Wikipedia:Arbitration/Requests/Case/Alex Shih

Case opened on 05:01, 12 February 2019 (UTC)

Case suspended on 05:01, 12 February 2019 (UTC)

Case closed on 18:34, 13 February 2020 (UTC)

Watchlist all case (and talk) pages: [/index.php?title=&action=watch Front], [/index.php?title=/Evidence&action=watch Ev.], [/index.php?title=/Workshop&action=watch Wshp.], [/index.php?title=/Proposed_decision&action=watch PD.]

Involved parties

 * , filing party

Statement by GoldenRing
The arbitration committee have recently disclosed that the resignation of Alex Shih from the committee and from his CU and OS permissions was a direct consequence of inappropriate discloses of private information in breach of the WMF's privacy policy, and therefore considered to be "under a cloud" (diff). Members of the committee have stated that the situation doesn't fit the cases where the committee may desysop sua sponte (diff), but owing to the nature of the violations the community lacks the information to determinate whether Alex should retain the trust of the community. Alex is now known to have lied about his reasons for resignation, both at the time (diff) and during his recent self-nomination in the Stewards' elections (diff). We consider this evidence of untrustworthiness, but each have our own view on the significance of this breach of trust.

In responding to this resignation, editors have argued both that the nature of the violations necessarily imply a loss of trustworthiness that is incompatible with access to the tools, and that the violations were related specifically to permissions that Alex has already resigned and that the administrator tools had not been abused.

We therefore request that the committee review the information available to it, both public and private, in relation to Alex's resignation and consider whether his actions, both as a Checkuser and during his resignation, were compatible with the expectations to which administrators are held, specifically as at WP:ADMINACCT and WP:ADMINCOND. We make this request without the assumption that Alex Shih must be desysopped. If the expectations of administrator conduct have been violated to the extent that a desysop is necessary, only ARBCOM is able to evaluate and act upon this; if those expectations have not been violated to that extent, then only a finding to that effect from ARBCOM will enable Alex to resume his normal activities as an administrator.

The above request has been prepared, and is submitted jointly, by, and. I (GoldenRing) will notify all those I have listed as parties, but will allow Vanamonde93 and Boing! said Zebedee to co-sign this request in place of the usual diff of notification (which they have agreed privately to do). I will say personally that I am sad to have to file this request, as I have always previously held Alex in high regard, but that I do feel there are issues of trust that need to be explored and resolved here. My co-signers have their own space below to add their own personal notes.

Statement by Alex Shih
The breaches of CheckUser policies is becoming a red herring in this case, as these breaches have already been reviewed by ArbCom and a conclusion was made back in September 2018 where I offered full explanation for the allegations and I am perfectly fine with these allegations and my full explanation to be made public because I don’t believe anything sensitive was involved. Moving forward, in short summary I interpreted the conclusion at the time as repeated mistakes, regardless of whether or not made in good faith, coupled with mishandling of several mailing list discussion content, that the committee thought I was being indiscreet, an admonishment that I freely accept as I had the mistaken belief over what is confidential information and what is not, and resigned right after.

The focal point here (for the merit of this case) should be based on whether or not I have breached basic policies in the context of Wikipedia, and I firmly believe that was never the case and trust that same conclusion can be made when evaluating the evidence. I do hope ArbCom can make the clear distinguishment here on whether or not I “violated privacy policy”, because that is a very serious accusation, or simply I have made several repeated mistakes in my role as a functionary and arbitrator, as I have described above. The second point seems to be on whether or not “community trust” is retained based on the fact I did not disclose the complete nature in which I resigned ArbCom. I do apologise for those who have been offended by the apparent lack of transparency, but I ask these folks to understand that had I done a full disclosure, I believed at the time it would have been equivalent to the same breach that I have been accused of doing by ArbCom, which is to reveal that a discussion is or has been taking place (despite of not revealing the content of the discussion) without the consensus of the committee.

I have no desire to edit nor participate in this any further. All of my actions were done in the best interest of the encyclopedia and I am not afraid to talk about them (which is why I openly edit under my real name), and I believe my work as an editor and administrator have reflected this. I have long disagreed that arbitrary titles on a website should be taken as seriously as it currently is, so do what you think is in the best interest of the encyclopedia, because I believe that is the only thing we should be considering. Alex Shih (talk) 00:45, 10 February 2019 (UTC)

Statement by Vanamonde93
I endorse the above request, obviously, as co-drafter. Filing this saddens me, but I think this is the only way for this incident to be given some closure. Vanamonde (Talk) 21:46, 9 February 2019 (UTC)

Statement by Boing! said Zebedee
I endorse the case request and the statement above as a co-signatory, and will add some more thoughts here tomorrow morning (UK time). I also echo Vanamonde's sadness. Boing! said Zebedee (talk) 23:01, 9 February 2019 (UTC) The basis of this case is AlexShih's resignation from ArbCom under a cloud, coupled with subsequent events. I accept that ArbCom has the full details while the rest of us do not, and that only ArbCom can judge the extent and nature of the CU violations at the time together with Alex's response. ArbCom, apparently, considered desysop at the time, but part of the reason for not going ahead with it was due to the committee's inability to bring cases of its own (if I'm understanding the recent discussion). So we had a situation where only ArbCom knew about the CU breaches, but only the community could bring a case - but ArbCom were not going to tell the community about it. I think there's something wrong there, and I hope the committee will consider that when they ponder this case. Does there need to be a process for when the knowledge of a policy breach is confidential, but those in the know can't consider a desysop case for technical reasons? The main point I wish to make (which several responding Arbs have already covered below) is Alex's behaviour since his ArbCom resignation. His explanation of his resignation shortly afterwards was false, and I'd say that's a breach of WP:ADMINACCT - which says "Administrators are expected to respond promptly and civilly to queries about their Wikipedia-related conduct and administrative actions and to justify them when needed" (my emphasis), so it's not only about tool use. (Incidentally, I'm disappointed that ArbCom, knowing that Alex's claim was not true, stood by and said nothing at the time.) But, far worse, Alex was dishonest in his run for Steward, again giving a false reason for his ArbCom resignation and failing to disclose that it was due to breaches of CU policy - a power that he would have been handed back had he succeeded in that endeavour. That was deliberate deceit and a serious breach of trust, and someone who is prepared to deceive the community in order to gain advanced tools has no place being an admin. So has Alex actually lost the trust of the community? I think the comments from en.wiki voters in the Stewards election (which would have risen by at least one had he not withdrawn before I was able to offer mine) make that clear. I'll add that until this affair came out, I have had no complaints about Alex as an admin (though I have seen what I think is a deteriorating attitude towards the Wikipedia community, so maybe there's a bit of burnout here?) I also want to put on record my opinion that ArbCom performed poorly in this case last year. Boing! said Zebedee (talk) 06:41, 10 February 2019 (UTC) Just one extra comment (agreeing with others here) that whether dealt with by motion, a shortened case or whatever, there's surely no evidence phase needed. ArbCom obviously already has the evidence of the CU violations, and the only other apparent issue (the dishonest explanations for the ArbCom resignation) is out in the open. Boing! said Zebedee (talk) 13:21, 10 February 2019 (UTC)

Clerk notes

 * This area is used for notes by the clerks (including clerk recusals).


 * Recused GoldenRing (talk) 22:22, 9 February 2019 (UTC)
 * I've notified all parties notified of the motion. --Cameron11598 (Talk) 03:45, 11 February 2019 (UTC)
 * Provided I don't hear anything from members of the committee asking to hold off on implementation, the motion will be enacted 24 hours from the 6th Support. --Cameron11598 (Talk) 20:07, 11 February 2019 (UTC)

Arbitrators' opinions on hearing this matter (8/1/0)
Vote key: (Accept/decline/recuse)
 * Accept. I'd normally wait for additional statements before deciding whether I think a case should be opened, but given the fact that the ArbCom is currently the only group that consider desysop cases and given all the discussion at Wikipedia talk:Arbitration Committee/Noticeboard it's quite clear that a case is needed. GorillaWarfare (talk) 22:09, 9 February 2019 (UTC)
 * When you say, "I have no desire to edit nor participate in this any further", do you mean you have no desire to edit Wikipedia at all or just specifically with regards to this matter? GorillaWarfare (talk) 07:01, 10 February 2019 (UTC)
 * Given that Alex Shih has retired, I think we should suspend the case with the understanding that it will be reopened if he returns to editing, and that he should not use his admin tools until the case is resolved. GorillaWarfare (talk) 20:31, 10 February 2019 (UTC)
 * Decline. As our policies stand, Alex Shih can only be desysopped by ArbCom – who will do so either to resolve a dispute or deal with a gross breach of trust.  This request falls into the second category.  On that basis, I would deny this request and not move to desysop.  A series of serious errors were committed by Alex Shih, but in a role that lends itself to them.  Checkusers are self-taught, loosely audited, and doing a job where committing errors may not change the result.  Alex Shih should have known better, but breached policy in the same way that several checkusers have done since the tool was introduced.  The difference here is that we allowed Alex Shih to carry on making the errors, if not make them in the first place.  Likewise, Alex Shih failed to disclose their many mistakes when resigning; but the resignation statement was pointedly brief and technically accurate.  I am not sure any of us would rush to disgrace ourselves on so public a forum.  Much of this sounds like I am excusing the administrator for their breaches of the checkuser policy; I do not.  However, I do not consider errors in one of Wikipedia's most difficult, obscure roles to cast doubt, in this particular case, on Alex Shih's fitness to serve as an administrator for reasons of character, judgment, or competence.  Although not generally inclined to leniency with administrator misconduct, I feel obliged to disagree with this movement for desysopping.   AGK  &#9632;  22:54, 9 February 2019 (UTC)
 * Addition: all that said, if Alex Shih continues to keep radio silence then we may be obliged to desysop on an interim basis.  AGK  &#9632;  23:02, 9 February 2019 (UTC)
 * Pardon, ? First, another error was discussed on our very noticeboard; and second, as I said without ambiguity, those errors stretch back to the tool's creation in 2006. Your monologue about a current agenda to tighten up on errors is fine, but don't misrepresent my assessment of this case request.   AGK  &#9632;  07:47, 10 February 2019 (UTC)
 * Checkuser lends itself to incorrect use more easily than other tools. Some parts of our Checkuser and Privacy policies appear not to have come naturally to other users, and require rigorous self-discipline from new appointees to the tool.  All this is why we carefully appoint new users and are severe with use that is less than perfect.   AGK  &#9632;  11:05, 10 February 2019 (UTC)


 * Accept. If it were just the CU mistakes conducted while a member of ArbCom then AGK's argument would be acceptable. But as Alex applied for a CU role after he had been made aware he had various weaknesses in that area which made applying for the tool inappropriate, and misled people about the reason he had resigned from ArbCom, we are into a gross breach of trust and a serious lack of judgement. I don't see it possible how in the circumstances the community could trust him right now. This can be resolved quickly by motion. SilkTork (talk) 00:50, 10 February 2019 (UTC)
 * Accept, but perhaps not for the reasons one would expect. I stand by my statement at WT:ACN, and I do not necessarily think that misuse of CheckUser data must lead to desysopping. Alex was and remains, by all accounts, a fine administrator when he does not have access to nonpublic information. At the time I wrote that statement, I expected to vote to decline a case request if one were filed, despite being Alex's harshest critic on the Committee in 2018, easily. Having said that, additional information has come to light that I believe warrants a case. I have been surprised at Alex's lack of response to community concerns on-wiki, including his statements made off-wiki that he does not intend further comment. I have also read 's comments on Alex's talk, and they are persuasive. To the extent there was an attempt to mislead the community while seeking election as a steward, through omission or otherwise, that speaks directly to "gross loss of trust". In that sense, my acceptance of this case is based not on the contents of the Ombudsman Commission report, as bad as they were, because I do not see them as particularly related to his role as an administrator or the community's ability to trust him with administrative tools. My acceptance of this case is based on Alex's conduct since then, which has evaded accountability and thwarted attempts by the community to evaluate whether he continues to hold their trust. I will await Alex's response to this case request. If he does not respond within a few days, I will move to resolve this case by motion. (Striking, as Alex responded in the long interim I was typing my initial comment.) ~ Rob 13 Talk 01:25, 10 February 2019 (UTC)
 * I also emphatically disagree with my colleague's statement above regarding CheckUser "mistakes". When the last group of CheckUsers was brought on, I organized an internal training to explain the role. One of my goals in that training was to put the "fear of God" into them, and I hope I succeeded. CheckUsers simply can't make mistakes, not when it comes to disclosure of non-public information. The lines are very bright, intentionally so. I am aware of no violations of the access to nonpublic information policy by any enwiki CheckUsers during my tenure on the Arbitration Committee other than Alex Shih. If I were aware of any such violations, I would have advocated immediate removal as a CheckUser, even on a first offense, if the violation involved the release of any private data. I am dismayed that such a statement would be made by an arbitrator, since it gives the impression that the Arbitration Committee would not treat violations of the access to nonpublic information policy with the appropriate severity. ~ Rob 13 Talk 01:33, 10 February 2019 (UTC)
 * All fair questions. I suppose I meant to say that Alex was and remains a fine administrator, completely ignoring his access to nonpublic information from start to finish. I know of no similar cases of lack of accountability disconnected from his role on the Arbitration Committee, including his access to CheckUser data. That is all I was saying. As for the latter questions, I think the Committee did treat the situation with the appropriate severity once it became known. Keep in mind that we cannot disclose internal deliberations related to Alex before his resignation, so I'm not at liberty to say exactly what we were doing at the time. I can say that my own position was that Alex should be removed from the Committee if he did not resign, and I was taking steps toward that. I will also make a personal commitment that auditing of CheckUsers on enwiki will improve. I have already said I will push to restore regular audits conducted by the Arbitration Committee or a subset thereof, but if for whatever reason that does not gain internal support, I will personally audit our local CheckUsers, including sitting arbitrators, and forward any problematic checks to the Ombudsman Commission for review. We have been too lax on auditing of enwiki CheckUsers. I am making a commitment that practice will change, even if it means doing the work entirely myself. ~ Rob 13 Talk 01:48, 10 February 2019 (UTC)
 * I take it you do not see value in a full case on this matter, in favor of dispatching this one way or the other by motion? ~ Rob 13 Talk 04:29, 10 February 2019 (UTC)
 * It's unclear to me how you believe I have misrepresented anything you said, mostly because I haven't particularly represented it at all. I did not summarize your statement in any way. I said I disagreed with its content and went on to reassure the community that I personally treat CheckUser mistakes as serious and have not seen any large number of them being made. I have no idea about what happened with Drmies in 2012; indeed, I wasn't even an editor then. But I seriously doubt it could be anything approaching the contents of the Ombudsman Commission report, which detailed repeated instances of bright-line violations of the access to nonpublic information policy, which continued even after they were discussed with Alex. ~ Rob 13 Talk 16:11, 10 February 2019 (UTC)
 * Please see my statement above, which I believe answers your point. I am happy to answer any other specific questions you have to the farthest extent I am able to under the access to nonpublic information policy. ~ Rob 13 Talk 16:35, 10 February 2019 (UTC)
 * Accept As I've said in the previous admin cases with which we've been presented, I have a low bar for acceptance. I agree with Rob, though, that Alex has been a competent administrator – hell, I couldn't work the DYK queue like he does, and the closures he makes seem to be just fine. I was annoyed that he resigned from the committee in the manner he did, with such a touching yet completely misleading statement, but fine, whatever. It allowed him to leave with grace, and the OC complaint was in the hands of the commission, and they were going to do what they were going to do. However, I was stunned that he tried to deceive the community by saying he resigned from the committee due to 'bureaucratic differences' (I think that's the phrase he used, if not, sorry). That couldn't stand, and that's why we released the statement we did. Since then, he's essentially called us liars, but I assure you the data is real, the breaches are real and serious, and if I could share it with you all, I would. But I can't, and you all elected me and the rest of us to our positions to do this kind of ugly but necessary work. I don't know if I'll vote to desysop, but the arguments to accept are persuasive and I do not like what I've seen from Alex in the last two weeks. Katietalk 02:06, 10 February 2019 (UTC)
 * Accept. Alex's entry into the Stewards election raises fresh issues and we'll all benefit from having a line drawn under this. We should aim to resolve this quickly by motion; given that much of the discussion will be private and we already have most of the evidence, a full case would be pantomime. If anyone wants to us to consider additional information, they probably ought to submit it now, either as a statement or by email. –&#8239;Joe (talk) 02:30, 10 February 2019 (UTC)
 * Accept. Alex volunteered to do a notoriously difficult job and was elected based on his record as an editor and admin. It turned out that there were some aspects of the job that were not a good fit. Well, that's probably true for most of us (oh god, please don't ask me about the fine points of DS procedure) but the particular aspects Alex was unsuccessful with are pretty core elements of the role. When told he was making too many mistakes, he removed himself from the situations where mistakes could be made and went back to doing the things he'd previously been successful at. So far, so good. I was not in favor of any further action at the time of the original discussion in August-September. (I do feel like a bit of a chump now for having said at the time that surely the ombudsmen wouldn't take too long...) I've never been bothered by his descriptions of his departure in on-wiki or off-site conversations - after all, resigning from things "to spend more time with your family" and so on is a well-established tradition of resolving messy situations without too much public fuss. But then going on to run for another position of trust, with access to many of the same types of private information he had access to before, with arguably even more responsibility and less oversight, knowing that there were outstanding concerns about his work in exactly that area - well, that's just really poor judgment. (Alex, if you were concerned that disclosing the background might itself be a privacy violation, you could have asked us first?) We certainly don't need a lengthy process here, but I can see why a re-review is in order in light of recent events. Opabinia regalis (talk) 06:36, 10 February 2019 (UTC)
 * Accept This case request comes amid a recent series of events and most likely others soon to come. I cannot think of too many cases where ArbCom was requested to decide on a matter related to ADMINACCT without there being a direct misuse of the sysop tools or sock puppetry. The issue of repeated 'poor judgement' and 'lost the trust of the community' has been regularly discussed within the community and this case will almost certainly test those aspects of the admin policy. The discussion at WT:AC/N and the statements made here sufficiently outlines enough concern for a case. Mkdw  talk 17:36, 10 February 2019 (UTC)
 * Not that I am aware. Mkdw  talk 17:36, 10 February 2019 (UTC)


 * Accept This needs to be looked at by the committee as a whole. RickinBaltimore (talk) 18:26, 10 February 2019 (UTC)

Temporary Injunction: Alex Shih
(1) Proposed:




 * Support
 * Proposed. This is an injunction, not a motion, and as a clerical matter the accepted case should still be opened.  No need to create anything other than the case's front page.  As I recall, the difference here from  is that SW was uncontactable prior to the matter reaching arbitration, rather than retiring after arbitration started.  In other words, SW had not been afforded the chance to respond.   AGK  &#9632;  23:07, 10 February 2019 (UTC)
 * I find it interesting that Alex so believes in his position that he immediately retires when the case is opened. Nonetheless, he's entitled to process, and if he returns, he'll get it. Meanwhile, we can move on and at least temporarily close this sorry chapter. Katietalk 23:18, 10 February 2019 (UTC)
 * Now second choice. Katietalk 02:53, 11 February 2019 (UTC)
 * Second choice - I prefer the option of only re-opening the case if Alex requests it. SilkTork (talk) 01:31, 11 February 2019 (UTC)


 * Oppose
 * Procedurally, an injunction cannot be issued until a case is opened, and this case is not yet opened. We should instead open and suspend this case by motion. I will propose something in a moment. ~ Rob 13 Talk 00:17, 11 February 2019 (UTC)
 * Not sure this needed anything other than removing the word "injunction", but anyway, support the motion version. Opabinia regalis (talk) 04:54, 11 February 2019 (UTC)
 * I'm not fussed about the bureaucracy around opening and shutting cases, but I prefer the second motion in that it allows Alex to return as soon as he likes without prejudice. This motion requires a case, and since both motions start with a temporary desysop, I don't see the need for a case unless Alex objects. WormTT(talk) 09:06, 11 February 2019 (UTC)
 * As I prefer the option below as a better overall decision. RickinBaltimore (talk) 11:14, 11 February 2019 (UTC)
 * Abstain


 * Comments

Motion to Suspend

 * Enacted --Cameron11598 (Talk) 04:52, 12 February 2019 (UTC)
 * Support
 * ~ Rob 13 Talk 00:31, 11 February 2019 (UTC)
 * First choice as this gives the option to Alex of re-opening the case, and so allows him to quietly return to Wikipedia without having to face a trial. SilkTork (talk) 01:31, 11 February 2019 (UTC)
 * First choice. Katietalk 02:52, 11 February 2019 (UTC)
 * GorillaWarfare (talk) 03:12, 11 February 2019 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 03:31, 11 February 2019 (UTC)
 * &spades;PMC&spades; (talk) 03:50, 11 February 2019 (UTC)
 * Opabinia regalis (talk) 04:52, 11 February 2019 (UTC)
 * I'm saddened to see how far and fast this has progressed, when I stopped on Wikipedia on Friday there was a discussion about Alex on the WP:ACN, and this morning there's an accepted case, Alex has retired and we have a motion to suspend and desysop. My concerns have been handled in this motion - it allows Alex to return if and when he wants (and that's something I'd like to see), with or without a case - completely at his choice for the first year. It addresses the community concerns or "loss of trust", by imposing the desysop. Oh, and just so it is on the record, I do believe Alex was truthful in his succinct statements about retiring - notwithstanding the ommissions, checkuser behaviour, and poor judgement running for steward - but he and I discussed and expressed infuriation about the bureaucracy of Arbcom. Further, I have seen no malice in Alex's actions - I hope this motion can be the end of it and Alex can have the option of returning to content editing. <b style="text-shadow:0 -1px #DDD,1px 0 #DDD,0 1px #DDD,-1px 0 #DDD; color:#000;">Worm</b>TT(<b style="color:#060;">talk</b>) 09:04, 11 February 2019 (UTC)
 * I'd have preferred to get this over with now and, judging by the sentiment of our comments above, Alex likely would have avoided a desysop that way. But it would be unfair to proceed in his absence. –&#8239;Joe (talk) 10:32, 11 February 2019 (UTC)
 * RickinBaltimore (talk) 11:14, 11 February 2019 (UTC)
 * Oppose
 * Oppose


 * Abstain


 * Comments
 * Re-framing as a motion. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 00:31, 11 February 2019 (UTC)

= Final decision =

Enforcement log
Any block, restriction, ban, or sanction performed under the authorisation of a remedy for this case must be logged at Arbitration enforcement log, not here.