to lift my block as the blocking admin. I can’t answer all of those questions since there are some things I don’t know (as I’m not a T&S member), but Joe has been handling these globally for a while and I trust his judgement on unlocks. This is of course, a different question than the desysop question, which is up to ArbCom. TonyBallioni (talk) 03:54, 23 May 2019 (UTC)

  These admin account compromises are occurring with some regular frequency so it might be worthwhile to outline a standard set of unblock procedures. In this case, it seems like all teams were working on the issue at the same time so we had some crossover. Personally speaking, I do not think compromised admin accounts should be unblocked until the account has been a) desysopped and b) unlocked by the WMF (if globally locked). This would constitute the bare minimum in my opinion. One situation where we need to be cautious about is a person who uses the same password for their email and their Wikipedia account (or a Wikipedia account that is accessed because of a compromised email account). Resetting the password with the recovery link being sent to their email could simply allow the account to be compromised again. This is why I think requiring some sort of additional verification to ensure the right person actually has control of the account would generally be a good idea. There are not too many benefits in rushing this process when compared to the risks. Mkdw ^talk 04:26, 23 May 2019 (UTC)

  Yes, I think we’re generally on the same wavelength. I agree that those should be the minimum for an unblock, my general sense is that once an unlock has happened and an desysop has occurred there’s reasonable assurance that the account is secure and the desysop provides an additional failsafe so that there’s not damage if the WMF was wrong. Resysoping is an entirely different matter and I’d agree 100% on needing more assurance before that happens, if at all. TonyBallioni (talk) 04:40, 23 May 2019 (UTC)

  Any volunteers to write up unblock instructions for a compromised admin account? Mkdw ^talk 04:49, 23 May 2019 (UTC)

Just want to point out that, once the account is desysopped, it's no longer an admin account, just a regular user account; there should be a single standard for unblocking of any account considered to be potentially compromised. Whether or not the account should be resysopped is an Arbcom decision, and it should probably be an arbitrator who writes up that process (whether publicly, or just on the arbwiki). Risker (talk) 05:01, 23 May 2019 (UTC)

Whether an account should be resysopped is currently a decision that rests with bureaucrats, the committee should be the one to lift their desysop motion and it would be helpful to make recommendations to the bureaucrat team following the investigation and nature of the compromise as to whether it is safe to restore privileges based on whether appropriate operational security was (or will be) observed. The committee does not currently have a mandate to create or recreate administrators. This was one of the points of contention in the last compromise drama. –xeno^talk 09:43, 23 May 2019 (UTC)

ArbCom absolutely does have the mandate to investigate claims of administrator misconduct, and it is administrator misconduct to fail to meet the minimum requirements for administrator account security. ~ Rob₁₃^Talk 13:36, 23 May 2019 (UTC)

Of course they do. –xeno^talk 13:41, 23 May 2019 (UTC)

I think it would be very dubious if any process wasn't public Nosebagbear (talk) 08:25, 23 May 2019 (UTC)

I think it would be dubious if any process was public. How do you propose we openly discuss the level of security of an individual Wikipedia account in public? That's asking for a second compromise. ~ Rob₁₃^Talk 13:36, 23 May 2019 (UTC)

• I think we are setting the wrong sentiment here. Unblock instructions are secondary to determining the two most important things: 1) if control of account has been regained at all 2) why the account was compromised in the first place. It makes no sense to draw up completely arbitrary standards when we do not identify the problems behind it. I think the focus should should not be "how to get rights back" but "why", and personally, it makes no sense that there isn't a proper mechanism to enforce that and this latest timeline is probably another example of that. In fact, the only reason this particular compromised/newly-uncompromised account (which mind you, we are still assuming it is with absolutely no on-wiki proof to show for it) didn't cause disruption is the overabundance of caution at VPT and the rapid handling by active administrators. The fact that most of under-the-radar action here was caught accidentally is a testament to the fact that the system is broken and the rapid reversal of mitigations in place is not helpful, since again, there is no clear determination of the account holder, so is there a point rushing the wrong processes? --qedk (t 桜 c) 08:57, 23 May 2019 (UTC)

• It's not yet been confirmed if Larry (the named user of the Nv8200pa account) is in control of their email account, but restoring email, password, and unlocking, does allow the person in control of either the email or the Nv8200pa account to make contact. We hope that Larry is in control of their email and will get in touch. If the attacker is in control of the email account, then I am assuming we will not get a response by email, but with password restored Larry should be able to respond via their Wikipedia account. In the meantime the admin tools have been removed. So, the door is open for a return, but the valuables have been put in a safe place. SilkTork (talk) 10:15, 23 May 2019 (UTC)

  If the attacker is in control of the email account, won't they be the one who gets the password reset details and then be back in control of the account? Boing! said Zebedee (talk) 10:28, 23 May 2019 (UTC)

Yes, that is possible, though it is considered an unlikely scenario. Meanwhile, if communication is restored, we have a better chance of returning the rightful person to the rightful account. SilkTork (talk) 11:30, 23 May 2019 (UTC)

Also note that locked account never receives an email (I tested it for pings, but I believe password resets are also ignored), so, to receive an email from Wikipedia, unlock is necessary. — regards, Revi 23:11, 23 May 2019 (UTC)

Actually, it sends password reset so T224341 filed. — regards, Revi 15:32, 28 May 2019 (UTC)

• This is Larry. I received my password reset and have successfully logged in with the temporary password and created a new permanent password. My Wikipedia Watchlist is gone. Can that be restored? Thanks. Nv8200p talk 11:01, 23 May 2019 (UTC)

Hi Larry. To start a discussion regarding your admin tools, you'll need to contact ArbCom via email at arbcom-en@wikimedia.org. SilkTork (talk) 11:30, 23 May 2019 (UTC)

Checkuser needed can a review that Nv8200p's current fingerprint is consistent with the one pre-20190522 be checked? — xaosflux ^Talk 12:21, 23 May 2019 (UTC)

 Confirmed entirely consistent with the user being back in control. Also note the relevant account name and the user making the edit above, Nv8200pa. -- zzuuzz ^(talk) 12:36, 23 May 2019 (UTC)

Thank you, timeline updated above, this should certainly resolve the "right person in control" matter - leaving the due-care/account security matter open. A post-mortem review is a good idea once this gets resolved one way or the other, so we can document the best-practices for some of these orders of activities, and make recommendations for improvements. (There are some quite valid order-of-operations and expectations of parties issues that have been raised). — xaosflux ^Talk 13:14, 23 May 2019 (UTC)

I put together a draft and circulated it to the committee for comments. It got a mixed response with some saying it was not needed. So, no proposal will be coming from the committee. I had originally suggested for someone else to volunteer because while desysop decisions and enacting them falls within the responsibilities of the Arbitration Committee, the committee follows the community's policy, such as Wikipedia:Blocking policy#Unblocking. So here's my copy of the draft. Others are welcome to improve and change it or decide it is not needed.

Note: To be placed under Wikipedia:Blocking policy#Unblocking

Administrator accounts that are suspected to have been compromised and have been blocked may only be unblocked by: The Arbitration Committee; or By an administrator or checkuser/oversighter (WP:CUBL/WP:OSBL), after the account has been desysopped and unlocked by the WMF or Stewards Please see WP:RETURN for the procedure to return administrative permissions.

Mkdw ^talk 17:24, 31 May 2019 (UTC)

• @Mkdw: even if not a strict policy, codifying the "compromised account" incident handling procedure is a good idea. It should contain when TO block, when TO unblock. The unblock should include some sort of confidence that the correct account holder is actually in control of their account (i.e. that the password reset if granted by WMF/devs has completed, and possibly some other verification such as a committed identity/public key/checkuser check) - thoughts? — xaosflux ^Talk 17:29, 31 May 2019 (UTC)
  • This was basically what I was going to propose when I got around to it (sorry for not replying Mkdw). Xaosflux, I think a KISS method makes sense. As Risker has pointed out, once a desysop occurs they are no longer admins so the same standards apply to them as everyone else. The current standard for that is the WMF (normally Joe) telling us they’ve reset the password and unlocked. There is no requirement that anyone be identified to the WMF anymore (outside board seats/committees) and it’d basically be impossible to obtain the level of assurance that you’d expect at a corporation for this reason without making requirements that would likely be viewed by many sysops as privacy violations. TonyBallioni (talk) 17:38, 31 May 2019 (UTC)

Stricter re-sysopping can be counterproductive

There has been an enormous amount of discussion on whether the Arbs (or indeed, anyone) has the right to prohibit resysopping on the basis of insufficient security standards taken. In effect, this is a judgement of security efficacy vs community judgement

I want to raise a separate point - that this form of assessment can be counterproductive at improving our security and reducing Administrator breaches

By requiring a consideration of whether the Admin was (too) careless, we discourage a full and frank discussion between them and whoever as to what happened. This makes it harder to consider what should be done in the future because we have less information. In fact, it might encourage future deception in order to ensure regaining the tools, which would be downright counterproductive.

This is a known consideration of pilot accident assessment - you actually get better future safety if a non-punitive approach is used. Nosebagbear (talk) 13:47, 23 May 2019 (UTC)

Well, people might lie, hardly seems a reason to do anything. The people who are liars, should not generally be trusted. Now, sure some liars will pull one over but the injury is that they are liars. As for the 'security', as the investigation will need to be private, at any rate, and the issues it explores will include, 'despite the breach, can one still reasonably demonstrate the level of care and trustworthiness in all matters that are expected', and the investigators will need to have their doubts in the matter overcome with that individual person, one investigation will hardly translate to another investigation. Alanscottwalker (talk) 14:07, 23 May 2019 (UTC)

Moreover, administrators have a strong incentive not to lie because the Arbitration Committee receives information from the WMF directly about how a breach occurred. Where the information received from the administrator and the information received from the WMF do not match up, the best-case scenario as a determination would have to be that we do not know for sure the cause of the breach, and so we cannot know if the account has been secured. That would presumably be a barrier to resysopping. ~ Rob₁₃^Talk 14:11, 23 May 2019 (UTC)

That being said, I wonder if an innocent until proven guilty approach would be best here. - Yuhong (talk) 00:35, 8 June 2019 (UTC)

SQL appointed trainee clerk

Original announcement

• Congratulations to SQL! From one three-letter user to another, I know you'll do a great job ^{but that's pretty obvious}. :D –MJL ‐Talk‐^☖ 04:14, 4 June 2019 (UTC)
  • We're taking over. AGK ■ 20:53, 5 June 2019 (UTC)
    • Now I feel some pressure. Congrats! -- Luk ^talk 11:49, 6 June 2019 (UTC)
      • Can I join in? NYB 15:04, 6 June 2019 (UTC)
        • NYB if you switch your signature –MJL ‐Talk‐^☖ 19:19, 6 June 2019 (UTC)

Desysopping of Od Mishehu

Original announcement

• I would like to thank the committee for taking this step. I was one of the CheckUsers who reviewed the data and I think this was the only acceptable option. TonyBallioni (talk) 17:20, 5 June 2019 (UTC)
• Has anyone considered removing their EFM rights? That's a pretty sensitive right imo, and a lot of damage can be done but also it's a great way to see what is in place to prevent abuse from their ip socking as well...Praxidicae (talk) 17:28, 5 June 2019 (UTC)

• I think removal of local permissions like EFM can be done as a regular admin action, ArbCom doesn't usually address those in desysop motions/remedies. Ben · Salvidrim! ✉ 17:30, 5 June 2019 (UTC)

• I considered it earlier, but was otherwise occupied; with there being CU-verified evidence of abuse, I've gone and done it. Especially since it was self-granted. ~ Amory (u • t • c) 17:41, 5 June 2019 (UTC)

Thank you, Amorymeltzer, for doing that. ~Oshwah~^{(talk) (contribs)} 10:36, 6 June 2019 (UTC)

• Has the disruptive editing egregiously affected content and not been amended? That is, is there any clean up required? - Sitush (talk) 17:31, 5 June 2019 (UTC)

• No, no cleanup is required. It was handled as it happened. Katie^talk 17:57, 5 June 2019 (UTC)

• Great, thank you. - Sitush (talk) 18:02, 5 June 2019 (UTC)

• Very disappointing. However I agree with Praxidacae, a user who violates trust at this magnitude should not hold any userright at this moment not talk of as sensitive as EFM one which requires high level of trust which doesn't exist here. – Ammarpad (talk) 17:36, 5 June 2019 (UTC)

I agree. ~Oshwah~^{(talk) (contribs)} 10:36, 6 June 2019 (UTC)

• Can the sole dissenting arbitrator clarify his views, kindly? ∯WBG^converse 17:37, 5 June 2019 (UTC)

  Even if there is reluctance to go into details, I also think it would be worth to clarify whether the oppose was (1) because the evidence did not convince Courcelles that Od Mishehu was responsible or (2) the evidence was clear but the actions weren't enough to desysop. Ben · Salvidrim! ✉ 17:44, 5 June 2019 (UTC)

  Absolutely. Wot you se. ∯WBG^converse 17:49, 5 June 2019 (UTC)

  That may not be possible. I cannot speak on behalf of Courcelles, yet they are presently inactive on all committee business. AGK ■ 20:52, 5 June 2019 (UTC)

  AGK was Courcelles being marked as "oppose" instead of "inactive" in the motion voting a clerical mistake? Ben · Salvidrim! ✉ 20:55, 5 June 2019 (UTC)

  Not at all. Courcelles cast a vote to oppose this motion. Courcelles is still inactive on all new arbitration business and may not visit this page in good time. I am just letting you know that you have asked questions of an arbitrator who is not currently active. AGK ■ 21:06, 5 June 2019 (UTC)

  Courcelles cast a vote to oppose this motion... I am just letting you know that you have asked questions of an arbitrator who is not currently active. <facepalm>Dan Murphy (talk) 21:09, 5 June 2019 (UTC)

  Shall I ignore the question, next time? AGK ■ 21:18, 5 June 2019 (UTC)

  (edit conflict)AGK, personally, thank you for taking the time to answer as best as possible anyways under the circumstances. I, at least, appreciate your communicativeness. Ben · Salvidrim! ✉ 21:41, 5 June 2019 (UTC)

  Nah. Just try not to mangle the English language and reason in your statements.Dan Murphy (talk) 21:37, 5 June 2019 (UTC)

  Notwithstanding the inherent irony of your statement, your attempt to normalise the notion of an arbitrator who selectively partakes in certain cases but hides under the guise of inactivity, when asked for accountability by the community, is well noted. And, then, you are one of the folks who will decide about whether some admin followed ADMINACCT, in potential future cases. Sigh. ∯WBG^converse 21:14, 5 June 2019 (UTC)

  I made no attempt to normalise anything. You are inventing a narrative. AGK ■ 21:18, 5 June 2019 (UTC)

  I am unable to see otherwise. If you are not willing to give such impressions, you might try to not frame such weird self-contradictory sentences. ∯WBG^converse 21:32, 5 June 2019 (UTC)

  ... and who hasn't edited in several months. * Pppery * _{it has begun...} 21:20, 5 June 2019 (UTC)

  That makes no sense. An inactive arbitrator cannot dissent. ∯WBG^converse 21:00, 5 June 2019 (UTC)

  I think it would be very useful to find out why Courcelles opposed in the vote, but I think it's unfair to assume that there is some sort of bad-faith stuff going on because there wasn't a quick reply. No reason why someone could not have taken part in the decision and then stepped away. --Tryptofish (talk) 21:28, 5 June 2019 (UTC)

  "And then step away" - Courcelles hasn't edited in months, however, which is one of the reasons why WBG above is so struck by an Arb marked as inactive for Arb business and who is not an active editor would, seemingly out-of-the-blue, cast an oppose vote on what seems like such an obvious call (per the evidence laid out below) and which was otherwise an unanimous vote. Ben · Salvidrim! ✉ 21:41, 5 June 2019 (UTC)

  Understood. And despite the indenting, I didn't mean to direct it specifically at WBG. And it is indeed useful to understand "an oppose vote on what seems like such an obvious call". It just doesn't mean that it's inconceivable that an Arb would briefly take part in one decision in the middle of an otherwise inactive period, nor does it mean that there was bad faith. --Tryptofish (talk) 22:11, 5 June 2019 (UTC)

  Not speaking for Courcelles at all, but in my experience it's easy to end up participating in matters being handled primarily by email while not doing much on-wiki. I don't edit from work and I hate editing from my phone, but in both contexts email is much more doable. Opabinia regalis (talk) 08:20, 6 June 2019 (UTC)

Under arcane ArbCom procedures, an Arb can be inactive, but become active on any issue by commenting on it. The checking of the findings by the functionaries and the resulting discussions and clarifications and double checking and looking into possible alternative explanations took the Committee some time. The functionaries passed their findings to us around 13 May, and we have been discussing it since then, with several emails to Od Mishehu. Courcelles commented on this issue while it was being discussed on the email list, and made it clear he did not support a desysop. At the time he said that, there were still a few other Arbs, including myself, who were not sure. After further investigation of the technical and behavioural evidence, and more emails to Od Mishehu, all the Arbs came round, though Courcelles was still logged as not supporting. I cannot say for certainty if at this point Courcelles would still be opposed to desysopping, but that was his last indicator to us, and it came, as I said, at a stage when we did not have the clarity we do now. Should we, in retrospect, have marked Courcelles as inactive? Well, that would have gone against his last statement to us, so I'm not sure. SilkTork (talk) 23:15, 5 June 2019 (UTC)

• You all do know that the first time he was caught socking as an administrator was in 2008, right? The odds he hasn't been doing this ever since are pretty much nil.Dan Murphy (talk) 17:56, 5 June 2019 (UTC)

  Dan Murphy, any links to relevant threads? ∯WBG^converse 18:03, 5 June 2019 (UTC)

  Arbcom-I, April 2008, leaked on Wikipedia Review in 2011 and there's a thread about it on Wikipediocracy as well. I'm too lazy to figure out who was on the arbcom then and which of those people are still around, but you could ask them for more details.Dan Murphy (talk) 18:09, 5 June 2019 (UTC)

  The 2008 ArbCom appears to have been comprised of Newyorkbrad, FT2, FayssalF, Sam Blacketer, Deskana, James F., YellowMonkey, Stephen Bain, Charles Matthews, Morven, FloNight, Kirill Lokshin, Paul August, UninvitedCompany, and jpgordon. Out of those, Newyorkbrad, FT2, Paul August, UninvitedCompany, and Jpgordon have edited within the past week. -A lad insane (Channel 2) 18:56, 5 June 2019 (UTC)

  I have no recollection of this issue. As it happens, April 2008 was the month in which I was threatened, doxed, and defamed on external sites and took a long wikibreak to deal with that, so I was probably distracted from regular ArbCom business. Newyorkbrad (talk) 19:10, 5 June 2019 (UTC)

  You were active on Wikipedia every day of April 2008, with the exception of a break between April 16 and April 22. The thread on Arbcom-I "Admin Od Mishehu and serious BLP vandalism" began on April 14 of that year.Dan Murphy (talk) 20:59, 5 June 2019 (UTC)

  That was 11 years ago, and arbcom-l had a great deal of traffic. I don't remember anything about the incident either. UninvitedCompany 22:29, 5 June 2019 (UTC)

  Thanks, Dan. ∯WBG^converse 20:47, 5 June 2019 (UTC)

  I barely remember what I had for lunch today. I wore the same shirt to the same meeting with the same people two weeks in a row, and I was even specifically trying not to do that! Expecting people to remember this one incident from more than a decade ago is not realistic. Incidentally, the old discussion was known and available to the current arbs for review, and speaking for myself, was totally irrelevant to my thinking, because, again, it was eleventy years ago. Opabinia regalis (talk) 08:20, 6 June 2019 (UTC)

• Some background/an explanation would be nice. GiantSnowman 18:10, 5 June 2019 (UTC)
  • What more background/explanation do you need? There is checkuser-verified evidence that Od Misheu has been disruptively editing Wikipedia while logged out on multiple occasions. This is a very clear violation of several policies and the evidence convinced 9 of 10 arbitrators that a desysop was warranted. They could easily have been blocked under the sockpuppetry policy. Checkusers will never connect an IP address with a registered user (this is certainly in the CU policy, but whether it is also covered in the Foundation's privacy agreement they are required to sign I can't remember) so they cannot give more information in that regard. Thryduulf (talk) 19:10, 5 June 2019 (UTC)
    • I dunno, I was just expecting more information than 'this long established admin has been socking, we've desysopped them'. GiantSnowman 19:44, 5 June 2019 (UTC)
      • Since the editing was done while logged out, that actually implicates the privacy policy so far as giving more detail or specifying which one(s), since that would reveal the editor's IP in violation of the privacy policy. If the socking had been done with alternate accounts, they could be named specifically. Seraphimblade ^{Talk to me} 19:50, 5 June 2019 (UTC)
        • How about some information about the nature or extent of the disruption and how long it's been going on for? That wouldn't need IP addresses to be revealed. Boing! said Zebedee (talk) 19:53, 5 June 2019 (UTC)
          • Exactly - something as factual and basic as "CU evidence clearly showed that OM had used X IP accounts over a period of X years to make approximately X edits which could be considered [vandalism/BLP violations/whatever - maybe give some examples so we get a flavour of what's gone on]". GiantSnowman 19:56, 5 June 2019 (UTC)
            • Plus the answer to WBG/Salvadrim's question above would be useful in determining whether the finding of wrongdoing was unanimous or not, and would not require revealing IP information. – Levivich 20:02, 5 June 2019 (UTC)
              • The data checkusers see when looking at accounts is only kept for 3 months, so CU wont be able to confirm anything going back further than that. If the IP used when logged out is static or it's part of a small/lightly used dynamic range then behavioural evidence might be able to take it back further than that. If however it's a highly dynamic IP on a busy range then determining which edits where made by OM and which were not without CU data will likely be very difficult unless there are very clear behaviour tells. Thryduulf (talk) 20:11, 5 June 2019 (UTC)
                • Yep, I appreciate that, but we could surely be told how long it went on up to the extent of the CU data - a few days, the whole 3 months? And the nature of the disruption? Boing! said Zebedee (talk) 20:14, 5 June 2019 (UTC)
                  • I can say what I reviewed since this was brought to checkuser-l: in my opinion there were multiple violations of different types on multiple occasions within the CheckUser data window. This was not simply based on IP: there was a device match and other indicators. Additionally, I am of the belief that there is non-technical evidence connecting improper logged out edits outside the CU period to Od Mishehu. Arbs have a fuller picture as this was discussed by them on arbcom-en as well, but that is what I can say from what I know. TonyBallioni (talk) 20:27, 5 June 2019 (UTC)
                    • Thanks Tony, that's valuable. Boing! said Zebedee (talk) 20:29, 5 June 2019 (UTC)
                  • As TonyBallioni suggests, the occasions mentioned in our motion were spread across the 3-month data retention window. The edits were in every case simple disruption or page vandalism. No abusive content was added; the edits were to non-article pages. In isolation, the edits would justify "revert on sight and warn the IP". The edits would not justify an immediate block (except for repeated offences). The edits were indisputably disruptive. I am sorry that I cannot provide more detail or simply link you to the edits; this detail of this matter is necessarily private. AGK ■ 20:48, 5 June 2019 (UTC)
                    • Thanks AGK, that's the kind of thing I was wanting to know. If it was article disruption, BLP, abusive or anything like that, I'd be asking about a block or ban. But I guess he's effectively on a firm warning now. Boing! said Zebedee (talk) 20:59, 5 June 2019 (UTC)
                    • Going by your descriptions, this is going to be the weirdest case of 'Good Hand - Bad Hand' sockpuppetry, I have ever seen. ∯WBG^converse 21:07, 5 June 2019 (UTC)
                      • That is not what GHBH means; see WP:GHBH for a definition. The IP addresses that were the subject of our findings did not engage in direct overlap with Od Mishehu's account. AGK ■ 21:26, 5 June 2019 (UTC)
                        • Thanks, my fault to assume of the intersections (as it occurs, almost always) :-( ∯WBG^converse 21:32, 5 June 2019 (UTC)
                        • Though I agree with your definition, note WP:GHBH doesn't describe any overlap. isaacl (talk) 02:40, 6 June 2019 (UTC)
                          • It does: accounts would be editing the same page or discussion. AGK ■ 08:14, 6 June 2019 (UTC)
                            • Nope; Using one account for constructive contributions and the other one for disruptive editing or vandalism. You're quoting some other stuff, from up-above, which ain't linked any with GHBH.∯WBG^converse 08:28, 6 June 2019 (UTC)
                            • The specific shortcut you referenced points to a definition of "good hand" and "bad hand" accounts that does not say "same page or discussion". This phrase is contained two bullet points higher up, as separate example of inappropriate use of alternative accounts. I partially retract my agreement with your definition; although that is one usage, I have seen the terms used as well just to cover the concept of someone having a separate account where they act poorly in unrelated ways to their other account. isaacl (talk) 10:08, 6 June 2019 (UTC)
• Why isn’t the user blocked? –xeno^talk 20:54, 5 June 2019 (UTC)
  • I am not sure blocking would prevent any sort of further disruption or otherwise accomplish anything. The desysopping, removal of extra permissions, and permanent blemish seems, in my judgement, a proportionate response to the behaviour recorded in the committee's findings. We would not, as a community, block this user if they were a non-administrator. This is just my view. The committee did not discuss the question because Level II procedures are pretty single-minded (and should really just be abolished as anything different to a normal motion). AGK ■ 21:00, 5 June 2019 (UTC)
• I don't know what's permitted to disclose, but has anything been heard from Od Mishehu themselves, by way of explanation?  — Amakuru (talk) 22:02, 5 June 2019 (UTC)
• @AGK: - while I too would like to know (to the degree possible) Courcelles' reasoning, I also wanted to give my appreciation for your response above - I felt it was as clear an explanation as the circumstances (of voting while generally inactive as an Arb) enabled. Nosebagbear (talk) 22:35, 5 June 2019 (UTC)
• I also reviewed the evidence related to socking here, and I would consider this rather iron-clad. I queried Od about the evidence when I was still on the Committee, and I personally found his response extremely lacking. It failed to explain the weight of the technical and behavioral evidence; there was plenty of both. The only reason I did not block at that time was because I felt the desysopping question needed to be answered first. Several experienced local CheckUsers weighed in with opinions that they would have blocked this editor had they not been an administrator at the time during discussions of the evidence. Now that the desysopping question has been addressed, I have blocked the account in my capacity as an individual CheckUser.
  Thank you to the Committee for pursuing this investigation to its conclusion. ~ Rob₁₃^Talk 22:54, 5 June 2019 (UTC)

  Thanks Rob, I think that ties it all up nicely. I get the point (expanded below too) that ArbCom does desysops and the Community does blocks - though that does gloss over cases where the Community can't see the evidence needed to decide on a block (in which case often ArbCom is the only body that can). In this case, however, the "individual checkuser" approach seems to be the key. Boing! said Zebedee (talk) 08:46, 6 June 2019 (UTC)

I have to say that I don't quite understand the Committee's response to what was discovered; mere admin-rights removal is nearly never appropriate, in my mind, except when those rights have been abused. If I logged out and then went around making edits like [1] or [2] (drawn at random from WP:AIV and a protected project page), either you also have reason to believe that I won't abuse further, and I should keep my user rights, or you don't have reason to believe that I'll stop abusing, and I should be blocked. I understand why more details aren't given, and I won't try to second-guess the {{checkuserblock-account}}. Nyttend (talk) 00:38, 6 June 2019 (UTC)

• @Nyttend: For what it's worth, if I had been on the Committee when the voting happened still, I would have supported desysopping but not a Committee-enacted block. It's not that a block isn't warranted. It's that it shouldn't come from the Committee, since it's really something for the community to handle (or the CU team, given the private info). The only piece that needed arb review was the sysop flag, and it's best to have as much separation of powers as possible in situations like these. Note that I've now blocked the account as an individual CheckUser, in keeping with that line of thought. ~ Rob₁₃^Talk 00:41, 6 June 2019 (UTC)
  • Ah, good point. Between AGK's view ("We would not, as a community, block this user if they were a non-administrator") and the actual result, I thought the committee was saying basically "We don't think further sanctions are appropriate" and I was somewhat surprised that you blocked when they hadn't. I hadn't thought of the difference between an Arbcom block and more "normal" block. Nyttend (talk) 00:54, 6 June 2019 (UTC)
    • @Nyttend: I can see where that would be confusing. I can't speak for any discussions after I left the Committee, but for what it's worth, there was zero talk about a block coming from ArbCom while I was still on the list, positive or negative. I would interpret AGK as speaking for himself, not the Committee, on this topic until a current arb says otherwise. ~ Rob₁₃^Talk 01:08, 6 June 2019 (UTC)
      • While making that comment, I was talking as a community member rather than as some kind of representative of ArbCom. AGK ■ 06:43, 6 June 2019 (UTC)
• To those who have expressed a concern about whether or not Arbcom should have blocked as well as desysopped, there is longstanding historical precedent here. On at least two occasions in the past that I am personally aware of, Arbcom has desysopped an admin for socking, while leaving the community to determine whether or not a block is appropriate. This is consistent with our community's blocking processes; Arbcom doesn't block accounts for socking, the community does. Even in cases where checkuser is used, it is a community blocking process that is activated: the majority of checkusers are acting "from the community" as opposed to acting as members of the Arbitration Committee. Even oversight blocks and checkuser blocks are issued by (specially authorized) community members, not by arbitrators acting in their role as arbitrators. It can get a bit kafkaesque when these groups intersect (we have active OS and CU who are also arbitrators), but the lines are pretty clear for those who hold the permissions. Risker (talk) 01:03, 6 June 2019 (UTC)
• Mmmm. There's a bit of the same thing that notoriously happens at AE from time to time - it doesn't matter how many people have looked at something and decided not to take action, because it only takes one to jump in. Opabinia regalis (talk) 08:20, 6 June 2019 (UTC)

  Well, the thing is either according to policies, or contrary to policies, or borderline. If it is according to policies, it is fine. If it is contrary to policies, it is not fine, and if the admin insists they were doing the right thing, at some point it will go to ArbCom to see whether them having access to the tools is still beneficial. Now, borderline means some people think it is according to policies and some think it is contrary to policies, and there could be some heated discussions about this, but in the end of the day if there is no consensus, it is no consensus, n'est ce pas? (A purely theoretical discourse, not specifically based on this case).--Ymblanter (talk) 08:31, 6 June 2019 (UTC)

  I also just don't think that's quite what happened here. ArbCom blocks aren't for CU evidence. That's what CU blocks are for, which are decided by CheckUsers as individuals or as a team. I don't consider the fact that ArbCom didn't block to represent the opinion that there shouldn't be a block. In this case, I blocked as an individual, but multiple opinions were given to me - on-list and privately - from other experienced locally-appointed CheckUsers that they considered the evidence and behavior sufficient to warrant an indefinite CU block. Given all that, "it only takes one" doesn't really apply here. ~ Rob₁₃^Talk 10:46, 6 June 2019 (UTC)

• So I just want to make sure I have this straight. Od Mishehu has repeatedly engaged in disruptive editing while logged out. The edits themselves were not egregiously harmful, consisting of nothing more than run-of-the-mill vandalism to articles non-article pages, but it obviously does mean that he has lost trust. The technical and behavioral evidence is incontrovertible, as determined by multiple checkusers and arbitrators; there is no reasonable doubt that these acts of petty vandalism were committed by Od Mishehu himself. His private correspondence with members of the committee failed to adequately address the concerns raised. As a result of his inappropriate conduct, he has been desysopped, and is now indefinitely block by BU Rob13 acting in his capacity as checkuser. Does that sound about right? Kurtis ^(talk) 07:31, 6 June 2019 (UTC)

  Yes, this is a good summary of the thread.--Ymblanter (talk) 07:49, 6 June 2019 (UTC)

  That's a shame. I liked Od Mishehu. Kurtis ^(talk) 19:25, 6 June 2019 (UTC)

  @Kurtis: Except that the edits were to non-article pages --DannyS712 (talk) 07:51, 6 June 2019 (UTC)

  Oh, sorry. Missed that part. Kurtis ^(talk) 19:25, 6 June 2019 (UTC)

• I obviously can't see the details, but I trust arbcom to have summarised them fairly. And on the summary facts as presented, the desysop was of course the right move.

I had many dealings with Od Mishehu over the years through their work at CFD. WE had occasional disagreements, but on the whole I found Od to to be a diligent and very competent admin who used good judgement in closing lots of discussions and handled lots of related technical issues with considerable skill.

So I am saddened to see the betrayal of trust by someone who I thought was one of the regular reliables.

Is there any way back? Or has the breach of trust been too severe? --BrownHairedGirl (talk) • (contribs) 09:39, 6 June 2019 (UTC)

Yes, it came as quite a surprise to me too, and seems out of character (at least the character I've seen under the Od Mishehu account). But there does seem to have been a suspicion that the same thing was happening way back in 2008. My guess is that an SO unblock appeal is the only way back now, though I doubt the community would be too welcoming as early as 6 months from now. Boing! said Zebedee (talk) 09:57, 6 June 2019 (UTC)

I don't think there is a way back to adminship, but I certainly hope there is a way back from the block - which (based on the public evidence) is not one that I would have placed. If there is no evidence of socking between now and then, I would expect to support an unblock with a strict one-account restriction in a few months. Thryduulf (talk) 10:04, 6 June 2019 (UTC)

The problem is, what is publicly revealed is (necessarily) very scant, and I could not myself offer a judgment without seeing the actual evidence. So I simply have to defer to one who can see it, and I consider BU Rob13 to be of sufficiently sound judgment to trust him. For any unblock consideration, I'd need to see a full revelation of what Od Mishehu was doing and the full extent of it, and I would not support an unblock without that level of honesty. Boing! said Zebedee (talk) 10:47, 6 June 2019 (UTC)

Please understand that such a "full revelation" could never include links to edits, since that would reveal Od's IP address. Even in this situation, that's not something we can reasonably condition an unblock on. ~ Rob₁₃^Talk 10:52, 6 June 2019 (UTC)

Oh, indeed. I was perhaps clumsy in suggesting "full revelation", when what I really meant was an explanation of the kind of edits Od Mishehu was making, the timescale of them, the reason for them... that would be a minimum for me. But saying that, those who know where to look can quickly discover exactly where Od Mishehu is (or was at a time in the past) located. Boing! said Zebedee (talk) 10:58, 6 June 2019 (UTC)

As I have nothing to add, I'll just echo what BrownHairedGirl said above, except that I don't remember having any disagreements with Od Mishehu. A respected co-worker at CFD, until now. – Fayenatic London 21:38, 9 June 2019 (UTC)

(edit conflict) The basic reason for my block is that I don't feel the editor can be currently trusted, but I too hope that trust can be regained. Based on the evidence I analyzed, Od disruptively vandalized content pages (not articles, but pages that contained content nonetheless). These are pages that our readers could navigate to from mainspace, in certain instances, not project pages. Moreover, he did not take ownership of this fact, instead continuing to evade scrutiny. Most importantly, he did this all from a role of extreme community trust, which makes the breach of trust so much worse. It will likely take an ownership of his conduct and a frank explanation of why he felt it necessary to start on the path to returning as an editor. ~ Rob₁₃^Talk 10:52, 6 June 2019 (UTC)

Thanks, Rob. That seems to me to be a reasonable approach. Thanks for acting on this issue, for being so diligent in explaining your actions. --BrownHairedGirl (talk) • (contribs) 15:43, 6 June 2019 (UTC)

• It is clearly incongruous for an absent Arbitrator to cast the only oppose vote in an otherwise clear cut case, appear to be not active in the case, and then not provide a response to questions despite WP:ARBCOND. What is actually worse is to see 3 Arbs attempting to explain this away as par for the course. Utter nonsense. No worries for me - all of these current members will be held to account when next they submit their names for AC membership. Leaky caldron (talk) 10:12, 6 June 2019 (UTC)
  • It's been just over 12 hours. I'm not even expected to respond to work emails that quickly, let alone calls for clarity on a vote I placed weeks ago. As noted above by an arbitrator, Courcelles expressed opposition to this desysop early on, before the Committee had the complete evidence and understanding of Od's conduct. He didn't just place a vote yesterday and then disappear. Maybe give it some time before you decide he's decided to refuse an explanation. We don't even know that he's seen this yet. You're literally demanding more from him than an employer would. ~ Rob₁₃^Talk 10:55, 6 June 2019 (UTC)
    • Another one carefully selecting the narrative. Leaving aside absence of a response which we might eventually get, the issue raised by others is voting in a case while shown inactive. Leaky caldron (talk) 11:29, 6 June 2019 (UTC)
      • Can you articulate why you believe that to be a problem? When arbitrators have sufficient time to be active on some but not all matters, why should they not do so? Should those who frequent AfD comment on all AfDs or none? Of course not; we're all volunteers. We contribute what we can. ~ Rob₁₃^Talk 11:41, 6 June 2019 (UTC)
        • "Can you articulate why you believe that to be a problem?" Because it's obviously causing an issue! Just like in the Rama case, a guy is down as one of the 2 drafting Admins. in the case and disappears for 3 weeks. It gives an impression of lack of situational awareness. To me; incompetence. Yes you are volunteers. Doesn't mean you cannot voluntarily get your collective acts together, does it? Leaky caldron (talk) 11:53, 6 June 2019 (UTC)
          • That pretty much reads as "I really like to criticize the Arbitration Committee" to me. The only issue I see here is you complaining, frankly. The outcome was correct here. The outcome in the Rama case was also correct, at least in my opinion. It took a bit to get there because people were busy. You are trying to demand that arbitrators place their volunteer positions above their own real lives, livelihoods, jobs, etc. That is just an entirely unreasonable demand. I don't see you volunteering to do that. ~ Rob₁₃^Talk 12:06, 6 June 2019 (UTC)
            • I volunteer when I can. Rarely. You are elected for trust and competence and if you set dates (as in the Rama case) there are enough of you to achieve that without messing about with piss-poor excuses. If you are too busy in RL, pack it in. Don't whinge at people like me who are holding you to the expectations of the job you scrambled up a greasy pole to get. Leaky caldron (talk) 12:14, 6 June 2019 (UTC)
              • This is a pretty strong stance. While arbitrators clearly have accepted responsibility beyond "we're all volunteers" this still strikes me as somewhere between ungenerous and strident and unlikely to land in a way to cause the desired change. Best, Barkeep49 (talk) 12:39, 6 June 2019 (UTC)
        • BU Rob13, in that case Courcelles cast a premature vote that didn't consider any subsequent details that the rest of the arbs considered, which would also be somewhat sub-optimal. Mr Ernie (talk) 11:53, 6 June 2019 (UTC)
          • @Mr Ernie: He gave a preliminary opinion, and didn't chime in before this needed to be acted on. Other arbs above noted that they probably could have put him as inactive instead of opposing given the new information, but then that creates a sub-optimal situation where you're giving the impression of unanimity when that's not the full picture. It's always sub-optimal to have an inactive arbitrator, since more voices are helpful. It's unavoidable when off-wiki life rears its head, though. Perfect ideals are hard to reach. ~ Rob₁₃^Talk 12:06, 6 June 2019 (UTC)

@BU Rob13: Should User:Od Mishehu AWB also be blocked? Boing! said Zebedee (talk) 11:38, 6 June 2019 (UTC)

• @Boing! said Zebedee: I haven't checked alt accounts, etc. Obviously, if he edited from that account, it would be block evasion, but I'm not super interested in chasing down all the accounts if there isn't evidence he's using them. Admins often have loads of doppelganger accounts to prevent impersonation, etc. It's probably a waste of time for me to chase them. ~ Rob₁₃^Talk 11:41, 6 June 2019 (UTC)

  I wasn't suggesting any negligence on your part and I'm obviously not suggesting block evasion, I'm just asking a question. It seems obvious to me that when the person is blocked, all their known accounts should be blocked. I will make it so. Boing! said Zebedee (talk) 11:46, 6 June 2019 (UTC)

  Of course! I didn't think you were. Feel free to block away for the other ones. ~ Rob₁₃^Talk 12:29, 6 June 2019 (UTC)

• @BU Rob13: - the issue of Courcelles casting a vote, becoming inactive and then more evidence being identified did leave you with an interesting dilemma there. Both false unianmity and the issue of an Arb casting a vote before seeing all evidence, are serious negatives. Perhaps a two line summary, but I can see the issues with that as well. Hmmmm. Nosebagbear (talk) 12:18, 6 June 2019 (UTC)
  • Yeah, it's a difficult situation. I think the community often doesn't fully appreciate that all situations the Committee face suck, barring those that we can easily redirect to other venues. If they didn't suck, they wouldn't have reached the Committee. We try to make them suck as little as possible, and that's really all we can hope to do. ~ Rob₁₃^Talk 12:29, 6 June 2019 (UTC)
• As folks are referring to it, I thought it may be helpful to copy my earlier comment to here. SilkTork (talk) 12:57, 6 June 2019 (UTC)

Under arcane ArbCom procedures, an Arb can be inactive, but become active on any issue by commenting on it. The checking of the findings by the functionaries and the resulting discussions and clarifications and double checking and looking into possible alternative explanations took the Committee some time. The functionaries passed their findings to us around 13 May, and we have been discussing it since then, with several emails to Od Mishehu. Courcelles commented on this issue while it was being discussed on the email list, and made it clear he did not support a desysop. At the time he said that, there were still a few other Arbs, including myself, who were not sure. After further investigation of the technical and behavioural evidence, and more emails to Od Mishehu, all the Arbs came round, though Courcelles was still logged as not supporting. I cannot say for certainty if at this point Courcelles would still be opposed to desysopping, but that was his last indicator to us, and it came, as I said, at a stage when we did not have the clarity we do now. Should we, in retrospect, have marked Courcelles as inactive? Well, that would have gone against his last statement to us, so I'm not sure. SilkTork (talk) 23:15, 5 June 2019 (UTC)

This comes to mind. The Blade of the Northern Lights (話して下さい) 04:18, 7 June 2019 (UTC)

• This should just be basic procedure 101. A jury does not vote before all evidence is heard. If substantial new evidence is heard, all votes made before the new evidence came to light should be declared void, and those arbitrators should have been required to check in again to either re-affirm or change their early (premature) vote. Therefore, because Courcelles did not check in to cast his "final answer", he simply did not vote for the purpose of reporting this decision. I'm fine with inactive arbs checking in out of the blue, but not if they can't get the timing of when they check in right. – wbm1058 (talk) 12:06, 7 June 2019 (UTC)

Just for clarity. All evidence was in, and most arbs had indicated their views (or given their votes). What happened between then and the final decision was that there was more discussion which helped to clarify matters for me (and I had more time available to examine the evidence and the analysis of the evidence), but if the decision had been taken at the point that Courcelles voted, the desysopping would still have carried with a significant majority. I think, other than discussion, all that Courcelles missed was the final response from OM, which, like the previous responses, was very limited (I think monosyllabic would be an appropriate description). It was a simple denial. I cannot speak for if Courcelles would have changed his mind or not from that.

However, as the community are raising some concern about a public posting in which a Arbitrator's vote has been shown, and the background is the vote was included in less than ideal circumstances, we could raise a motion in which we vote on how to deal with such matters going forward (no votes given more than one week previously to be included if there has been no contact from the Arbitrator for more than one week, for example). In this case the vote was not significant, but if a vote is decisive, and the person casting the vote did so at an earlier stage of discussion, and they are unable to be contacted, then the decision to keep or remove the vote does become significant. SilkTork (talk) 16:01, 7 June 2019 (UTC)

I think passing such (a similarly worded) motion will be beneficial for the project.--Ymblanter (talk) 16:33, 7 June 2019 (UTC)

For the off-wiki decisions, is there a vote to close, as for on-wiki Arbcom cases (I guess it would be a vote to post the announcement, but separate from the underlying vote in the matter at hand)? If an arb votes in the matter and later votes to close, that's basically a confirmation of their initial vote. So if an arb doesn't vote to close, and the "votes to close" were disclosed to the community, the community would know that the vote wasn't confirmed (for whatever reason). That might clarify situations like these to the community. – Levivich 08:08, 8 June 2019 (UTC)

• Despite assuring the community in his statement that he is been back to full activity all this year, and am fully prepared for the workload and e-mail volume if I get elected again, Courcelles has only edited 4 times since winning election to the committee. This is a disservice to those who took him at his word and unfair to someone whose spot on the committee he took. And now we can't get an explanation from him on why he voted before all the evidence in this incident was known. ArbCom's likely got a very difficult case coming up, and with the loss of Rob and now Joe's Wikibreak, they will be quite shorthanded. Mr Ernie (talk) 13:05, 7 June 2019 (UTC)

  It didn't help that the community agreed last autumn to reduce the number of seats on the Committee. Courcelles was last-in at 70.23%, narrowly beating out DGG (69.39%). Some consideration should be given to the concept of having one or two "alternate jurors" available to fill seats vacated by resigning members. Alternates would be required to run for reelection at the next election. – wbm1058 (talk) 13:42, 7 June 2019 (UTC)

  As Iridescent explained at a recent case, more arbs does not equal quicker response times or more efficiency because of more hands. It means more of a wait and less getting done on ArbCom because the number required for a majority increases and all active arbs are treated as default oppose when doing the math to calculate votes. The advantage of having more arbitrators is diversity of opinion and making the committee move more slowly on things that maybe shouldn't be rushed into. It is not in creating less work. TonyBallioni (talk) 13:45, 7 June 2019 (UTC)

  So rather than starting out with a too-large and inefficient Committee with the anticipation that members resigning and/or going inactive will inevitably reduce it to the "just right" size, start out at the "just right" size with alternates available to keep it "just right" after the likely resignations or unanticipated issues causing inactivity happen. It's been over ten years since Jimbo last appointed someone to fill a vacant seat, and I think a standard procedure for filling seats with alternates who "just missed out" in the last election is worth consideration. – wbm1058 (talk) 14:21, 7 June 2019 (UTC)

  I think there's a stronger argument for actually reducing the number of seats on the committee to under the current limit. I think the ideal is probably 7-9 active members, which means you could probably reduce the size to 11 or 12 total arbs. For those who worry about needing more, 13 is a good number because it adds in extra cushion. TonyBallioni (talk) 14:26, 7 June 2019 (UTC)

• 1)I thought it was dumb to have less members. 2) But at any rate, I can't see why this matters, it is and was always a majority of the ctte that carries the motion, so some members might be 'wrong', they might be lazy, they might be distracted, they might just see it differently, or whatever, and it is going to be usually a discussion where members don't just vote, they talk to each other, and over a long or short period of time a majority forms, and it's done, every extra vote after the majority is fundamentally superfluous, because the outcome is the same, regardless. -- Alanscottwalker (talk) 13:19, 7 June 2019 (UTC)

  In clear-cut cases like this that's true, but there have certainly been cases where a majority has been reached for a particular position, and later interventions subsequently cause enough arbs to change their mind as to affect the outcome of something. Wikipedia:Arbitration/Requests/Case/Infoboxes/Proposed decision#Proposed remedies is probably the highest-profile example of this happening. ‑ Iridescent 13:53, 7 June 2019 (UTC)

  Well, the final tally here suggests at the end it was about as clear cut as clear cut gets. (Although, I could have made my last comment clearer because there is an unspoken ellipses. It was meant to read: 'every extra vote after, in the majority') Alanscottwalker (talk) 14:01, 7 June 2019 (UTC)

As regards size of the Committee. We currently have only eight. And there is an expectation that everyone takes part in everything. That puts a significant burden on the eight. In my previous time on the Committee there were more of us, and people specialised, and if tired or stressed someone could drop out for a bit. With the current burden being carried by such a small number, I would not be surprised if one or more members drop out before the end of the year. When tired or stressed, a person is less able to cope with the criticisms that come as part of being on the Committee. The current appeal procedure where we need to have four arbs voting on every appeal slows thing down considerably. It's not a large amount of Committee members that slows things down (I'd like to swear very loudly here because that notion is so mad!) it's the set up of the Committee which may slow things down. We should have quorums appropriate to the size of the Committee. If you have 4 Committee members and you require a quorum of four for every decision, then you have to wait until every Committee member has gone through every discussion before all decisions are finalised. In real life that is not going to happen as new stuff is coming in all the time. If you have a Committee of 8 with a quorum of four you are halving the workload and increasing the likelihood of decisions being made speedily. A Committee of 16 with a quorum of four works even better. A Committee of 16 with a quorum of three works better still. And we don't need every Arb to be sitting on every case. Up the Committee to 16, have a quorum of three for appeals, and restrict cases to only eight Arbs at any time. Less pressure on everyone. Work gets done quicker. And, yes, if someone resigns or goes ill, it would be helpful if they were replaced by the next person on the list. SilkTork (talk) 16:01, 7 June 2019 (UTC)

@SilkTork: in general I'd support expanding the committee to many more members, however in current practice I don't - and the reason is that every single person on the committee also become an investigator and stand-alone functionary (that is: they become checkusers and oversighters) not just dispute resolution experts. I'd want to see the CU/OS process devolved to dedicated teams unrelated to "being on the committee" before expanding the committee. (Wherein the OS/CU teams would simply report findings to the committee as evidence for their primary dispute resolution mandate). — xaosflux ^Talk 17:26, 7 June 2019 (UTC)

I was the one who made the proposal to reduce the size, but I don't think that it really took into account this many people having gone long-term inactive or resigning. I think this should be revisited next year (but I don't think that we should just blindly revert back to how things were). --Rschen7754 18:29, 7 June 2019 (UTC)

The idea of replacing resigned or long-term inactive arbs with the next in the ArbCom election list (if they're still willing) seems like a good one to me. Boing! said Zebedee (talk) 18:47, 7 June 2019 (UTC)

Perhaps just as bad, the only way to reduce the workload is to just say 'no', 'no', 'no', 'regardless of merits, just can't take it on.' Not that you would set out to be the ctte of 'no' but that is the incentive. -- Alanscottwalker (talk) 21:25, 7 June 2019 (UTC)

• Further comment on size. I always thought size was a good release valve related to the voluntary nature of the position. Sure it would be great to have 24/7 on duty working away for two years, but that seems wholly like a kind of servitude. Because the 'check-out' is bound to occur and we don't want servitude (or maybe we do, but don't want to admit it) we need room (capacity) for checking out. Alanscottwalker (talk) 20:03, 7 June 2019 (UTC)
• Some other comments on size. Not only has it always proved the case that not all arbs are able to complete their term, it is also the general practice of arbs to become temporarily inactive when they will be away from WP for more than a day or two, or be in a situation where they do not have secure communications. Some have even taken vacations where they have no internet access at all. Additionally, for some case a substantial number of arbitrators have considered themselves involved and been inactive for that particular matter. Even more important, there is a need for diversity of opinion. Arb com does not work by consensus. There is always an attempt to find a generally agreeable solution--and if possible a unanimous solution--but since it is the purpose of arb com to provide a definitive settlement of a dispute, remaining disagreements are handled by voting. In most non-obvious cases, although there is almost always unanimous agreement of the general principles, the specific sanctions are rarely unanimous, and sometimes come down to a bare majority. Most of the maters arb com ow deal with in formal cases , and especially the applicability of sanctions, rely a great deal of personal judgment about what it likely to be effective, or on whether the likelihood of recurrence will make indefinite sanctions necessary. There is no way of making such judgments in an objective fashion. The arbs have to go by their views. In such situations a larger panel is fairer and offers the only possibility of reasonable consistency. All proposals to use a subcommittee or a small quorum will be subject to variation with the individuals involved--which is also why most judicial or quasi-judicial systems do not make final decisions except by a panel. The real limitation on the size of the panel is that larger panels in such a system always take more time to decide. DGG ( talk ) 05:09, 8 June 2019 (UTC)
• Looking at this the only question I have is why User:Courcelles opposed this? Govindaharihari (talk) 05:29, 8 June 2019 (UTC)
  • I agree that this is an interesting question. I'd also like to commend ArbCom, however, on deciding not to treat Courcelles as inactive and thereby be able to publish a unanimous decision. Courcelles apparently did not comment late in the discussion and may have come to a different view – certainly some arbitrators took more time and evidence to be convinced – but the explanation for the "oppose" vote is for Courcelles to provide. As has been suggested above, ArbCom might want to look at its internal procedures allowing votes to stand even when evidence becomes available subsequently – indeed, that has been a problem on-wiki in cases in the past. Calling for an affirmation before a decision is implemented that conclusions / votes stated before X date have not changed might be one option. However, under the current procedures (as far as I understand them and as described in this thread), I think ArbCom took the correct action in posting Courcelles' opposition and I thank them for so doing. This was the transparent approach, chosen even when it was likely to provoke questions and possibly criticism, and when the expedient and likely uncontroversial placing of Courcelles as inactive was possible and appeared to carry little risk. After all, unless Courcelles objected publicly to being categorised as "inactive" rather than "opposed" in the posting of the decision to WP:ACN, the community would be unaware of the dissenting opinion / view. I am aware that treating ArbCom as not part of the community can be a problem, and certainly treating Arbitrators as individuals as not part of the community is unfair, unkind, and can be a cause for distress. Acting on a case like this does demonstrate that the Committee is separated somewhat from the community by its special authority, but by deciding only the sysop question and leaving the block question to the community and its checkusers – and thus only acting to the extent needed even if its power extends further – shows that the Arbitrators respect their community-derived mandate does not place them above the community, and that being part of the community means not usurping decisions that are resolvable through community mechanisms. I short, I'm interested to hear from Courcelles and I might have questions (such as the one below) but overall I'm grateful for and impressed by the individual Arbitrators / ArbCom collective decisions and comments here. EdChem (talk) 07:44, 8 June 2019 (UTC)
• BU Rob13 has imposed a CU block now that ArbCom has desysopped and it is suggested that other CU would have done the same if Od were not an admin. I would like to know why such a CU block was not imposed first and then the matter referred to ArbCom for consideration of Od's sysop status. If ArbCom were never going to block, and with recent changes preventing self-unblocking, why would the CU team not act on their authority to block on CU evidence? ArbCom would hear from Od and if they were convinced that the block was unjustified, retain the sysop flag and unblock... and if not, remove the sysop flag and leave the CU team and / or community to deal with any later unblock request. In the unlikely event that ArbCom favoured the desysopping but viewed the block as excessive, they could discuss with the CU team off-wiki, or open a community discussion providing whatever information is available for public viewing. Blocks are meant to stop disruption and waiting to impose one based on CU evidence while ArbCom deliberates on the sysop flag risks disruption, such as the now-alerted user who knows a desysop is likely coming to activate a sleeper and grant it rights. If the CU evidence justifies a block, impose it – whether the editor is "ordinary" or an admin or has advanced permissions or is an oversighter or a bureaucrat or a founder, etc – and then pass on the question over the advanced permissions... doing otherwise is to act like some editors are more equal than others. EdChem (talk) 07:09, 8 June 2019 (UTC)
• "Log in every day for two years" is a lot to ask of a volunteer. Staggered terms have obvious advantages, and we don't want to have elections too frequently (and one-year terms with elections every six months is probably too frequently). I agree that to compensate for the seemingly-necessary two-year commitment, the system must be set up in a way that makes it easy for an Arb to step away and come back. That means having a pool of alternates ready to step up when someone steps down. We can't have too many people making every decision, but I think SilkTork's suggestion of panels could provide the deep roster without requiring a long bench. We can have an arbitrator pool of many members, but individual cases are solved by smaller panels of arbs. The (very real) bias problems inherent in smaller panels that DGG raises can also be ameliorated. One way is to have panels formed at random by lottery from among active arbs (this may not be the best way, though, as we may want subject-matter-familiarity to influence panel composition). Another way is to allow appeals from a panel to the "full Arbcom" (all active arbs). This is basically how many courts work. Also, xaos's suggestion of separating the sensitive-private-info, off-wiki functions (the email list stuff), and the on-wiki dispute resolution functions, seems like it could help with the too-many-emails issue. The email list stuff could be handled by a particular panel of arbs, and that way not everyone will get 100 emails a day, just a lucky few :-). But those arbs, conversely, would not have to be part of on-wiki case panels, except maybe for an appeal to the full arbcom. That seems to be the best of all worlds: small-enough panels (a short bench) to be able to process matters without waiting for 10 people to chime in; a division of labor so no individual volunteer has to follow two cases, three appeals, and a bunch of emails every day; a deep roster so volunteers can take breaks when needed; and protections (random panel selection, appeals) against small-panel bias. Has anything like this (separating the list and dispute resolution work, having smaller panels) been tried before and how did it work, or what were the problems? – Levivich 08:08, 8 June 2019 (UTC)

Wikipedia:Arbitration/Requests/Case/Rama closed

Original announcement

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Is the edict here that a TP cannot be watched factually correct? [3]? It seems improbable. Leaky caldron (talk) 15:25, 6 June 2019 (UTC)

This case is now closed and pages relating to it may no longer be watched refers to users, such as arbitrators or clerks or parties or whomever. Hence, the helpful links about where various things should instead be directed. ~ Amory (u • t • c) 15:37, 6 June 2019 (UTC)

(ec) Seems you're reading too far into this. GoldenRing already explained to you what it means I believe that message has been used for a long time without anyone interpreting it that way. It's just a simple reminder that the page may no longer be watched hence any new post may go unnoticed. (People generally unwatch case pages once they're closed). It's similar to several other messages elsewhere like this and this. In short, it's not an edict to not watch the page, that's quite unreasonable ruling to even make in the first place not talk of being impossible to enforce technically. – Ammarpad (talk) 15:44, 6 June 2019 (UTC)

"May" can mean "have permission to" though... if it had said "might no longer be watched" there would be no confusion.-- Pawnkingthree (talk) 15:56, 6 June 2019 (UTC)

This isn’t a big deal... TonyBallioni (talk) 15:58, 6 June 2019 (UTC)

Indeed it is not. -- Pawnkingthree (talk) 16:01, 6 June 2019 (UTC)

The FACT is that the page can be watched. So the FACT is that it is wrong. Ammarpad I do not read "too far" into anything - I just read. And GoldenRing did not "already" explain anything to me. He replied to someone else's' question. But thanks for you ever so carefully crafted explanation of a palpable error. Oh, and Tony thanks for your helpful input. I don't think that it is a big deal either. Just a minor point that could be easily fixed but would probably need a 10 week RFC :). Leaky caldron (talk) 16:11, 6 June 2019 (UTC)

(ec) You throw a lot of innuendos around. (Many of them toward multiple editors on this page alone). It seems it's one of the things that you enjoy doing here. Good for you. – Ammarpad (talk) 16:22, 6 June 2019 (UTC)

No, you are just reading the sentence incorrectly. I may not see it is perfectly grammatical English and may is being used in the same sense here. The point of my previous statement is that in light of the explanation given by both Amory and myself, there really is no need to even be having this discussion. TonyBallioni (talk) 16:16, 6 June 2019 (UTC)

The solution to that is in your own hands, or fingers. Leaky caldron (talk) 16:20, 6 June 2019 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

On the PD talk page, it was noted that an IP had posted an offensive message on Rama's talk page. The post was immediately reverted by (I think) non-admin ArbCom Clerk Bradv, which was unquestionably the first step to take... but it seems to me to be a comment that falls under WP:RD2. I suggest that trolling like that post should be removed from being viewable in the page history, and that leaving it accessible is disrespectful to Rama. No matter the view taken of Rama and the case, the comparison made in the IP's comment was grossly offensive, unjustifiable, unfair, and malicious. I agree on the ignoring part of RBI and Arbitrator Mkdw blocked the IP, but I suggest that the "R" part is incomplete without RD2-ing of the IP's post... or is there something I'm missing? EdChem (talk) 06:00, 8 June 2019 (UTC)

Hi EdChem. The message was definitely a violation of Wikipedia's civility policy and made personal attacks. Removing it was definitely the right thing to do. However, this falls under "ordinary incivility", which is not eligible for revision deletion. This is not something that falls under being "grossly insulting, degrading, or offensive". ~Oshwah~^{(talk) (contribs)} 06:08, 8 June 2019 (UTC)

Thanks for your response, Oshwah, but I disagree. Comparing a user to being a member of an organisation that was involved in committing crimes against humanity and deliberate / planned and targeted torture, enslavement, and mass murder of millions is not "ordinary incivility." It is, in my view, unquestionably "grossly insulting, degrading, or offensive" to apply such a comparison to Rama. EdChem (talk) 06:55, 8 June 2019 (UTC)

No opinion either way, but I will just point out that you are doing the very opposite of the I in RBI at the moment — Martin (MSGJ · talk) 20:12, 8 June 2019 (UTC)

For what it’s worth, it’s been revdel’d. I agree with Oshwah that it doesn’t qualify, but it also falls in my camp of “meh, it’s been hidden, no good is done by reversing” I’ll note that both the post here requesting revdel and the actual revdel itself actually increase the number of people who saw the comment as the surest way to get admins to look at a revision is for another admin to hide it. TonyBallioni (talk) 20:21, 8 June 2019 (UTC)

Concerns of this nature in the future should be submitted to the clerks (or the Committee, if oversight is requested). If discretion is required, the clerks can always be contacted at clerks-llists.wikimedia.org. Taking off the clerk hat for a moment, I would suggest that referring to specific incidents of incivility on a page watched by over 1000 users is counterproductive. Best, Kevin (aka L235 · t · c) 20:43, 8 June 2019 (UTC)

• Glad it has been removed
• I wouldn't have raised it here but a Clerk removed the post and an Arbitrator blocked for it, yet neither acted on RD2, and I suspect not about how Rama might feel about it. Further, I am astonished that the comment was not seen as grossly offensive by some. I find the fact anyone had to ask is disappointing... as is the suggestion that seeing the comment is more problematic than WP and its admins saying to Rama, to those seeing the comment, and to Wikipedians in general that the comment was "ordinary incivility." If ArbCom really thinks civility needs addressing then surely it can agree the comment in question was grossly offensive, degrading, or insulting? EdChem (talk) 09:17, 9 June 2019 (UTC)

  @EdChem: I can't see the comment anymore, but my initial reaction was that it did not qualify for RD2, which specifically excludes "ordinary" incivility, personal attacks or conduct accusations, and so I didn't request revdel at the time. Perhaps there needs to be some clarification sought on what constitutes "ordinary" incivility, bearing in mind of course that routinely hiding personal attacks from view makes it more difficult to enforce the civility policy. – bradv🍁 13:21, 9 June 2019 (UTC)

  Bradv, it appears that the majority of admins / editors find that the troll's post was "ordinary incivility." I fail to see how claiming a named editor is associated with / effectively a member of a banned criminal organisation known for torture, mass murder, and crimes against humanity is not a BLP violation, which qualifies for RD2 even if the claim is not grossly degrading / offensive / insulting. It has been suggested on my talk page that if I want such things covered by RD2, I need to start an RfC. My view is that if standard practice does not view such comments as covered by RD2, its text needs change as practice does not match what is documented. We also need to remember how the target of such attacks might feel, and how onlookers perceive our choices given the wording of RD2... because if the claim made is not grossly offensive / insulting etc, then my understanding of the phrase is significantly different from that of many others here. EdChem (talk) 15:34, 9 June 2019 (UTC)

• RD2 and RD3 are wildly open to interpretation. Beeblebrox (talk) 17:49, 9 June 2019 (UTC)

Didn't see the revdel'd comment but if the "organization" Edchem mentions is the one best known for WWII, I thought per Godwin that comparing the other person to them was supposed to be the outcome of every extended internet disagreement. And "BLP violation" against an anonymous editor? Meh.

Regarding the arb case, the outcome is a bit saddening, so my sympathy to Rama. The initial IAR was perfectly good, but fighting over it against community pushback wasn't going to work. And the initial AfD was at best tasteless. 173.228.123.207 (talk) 18:52, 9 June 2019 (UTC)

Request for ArbCom to comment publicly on Fram's ban

Not sure where to put this; it isn't a case request. I don't think, anyway. I'd appreciate a clerk moving it to where it should go if it doesn't go here, or telling me where to move it to.

I do not need to know why Fram was banned; possibly privacy-related info that is none of my business. But I do think ArbCom needs to find out what is going on, and publicly state whether or not they agree with Fram's recent 1 year WMF office action ban from en.wiki (and en.wiki only), talk page access removal, and desysop. If you guys say this is OK, I'll defer to that (tho I suspect others might not). But I'm having a hard time imagining a situation where a 1 year local ban should be an office action, rather than something that ArbCom deals with. --Floquenbeam (talk) 18:24, 10 June 2019 (UTC)

@Floquenbeam: I'll send an email off to the clerks list and see if I can get some information from the arbs on where this should go for you. --Cameron₁₁₅₉₈ ^(Talk) 18:43, 10 June 2019 (UTC)

Cameron11598, this is probably the best place. Floq, This has come from WMF, not Arbcom, so I'm not sure that we are the best people to answer you. Perhaps your first port of call should be one of the WMF community liaisons? Worm^TT(talk) 18:57, 10 June 2019 (UTC)

@Worm That Turned: Previous observation tells me that they will tell me nothing, as I don't have standing. I will just be assured that the siteban was necessary to protect the community, or some such. Plus, about the only person in the whole WMF I actually trust implicitly is @Mdennis (WMF):, and I doubt she's in the loop with T&S blocks. ArbCom, on the other hand <suckup>I do trust</suckup>. My whole concern is that being sitebanned from en.wiki only has just got to be ArbCom's business, so I'm asking you guys to find out what's going on, and give it a thumbs up or thumbs down. I gather from your response that, at the very least, you guys weren't given a heads up ahead of time? --Floquenbeam (talk) 19:03, 10 June 2019 (UTC)

Worm, looks like I was right. --Floquenbeam (talk) 21:04, 10 June 2019 (UTC)

Were you guys aware of anything relevant surrounding the locus? ∯WBG^converse 19:12, 10 June 2019 (UTC)

• Guys, you have regular meetings with T&S - like Floquenbeam, I ask if you can give us a in-the-know "was reasonable/wasn't". If you weren't in-the-know, then could you both let us know that, and then put it on the next agenda? Nosebagbear (talk) 20:31, 10 June 2019 (UTC)
  • Posting up here because I'm specifically answering the question about a heads-up from the WMF, and the thread meanders a bit. Arbcom found out this was implemented at the same time everyone else did. We did hear in advance that an action to do with Fram was under consideration - "we" in the person of me, because this was a regularly scheduled conference call and I happened to be the only arb available to participate - so I feel obliged to clarify my own understanding here, but I'm speaking for myself only. (The rest of the arbs did have the minutes from the meeting available shortly after.) The call was not specifically about this; it was pre-scheduled and this was one of several unrelated items. That information was not final and did not get into the specifics of their investigation. In past conversations - speaking in general, not about any individual, and IIRC around the time of the first project-specific ban on another wiki - I've expressed significant skepticism about the concept of a ban that is both WMF-originated and project-specific. But that is a general opinion on philosophical grounds, not based on the details of any specific instance. (I will say this - I get that it's frustrating to see an unexpected and dramatic action and not hear any explanation from whoever enacted it. But arbs are often on the other side of that fence and it's also frustrating to be unable to publicly articulate the basis for a decision. Not an argument for anything in particular, just sympathy for both sides of this awkwardly placed fence. Also, in the interest of full disclosure, I've had some differences with Fram in the last couple of years - in fact I recall he came to my talk page just to tell me "fuck you", which if nothing else came in handy when I later wanted to defend someone else saying "fuck you" - but I obviously don't consider those differences banworthy.) Opabinia regalis (talk) 09:10, 11 June 2019 (UTC)

I think an ArbCom comment is needed here. Were you consulted or even informed about this ban? Are you aware of why WMF took direct action rather than referring the matter to you? Do you support the actions WMF have taken? No doubt there are other questions ArbCom could usefully address, and others will do doubt add to the list. This appears to many of us to be a case of WMF having usurped your role. I for one would be grateful for any comments/explanations you can provide. WJBscribe (talk) 22:29, 10 June 2019 (UTC)

And if, as I suspect, the answer to the above questions is "no" / "no" / "no comment", could we further request that you proactively contact the WMF and obtain a confidential summary of the reasons for the block? You could then, without revealing any further details, say whether you agree that it was justified or not. As ArbCom are subject to confidentiality agreements and inside the Chinese wall as far as privacy goes, it seems to me it would be reasonable for WMF to furnish them with such information if they requested it. They probably wouldn't play ball but it might be worth a shot, because it would go a long way to allaying community annoyance. Thanks  — Amakuru (talk) 22:34, 10 June 2019 (UTC)

Add me to those saying this needs an Arbcom comment. From what little I can glean from the WMF's non-answer answer, this appears to be a case of the WMF unilaterally banning one of our longest-serving admins on the basis of an anonymous complaint. Had it been a permanent global ban I'd shrug and assume there was good reason this couldn't be discussed publicly, but since this is specific to a single wiki and time-limited, it looks a clear-cut case of the WMF unilaterally carrying out a hit on an editor against whom they've taken a dislike because they didn't trust the committee to come to their preferred conclusion when provided with the relevant evidence. ‑ Iridescent 22:36, 10 June 2019 (UTC)

• I was on the committee when the first office bans occurred. Unless things have changed, they would tell us absolutely nothing about why they were issued (even in cases wher ewe pretty much knew anyway, having forwarded the matter to them). Standard CYA behavior, almost certainly at the behest of legal counsel, so I doubt the committee can crack the code here either. Beeblebrox (talk) 22:38, 10 June 2019 (UTC)

  They've already told us this ban was as a result of "a complaint from the community", but having gone through every post by Fram in the last couple of months I can see no complaint of any kind being made about him (other than a relatively minor scuffle here which certainly isn't ban-worthy). I can't see how this could be a legal issue, or the ban would be global and permanent; a non-appealable one year ban on a single project just doesn't cut it if there's a genuine legal concern. ‑ Iridescent 22:45, 10 June 2019 (UTC)

  I didn't mean to imply there was a legal concern witht he ban itself, but rather that counsel would've advised the WMF not to ever explicitly say why they banned anyone, which they never do. So unless Arbcom was where the complaint originated, they likely know no more about about it than the rest of us. Beeblebrox (talk) 22:50, 10 June 2019 (UTC)

  This would be completely reasonable if Fram had been globally banned. But he hasn't been, only from enwiki, and only for a year. If I was a member of ArbCom I'd be asking to know why - if nothing else, for information in case I was involved in a future ArbCom case involving Fram when his ban expires. Since they apparently haven't been, this makes no sense. Black Kite (talk) 23:03, 10 June 2019 (UTC)

  If you are right then, were I an Arbitrators, I would find my position untenable in such circumstances. The "non-answer answer" suggests that action was taken because this fell into the category of "cases where local communities consistently struggle to uphold not just their own autonomous rules but the Terms of Use, too." It is hard to see how enwiki hasn't delegated responsibility for these issues to ArbCom insofar as they relate to user conduct. Therefore, WMF Trust & Safety appears to be saying that, in Fram's case, ArbCom was failing to uphold our rules or the Terms of Use. That would make life pretty impossible for any sitting Arbitrators. WJBscribe (talk) 22:58, 10 June 2019 (UTC)

  @WJBscribe: - I'd have to disagree, despite the dramatic statement it might make. They'd be in an impossible position if the people who gave them their authority lost trust in them. The WMF doesn't give ARBCOM authority. The Community does. I have not lost trust in them, and losing them would not help en-wiki. Nosebagbear (talk) 23:02, 10 June 2019 (UTC)

• As for who was involved in this action, it looks like everyone in the office, including Legal, Maggie Denis and the ED. Alanscottwalker (talk) 22:48, 10 June 2019 (UTC)

Am I misreading the stuff at that link, or does it indicate that there is no appeal available for WMF Office bans? I can't see how that can possibly be acceptable. rdfox 76 (talk) 00:24, 11 June 2019 (UTC)

Yes, that is the policy. Alanscottwalker (talk) 00:27, 11 June 2019 (UTC)

Well, that's... disturbing. Am I the only one who finds it uncomfortable to think that the WMF can issue bans with no option for appealing them even to the WMF itself? rdfox 76 (talk) 00:31, 11 June 2019 (UTC)

There's always the Board of Directors, with longtime community representation there. They have override if necessary, and the Staff have to disclose to them anything they legally can. Georgewilliamherbert (talk) 02:01, 11 June 2019 (UTC)

• The committee has been following this thread and the one at BN. We are discussing the matter. Mkdw ^talk 23:17, 10 June 2019 (UTC)
• To add to Mkdw's comment, there is not a lot the Committee can say at the moment without discussing the matter first with the Foundation. However, I feel that I can say that this is not something the Committee asked the Foundation to do, and the Committee received the news regarding the ban at the same time as the community. I have written to the Foundation as an individual asking for the ban to undone, and for any concerns regarding Fram to be brought to ArbCom for the Committee to deal with appropriately. However, at this stage, I am unsure exactly what recent concerns the Foundation have regarding Fram that prompted them to ban him. And if they are local concerns, then either ArbCom should deal with the matter or the Foundation should decide that they will deal with local matters on en.wiki, and disband ArbCom and set up an appropriate interface between the community and the Foundation where matters like this can be openly discussed. There is no benefit to anyone in having secret bans. That merely creates a state of fear. SilkTork (talk) 00:06, 11 June 2019 (UTC)

  SilkTork, I think I can imagine a situation in which law enforcement was involved, and it would not be appropriate to expect the usual dispute resolution processes to be followed, culminating in an arbcom case, but I believe the office has explicitly excluded legal considerations. If it is community related rather than legal, I'd be hard-pressed to disagree with your conclusion as unsettling as that would be. S Philbrick(Talk) 01:21, 11 June 2019 (UTC)

Agreed. For an organisation that champions freedom of expression, openness and diversity, to be meting out justice in a manner similar to a tinpot repressive dictatorship is bizarre and unbefitting of its aims.  — Amakuru (talk) 00:15, 11 June 2019 (UTC)

• I feel like I have fallen through the looking-glass on all this. Up is down, down is up. I would welcome someones in a position of trust here on EnWiki - the CUs or ArbCom or the Stewards or somesuch - that they be given access to the details of the case and let us know if they agree or not. This action means that any editor on Wikipedia can be banned in an extrajudicial fashion with no possible appeal...I didn't even know that was possible. And all of a sudden I feel like we are all on shaky ground... Shearonink (talk) 03:19, 11 June 2019 (UTC)
• I shall be inactive on ArbCom until this matter is resolved one way or another. There is frustratingly little information at the moment, but hopefully the Foundation will soon be able to explain this action to everyone's satisfaction. SilkTork (talk) 06:12, 11 June 2019 (UTC)
• Add my voice to the call for arbcom to (a) be notified of the reasons and (b) communicate whether justified or not (happy not to know if the committee is satisfied one way or the other). I can't think of a reason for a 1-year ban restricted to en.wp based on my previous experience with the committee and office-bans etc. Cas Liber (talk · contribs) 08:04, 11 June 2019 (UTC)