Wikipedia talk:Arbitration Committee/Noticeboard/Archive 35
This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 30 | ← | Archive 33 | Archive 34 | Archive 35 | Archive 36 | Archive 37 | → | Archive 40 |
Arbitration motion regarding discretionary sanctions
We need more check users after Mike V's inactivity
When I posted this at WP:AN, one administrator asked me to ask ARBCOM. --Marvellous Spider-Man 04:08, 10 June 2017 (UTC)
- Thank you Marvellous Spider-Man. This is definitely something that's on our radar. We are nearing the completion of a review of functionary activity levels which should provide us with some information in looking ahead to CUOS2017 and when that might need to occur. Mkdw talk 04:29, 10 June 2017 (UTC)
- There's a persistent lack of Clerks and Admins at SPI; having more Checkusers won't necessarily alleviate that backlog. Note there is only one endorsed Checkuser request at the time of me writing this; the remainder require Clerk or Admin assessment.--Jezebel's Ponyobons mots 18:23, 15 June 2017 (UTC)
- What Ponyo said. Behavioral evaluation is the bulk of SPI work, and we desperately need more attention there. Katietalk 18:31, 15 June 2017 (UTC)
- Also, the Checkuser statistics since Mike V's departure have not changed. The total number of checks run prior to Mike V's departure were 4,287 (Sept. 2016), 4,023 (Oct. 2016) and 4,344 (Nov. 2016). The total number of checks run in the last three months were 4,309 (Mar. 2017), 4,327 (Apr. 2017) and 4,382 (May 2017).--Jezebel's Ponyobons mots 18:34, 15 June 2017 (UTC)
- Admin can't endorse for CU, only clerks can. Unless that has changed since I was a clerk. Dennis Brown - 2¢ 22:52, 15 June 2017 (UTC)
- I believe you're right, but we are in desperate need of admins to review the cases in "CU completed", "Open", "CU declined" and "Awaiting admin" status, which form the bulk of the SPI cases. It's a pretty draining and soul-crushing endeavour, as you know, and the burn-out rate is high, so I can understand why there is a perpetual shortage of willing and capable persons. --Jezebel's Ponyobons mots 23:12, 15 June 2017 (UTC)
- Last time I overstepped and did a small clerk thing in an obvious case, I got my head bit off unnecessarily by a CU. It was just second nature since I had the SPI scripts in Javascript. I stay away completely now. Dennis Brown - 2¢ 16:56, 16 June 2017 (UTC)
- I remember. I can only speak for myself, but if you ever want to dip your toe back in to help out with some of the admin stuff, I'd certainly appreciate your efforts (though I understand completely why you don't have any desire to return to the SPI arena).--Jezebel's Ponyobons mots 17:04, 16 June 2017 (UTC)
- What Ponyo said. —DoRD (talk) 20:15, 17 June 2017 (UTC)
- Last time I overstepped and did a small clerk thing in an obvious case, I got my head bit off unnecessarily by a CU. It was just second nature since I had the SPI scripts in Javascript. I stay away completely now. Dennis Brown - 2¢ 16:56, 16 June 2017 (UTC)
- I believe you're right, but we are in desperate need of admins to review the cases in "CU completed", "Open", "CU declined" and "Awaiting admin" status, which form the bulk of the SPI cases. It's a pretty draining and soul-crushing endeavour, as you know, and the burn-out rate is high, so I can understand why there is a perpetual shortage of willing and capable persons. --Jezebel's Ponyobons mots 23:12, 15 June 2017 (UTC)
- Admin can't endorse for CU, only clerks can. Unless that has changed since I was a clerk. Dennis Brown - 2¢ 22:52, 15 June 2017 (UTC)
- Also, the Checkuser statistics since Mike V's departure have not changed. The total number of checks run prior to Mike V's departure were 4,287 (Sept. 2016), 4,023 (Oct. 2016) and 4,344 (Nov. 2016). The total number of checks run in the last three months were 4,309 (Mar. 2017), 4,327 (Apr. 2017) and 4,382 (May 2017).--Jezebel's Ponyobons mots 18:34, 15 June 2017 (UTC)
- I agree with Ponyo. More admins and trusted users good at spotting similarities are needed to deal with requests that do not need CUs, or where CUs have already run a check, in my opinion... Those are, IMHO, the areas where we do tend to have backlogs... Salvio Let's talk about it! 17:13, 16 June 2017 (UTC)
- Perhaps, in addition, something could be added to the Administrators' Newsletter requesting for additional support and help. Mkdw talk 18:41, 16 June 2017 (UTC)
- Agree with Ponyo and Salvio with a bit more food for thought. The majority of current clerk candidates are not being bold enough. SPI is a place where you learn as you work and they won't gain enough experience if they aren't engaging cases; this is how they will learn who various sockmasters are as well as what to do. I see a minority that are SPI regulars. I vetted the list a few weeks ago and pondered leaving a post along the lines of "please engage in cases so that the clerks and checkusers have more to gauge when reviewing your applications". Some of them need to wake up a bit. My words a few years ago apply to non-admin clerks as much as DoRD's words apply to non-clerk admins, a view that I endorse. Perhaps the checkuser team should leave a message addressing the candidates.
- Perhaps, in addition, something could be added to the Administrators' Newsletter requesting for additional support and help. Mkdw talk 18:41, 16 June 2017 (UTC)
- What Ponyo said. Behavioral evaluation is the bulk of SPI work, and we desperately need more attention there. Katietalk 18:31, 15 June 2017 (UTC)
- There's a persistent lack of Clerks and Admins at SPI; having more Checkusers won't necessarily alleviate that backlog. Note there is only one endorsed Checkuser request at the time of me writing this; the remainder require Clerk or Admin assessment.--Jezebel's Ponyobons mots 18:23, 15 June 2017 (UTC)
- I would also add that some candidates, past and present seem to slow down as if awaiting permission to proceed beyond the gate after they add their name to the list. I would suggest that the placard at the top of this thread be reworded to encourage them to get busy working cases.
— Berean Hunter (talk) 20:45, 16 June 2017 (UTC)
- I would also add that some candidates, past and present seem to slow down as if awaiting permission to proceed beyond the gate after they add their name to the list. I would suggest that the placard at the top of this thread be reworded to encourage them to get busy working cases.
Motion: Activity
Just confirming that I was made aware of this prior to this action taking place, and have no issue with resigning checkuser per lack of use. Cheers! Reaper Eternal (talk) 19:59, 19 June 2017 (UTC)
- Thank you for your service as a CU, and SPI clerk before that, Reaper Eternal - I always enjoyed working with you. I'd like to thank Elockid and FloNight for their service, as well. —DoRD (talk) 02:36, 20 June 2017 (UTC)
GoldenRing appointed trainee clerk
- One of the few admins that are willing to take decisions that matter. Congratulations! Capitals00 (talk) 17:01, 19 June 2017 (UTC)
- No worries from me. Dennis Brown - 2¢ 17:14, 19 June 2017 (UTC)
- Good luck, GoldenRing! Lord Roem ~ (talk) 17:26, 19 June 2017 (UTC)
- Just remember that behind every user name is an actual human being. The Rambling Man (talk) 19:27, 20 June 2017 (UTC)
Recap of RfC discussion regarding WP:Outing and WMF essay
The RFC discussion regarding WP:OUTING and WMF essay about paid editing and outing (see more at the ArbCom noticeboard archives) is now archived. Milieus #3 and #4 received substantial support; so did concrete proposal #1.
Milieu 3: "
The balancing COI and privacy/outing means that the only option is that people investigating COI must submit information in private to the relevant people. Currently this is the arbitration committee and/or the WMF, but other bodies could be considered if there is consensus for this."Closing rationale:
There is consensus for the proposal with the obvious caveat, that this approach needs a lot more details and clarification.Many have clarified that other bodies shall only refer to editors who have been vetted by the community to handle sensitive and personally identifying information.There has been concerns about the use of the word only as it seems to nullify on-wiki processes based on CU and behaviorial evidence.
"
We need to balance privacy provided to those editing in good faith against the requirements of addressing undisclosed paid promotional editing. To do so can be achieved with a private investigation with some release of results publicly to help with the detection of further related accounts. These details may include the name of the Wikipedia editing company with which the account is associated (such as for example the connections drawn here)"Closing rationale:
There is consensus for the above proposal, with a condition that the proposal must be clarified to remove vaugeness, and that any information released must be limited to "employer, client, and affiliation".
More specifically, the information that is to be clarified is:
- Who is doing the investigating? (this looks like it's covered by Milieu 3)
- What information is to be released? The proposer has stated in the discussion below (and other editors agreed) that the information that is released is to be limited to "employer, client, and affiliation". This renders the argument of wp:outing invalid, which really was the only argument brought up on the oppose side.
--George Ho (talk) 23:57, 21 June 2017 (UTC); expanded, 00:03, 22 June 2017 (UTC)
- User:George Ho many thanks for the recap. Doc James (talk · contribs · email) 05:02, 22 June 2017 (UTC)
- You're welcome. :) I've given more copies of the recap at Meta-wiki's Wikimedia Forum and the Board of Trustees' Noticeboard. --George Ho (talk) 05:05, 22 June 2017 (UTC)
- User:George Ho many thanks for the recap. Doc James (talk · contribs · email) 05:02, 22 June 2017 (UTC)
Question about changing header wording at Arbitration Enforcement
Is it appropriate to change header level names at Arbitration Enforcement to say "Defendant" and refer to other Wikipedia users as "Prosecutor" and "special prosecutor", as was done, here DIFF? Sagecandor (talk) 17:03, 22 June 2017 (UTC)
- AE is run by admins, not ArbCom or clerks, so I'm not sure why this question was posted here. --Floquenbeam (talk) 17:18, 22 June 2017 (UTC)
- I don't know why they posted here, but it's aperfectly terrible idea. AE isn't a court of law. Beyond My Ken (talk) 17:20, 22 June 2017 (UTC)
- Sorry, I thought it was a clerks issue. My bad. But thanks for your attention. Sagecandor (talk) 17:33, 22 June 2017 (UTC)
- Well, it's a subpage of Wikipedia:Arbitration, and the entire Arb namespace is usually managed by ArbCom or their clerks directly, so it's not unreasonable to ask ArbCom or their clerks about what constitutes an acceptable format, I guess? But yeah, this could really have been asked on AE's talk page. ☺ · Salvidrim! · ✉ 17:51, 22 June 2017 (UTC)
- Alright, I know for next time, thanks. Sagecandor (talk) 17:56, 22 June 2017 (UTC)
- I don't know why they posted here, but it's aperfectly terrible idea. AE isn't a court of law. Beyond My Ken (talk) 17:20, 22 June 2017 (UTC)
Guido den Broeder
You may be aware by now that User:Guido den Broeder, who was unbanned a second time by ArbCom in April of this year, has been just rebanned by community consensus after a WP:AN discussion (Wikipedia:Administrators' noticeboard#User Roadcreature / Guido den Broeder). Some concern and/or surprise about the unban and lack of communication surrounding it has been expressed there. I also note that Guido den Broeder still denies being the same as the editors blocked in 2015 as being his sockpuppets, and that he claims that you had taken this into consideration at the time of the unban. In general, when people try to get some restriction lifted at AN, denial of previous problems is a major red flag, but I see very little understanding of why he was banned in the first place ("In conclusion, I have yet to see a single diff of any wrongdoing on my part. The same was true in 2009. People just jumped on the bandwagon then, as you are doing now."). If Guido den Broeder told you the same things he told at the AN discussion, then why was an unban granted (and who voted in favour of it)? Fram (talk) 07:12, 20 June 2017 (UTC)
- Yes, I would be interested in knowing this too. Although ArbCom was a venue for overturning a community ban until recently, it would stand to reason then when a community ban was being addressed, ArbCom should have communicated with the community when it overturned, and the ban. This was not done, which I feel is a fairly serious oversight on ArbCom's part, one which really ought to be explained. This is especially the case since the community saw fit to re-establish the ban almost immediately that it became aware of it. Such a disconnect between ArbCom and the general community is unusual, and, I would hope, will not be a frequent occurrence. Beyond My Ken (talk) 08:16, 20 June 2017 (UTC)
- Thats not really their fault. His current ban (which was lifted) was an arbcom ban placed by arbcom. His original community ban was lifted in 2009 by arbcom, (before they rebanned him) because the process at the time was that community bans were appealable to arbcom. Its unfortunate, but its the way the UNBAN policy has evolved. Only in death does duty end (talk) 08:31, 20 June 2017 (UTC)
- OK, thanks for the info, I appreciate your straightening that out for me. Beyond My Ken (talk) 19:16, 20 June 2017 (UTC)
- This thread could have ended here, because the answer is "what OID said". Opabinia regalis (talk) 09:37, 25 June 2017 (UTC)
- If you ignore the original questions, yes... My original post ended with "If Guido den Broeder told you the same things he told at the AN discussion, then why was an unban granted (and who voted in favour of it)?", to which "what OID said" was not an answer (nor an attempt to answer these questions, they were a reply to BMK). If that's the reading skill you applied to this unban request, I understand that it was granted. Fram (talk) 08:24, 26 June 2017 (UTC)
- Thats not really their fault. His current ban (which was lifted) was an arbcom ban placed by arbcom. His original community ban was lifted in 2009 by arbcom, (before they rebanned him) because the process at the time was that community bans were appealable to arbcom. Its unfortunate, but its the way the UNBAN policy has evolved. Only in death does duty end (talk) 08:31, 20 June 2017 (UTC)
I was wondering why ArbCom members (who are active elsewhere, so not a collective holiday in the sun) didn't respons here so far, but I notice that we have WP:ADMINACCT but not WP:ARBCOMACCT. That you have the right to grant unbans, doesn't mean that you don't have to be accountable for them. Fram (talk) 07:12, 22 June 2017 (UTC)
- Opabinia regalis did the unblock [1] and left this message [2], so at a minimum, WP:ADMINACCT applies to the unblock because it requires the admin tools. The community would expect an explanation from any admin, without disclosing private details. BTW, I didn't vote or close at that discussion, so uninvolved here. Dennis Brown - 2¢ 14:39, 22 June 2017 (UTC)
- I mean, if we get technical, the admin tools were used to enforce an ArbCom decision, so that's the only WP:ADMINACCT needed. As for WP:ARBACCT, I don't think there is any mechanism to compel a response from ArbCom about their privately-discussed decisions (other than your vote in the next elections). ☺ · Salvidrim! · ✉ 15:03, 22 June 2017 (UTC)
- No, but as a matter of form, it is good to explain what they can, parts that aren't privacy related. It could be as little as a paragraph or two, and doesn't require minute detail. It is a curious situation and it is reasonable to ask for clarification. The question was asked in good faith, and is reasonable. It would be unreasonable to flatly ignore the request. Dennis Brown - 2¢ 15:36, 22 June 2017 (UTC)
- Of course it would be "best form" for ArbCom to respond, I'm just pointing out they are not bound to do so. Their only "statement" on the matter will perhaps remain the unblock notice. ☺ · Salvidrim! · ✉ 16:04, 22 June 2017 (UTC)
- Everyone and their dog demands an explanation about why their request was declined, or why someone else's request was granted or declined. I would not even put Fram's request and 48-hour follow up for an explanation within our queue of the next ten requests by order of being received. Any expectations that members on the committee should not edit unless all requests have been answered is unrealistic. Additionally, the presumption that any request not answered within 48 hours automatically equates to being "ignored" or skirting accountability is unfounded. The fact that a nearly identical inquiry was asked and answered in April 24, 2017, when the unban occurred, provides contextual background and possible explanation on how it was prioritized in terms of urgency.
- For myself, I spend time everyday reviewing ArbCom material. In regards to the sentiment that I (or other members of committee) shouldn't be editing unless there are no outstanding requests or on a "collective holiday in the sun", I find disgusting and in opposition to the spirit and purpose of Wikipedia. Yesterday, I took some time to clean up a good article I wrote during my own personal time; outside of the time I already set aside for ArbCom (acknowledging that I signed up for it), and other things in my daily life such as work, family, and so forth. I in no way feel any regret for spending some of my own time doing that when there were still some tasks left in the queue. It is for these reasons, I have to disagree that this request was one hundred percent asked in good faith given the unpleasant assignment of expectations and accusations that were placed alongside it. Mkdw talk 22:05, 22 June 2017 (UTC)
- Perhaps if everyone wasn't so unfailingly snarky and unpleasant every single time when demanding a response, they'd respond quicker? I mean, I guess this looks like sucking up so I get leniency in my own inevitable ArbCom desysop case, but Jesus, it's always like that. That's been my experience in real life: if you're snotty to someone, they're less likely to respond to you really fast. Why should it be different here? --Floquenbeam (talk) 15:57, 22 June 2017 (UTC)
- I have made something like 25 edits this month, because I'm extremely busy in real life (in a good way!). I haven't even read the re-ban thread yet (though surely, as the unblocking admin, it would have been appropriate to at least notify me...) In the meantime I recommend mulling over the answer you already got two months ago ;) Opabinia regalis (talk) 16:32, 22 June 2017 (UTC)
- I was the closer of the latest thread - I remember GvB from his original incarnation and the disruption he caused then. Of course there may have been some good reasons why he was unblocked, despite never having (publicly) admitted to any of his previous failings, but I am surprised this was not more widely advertised; because perhaps, just perhaps, he could have been steered back onto a good course before the latest spree of nonsense occurred. Of course, by the time most people were aware, it was too late. Black Kite (talk) 16:38, 22 June 2017 (UTC)
- I wasn't a voting member on the request but I have looked over the request and responses. The recommendation that notices be placed here at ACN for similar unbans has been made and likely will occur in the future. Other unbans granted by the committee have been posted here, especially ones that denote restrictions, so the practice of doing so should continue for consistency and for other benefits as well. I do see at the time that the editor had some restrictions placed on them: a one account restriction and a topic ban, which to me seems to contemplate the past socking findings and their areas of disruption. It was logged on their user talk page and at ER. I don't see their recent edits as defensible and it's unfortunate that things went the way they did. Unbans are particularly challenging because things like the standard offer and second or third chances, especially when an amount of time has elapsed, must rely on a certain amount of good faith and the word of the editor to return to productive editing. Sometimes they work out and sometimes they don't. Mkdw talk 16:58, 22 June 2017 (UTC)
- I wanna make it clear that I absolutely don't fault the Committee for considering the appeal and working with the user -- in fact, I strongly commend ArbCom for its openness to rehabilitation, of which I am a strong proponent. Sometimes it works, sometimes it doesn't, and most times you never know for sure until you try. We can argue the merit of this particular case (or its context w/r/t to annoucement or community discussion or whatever) all day long but not of the principle that rehabilitation attempts are a good thing even when the results sometimes sour into disappointment. ☺ · Salvidrim! · ✉ 17:46, 22 June 2017 (UTC)
- And I want to make it absoluetly clear that while I absolutely don't fault the Committee for considering the appeal, I absolutely do fault them for not working with the community on this one, and for granting the unblock despite the clear evidence that the chances of this working out were next to nil, and the chances of this being a waste of time and a shitfest were very high. ArbCom supposedly is chosen for their ability to thoroughly consider tough situations, but I see zero evidence of them making a careful deliberation here and looking at the situation with a critical mind. "We unban because we can and let the community deal with it afterwards" is not what ArbCom has been installed for. Fram (talk) 07:19, 23 June 2017 (UTC)
- I wasn't a voting member on the request but I have looked over the request and responses. The recommendation that notices be placed here at ACN for similar unbans has been made and likely will occur in the future. Other unbans granted by the committee have been posted here, especially ones that denote restrictions, so the practice of doing so should continue for consistency and for other benefits as well. I do see at the time that the editor had some restrictions placed on them: a one account restriction and a topic ban, which to me seems to contemplate the past socking findings and their areas of disruption. It was logged on their user talk page and at ER. I don't see their recent edits as defensible and it's unfortunate that things went the way they did. Unbans are particularly challenging because things like the standard offer and second or third chances, especially when an amount of time has elapsed, must rely on a certain amount of good faith and the word of the editor to return to productive editing. Sometimes they work out and sometimes they don't. Mkdw talk 16:58, 22 June 2017 (UTC)
- I also wasn't active in the unblock request, so I'll leave that to someone with more information than I have, but out of curiosity: is posting at ACN the best place for you all when we grant an unban request? I agree that we should be more clear in communicating when we've unbanned someone, and figured while we're discussing it I should make sure that there isn't a more useful area to post (or cross-post) this kind of thing. GorillaWarfare (talk) 17:51, 23 June 2017 (UTC)
- ArbCom decisions usually have a post on WP:AC/N detailing the outcome (and sometimes who votes/recused/inactive, etc?), with a "discuss here" link to WT:AC/N and a cross-post on AN; and past unbans have been "announced" the same way. ☺ · Salvidrim! · ✉ 18:01, 23 June 2017 (UTC)
- Right, I just wanted to check that that is actually the best practice or if there would be a more helpful place to post/cross-post. GorillaWarfare (talk) 18:29, 23 June 2017 (UTC)
- That sounds reasonable to me GorillaWarfare. I think it is impossible for Arbs to always know which are going to be controversial and which aren't, but announcing them at ACN makes it transparent, as some might care to monitor their progress. Knowing the basics is fine, knowing who participated would be nice but not sure we can force that issue. We don't need a blow by blow as to deliberations, but in the event Arb knows it will be controversial, a bit more info would be helpful. This may actually prevent some surprise problems, and remind us that Arb is a part of the community. Dennis Brown - 2¢ 18:11, 23 June 2017 (UTC)
- It's true that unbans/unblocks have been a bit scattered - for awhile there was a log at Wikipedia talk:Arbitration Committee/Ban appeals, but it fell into disuse for reasons I don't remember. I believe I've always logged unbans that were associated with specific cases, but haven't been consistent about other stuff. Opabinia regalis (talk) 09:35, 25 June 2017 (UTC)
Reply
- +1 to everything Mkdw said above. It is not really reasonable to start getting agitated because nobody has replied to a non-urgent question within 48 hours of its asking. Especially when the previous iteration of the same question was already answered in this thread on my talk page from April. (Maybe that was just a forgettable conversation - unless I missed something, Fram didn't mention it here or at AN.) And given that said AN thread was open for under 24 hours and nobody except the soon-to-be-banned subject of the thread troubled themselves to notify the admin who unblocked, the carrying-on here about communication and speed of response is... well, as my mother would say, "that's a bit rich, isn't it?"
- As a matter of philosophy, I agree with Salvidrim about the value of second chances. People get older, mellow out, develop new interests and lose old ones.... in considering an unban/unblock/other type of appeal request, I tend to focus on the practical aspects of how the editor wants to contribute in the future. I almost always ask appellants to describe specifically how they plan to contribute and what they intend to do to avoid previous conflicts and trouble areas. I almost never ask someone to "admit their failings", publicly or otherwise. Sometimes their plans work out, sometimes they don't. FWIW, I had a look through my archives and since I've been on the committee, I believe we've unbanned eight editors (excluding BASC residuals in 2016, and routine CU/OS block appeals), of which two were subsequently re-banned.
- I do want to return to the point I raised in the first paragraph about the AN thread re-banning Guido. I'm not going to defend his editing, but it makes me cringe to see a poll of ten people that lasted about 13 hours uncritically described as a decision made by "the community" (in bold italics, no less!). Yes, everything on Wikipedia involves self-selected samples, and yes, we're not a bureaucracy, but I really don't like seeing those corners getting cut in cases where there's no particular urgency. An editor who's made a couple hundred edits in two months on a small handful of low-profile topics is not creating an emergency, even if they're being disruptive and unhelpful. For one thing, I have a self-interested reason to object, because (and I'm speaking in general, not about Guido here) editors who are on the receiving end of this kind of treatment then tend to email arbcom about how unfair it all is, procedures weren't followed, I didn't get my day in court, I'm appealing to Jimbo, I'm alerting the press, I'm suing, I'm telling the UN about this gross violation of my human rights, etc etc etc. More importantly, decision-making on the dramaboards is highly inconsistent and rarely critically evaluated - they don't get that name for shits and giggles - and we shouldn't be exacerbating those problems where the situation doesn't warrant it. Opabinia regalis (talk) 05:27, 23 June 2017 (UTC)
- I closed the thread at that point because it was obviously a simple open and shut case, and would inevitably have resulted in a simple pile-on of votes, which is a both a waste of editors time and unfair to GvB as well. Are you saying that the editors who posted there were wrong? If so, why? I'm a little disturbed by your claim that "decision making on the drama boards is highly inconsistent" - we're here because people think that decision making by ArbCom is highly inconsistent as well... Black Kite (talk) 00:28, 24 June 2017 (UTC)
- 24 hours is a good minimum IMO, unless there's something actually urgent happening (and someone making a bunch of comments and being annoying is not urgent).
- I think that all levels of Wikipedia dispute resolution suffer from inconsistency and lack of systematic review. There is remarkably little feedback on decision-making processes about what was and wasn't effective in the longer term. A few weeks ago I tried to review a small subset of arbcom remedies - recommendations for community RfCs - and posted it here, because I was prompted by an earlier thread there. Admittedly not the most visible place, but it got no reaction. Too bad. I keep meaning to follow up with other types of remedies, which are much easier to sort out from fairly structured case pages than the mess at AN/ANI. Opabinia regalis (talk) 09:35, 25 June 2017 (UTC)
- After reading what you said here, I took a look (for the first time) at what you posted that had gotten no reaction. To my eyes, a lot of it looks like situations where a decision said something like "the community is encouraged to..." and in some cases the community may have blown it off. I tend to think that, unless ArbCom wants to start mandating RfCs (obviously not likely), that's what you are stuck with, with a volunteer community. But if there's an urgently felt need for an RfC, someone who feels that need probably will start it. So maybe that's information that the community no longer felt that hot and bothered, once the case was decided. --Tryptofish (talk) 22:44, 25 June 2017 (UTC)
- Well, it's pretty easy to semi-automatically sort out the remedies, but not so easy to find the RfCs, so mostly I was hoping someone knew of some that I missed. But yeah, that's what I wanted to know - the extent to which these kinds of remedies are actually useful, or whether there are specific types of cases more likely to get follow-up. Opabinia regalis (talk) 04:48, 26 June 2017 (UTC)
- After reading what you said here, I took a look (for the first time) at what you posted that had gotten no reaction. To my eyes, a lot of it looks like situations where a decision said something like "the community is encouraged to..." and in some cases the community may have blown it off. I tend to think that, unless ArbCom wants to start mandating RfCs (obviously not likely), that's what you are stuck with, with a volunteer community. But if there's an urgently felt need for an RfC, someone who feels that need probably will start it. So maybe that's information that the community no longer felt that hot and bothered, once the case was decided. --Tryptofish (talk) 22:44, 25 June 2017 (UTC)
- I closed the thread at that point because it was obviously a simple open and shut case, and would inevitably have resulted in a simple pile-on of votes, which is a both a waste of editors time and unfair to GvB as well. Are you saying that the editors who posted there were wrong? If so, why? I'm a little disturbed by your claim that "decision making on the drama boards is highly inconsistent" - we're here because people think that decision making by ArbCom is highly inconsistent as well... Black Kite (talk) 00:28, 24 June 2017 (UTC)
TLDR (long post by myself removed), simple questions:
- Who in Arbcom voted for or against the unban (plus recusals, inactive, ...)
- Did Guido den Broeder deny the socking?
- Were you aware of his COI editing problems?
- Did he acknowledge any problem at all with his pre-ban editing?
- Did he indicate which topics he wanted to edit?
Shouldn't be too hard... Fram (talk) 07:26, 23 June 2017 (UTC)
- @Ophabinia Regalis--
but it makes me cringe to see a poll of ten people....as a decision made by "the community"
--Contact the WMF. They will more than gladly accept you on the pay-rolls! Usually, this has been the foundation's standard refuge in denying any result arising out of an en-wiki consensus!Also, I must have blatantly mis-read the snowball clause.And yeah, while I am not seeking any time-frame as such (it's perfectly understandable that many editors are busy in their real life) but I am necessitating my support for proper answer to Fram's questions.Cheers!Winged Blades Godric 12:32, 23 June 2017 (UTC) - I asked on the mailing list if anyone minds that their support is posted... obviously I supported, but you presumably guessed that. The rest of this is backseat driving. Like I said last time this came up, we didn't just fall off the turnip truck. If the appeal had been "I want to write an article about my micronation!" obviously it would have been declined. He had a one-account restriction he didn't violate, so all this stuff about socking is a distraction. He was topic-banned from articles where poor editing is a serious problem - medical articles matter in ways that differ from most topics, especially micronations and actresses. Opabinia regalis (talk) 09:35, 25 June 2017 (UTC)
- And I supported. I agree we definitely should have stated publicly that we'd unblocked, and I've suggested we put this in our procedures somewhere. I'm not going to discuss his email for obvious reasons. We decided to offer him another, probably last, chance assuming correctly that if there were problems they would be dealt with (and yes, that would have been easier if it had been announced). As for AN, waiting for 24 hours before making a final decision seems the optimal way to handle AN cases although circumstances can require faster action at times. I don't think this was one of those situations. Doug Weller talk 13:51, 25 June 2017 (UTC)
- "Backseat driving"? You have some guts, but you clearly don't know when to shut up. Let's see what a backseat driver is, shall we? "The term is also used allusively for any person who intervenes with advice and instructions in affairs they are not responsible for, or subjects they may not understand well." Considering that I have dealt with GdB for years, was (at least partially) responsible for his RfC and for his first reban and his second reban, I think this is an issue I understand quite well and take responsability for. Anyone may criticize ArbCom actions and communication, that should never be called "backseat driving", but to address it to people who actually know the case (and criticized it from the start of your interference, not only after it ended badly), is especially unwarranted and shows a clear lack of transparency and accountability.
- That you need to "ask" whether anyone minds that their support is posted is also quite bewildering. Any ArbCom member who voted in an ArbCom decision but refuses to have his vote made public should immediately resign from ArbCom, as there is no reason at all why such a vote would need to be secret.
- Looking back at my questions:
- Who in Arbcom voted for or against the unban (plus recusals, inactive, ...): no answer (collectively)
- Did Guido den Broeder deny the socking? "a distraction"???
- Were you aware of his COI editing problems? No answer
- Did he acknowledge any problem at all with his pre-ban editing? No answer
- Did he indicate which topics he wanted to edit? No answer
- But we do now know that you consider poor editing of the BLPs of 11-year old "beautiful girls" not potentialy "a serious problem". You didn't fall of the turnip truck, you were driving it blindfolded, singing "lalala" with your fingers in your ears.
- What made you (everyone who supported this) believe that he would this time, after 8 years of problems, suddenly become a productive, useful contributor, if you had no indication of what he would edit and what he would avoid (apart from the very limited topic ban you gave him), how he would avoid the original problems (which he still considers a witchhunt, just like the CU sock block was wrong). None of you have given any indication that you looked for more than 5 minutes at the original ban discussions and took any of it in any way serious. Never mind looking at how he went off the rails since the 2008 blocks, including his 2014 takedown notice to Google for dozens of Wikipedia pages featuring his name[3] (the same name he now used again for his username, now he again needs some publicity), his creation of a "second generation" wiki which happily constantly imports articles from wikipedia without the necessary attribution (e.g. this, his socking, and his creation of a micronation where he is the prince-monarch and which is the gateway to a parallel universe, and which his socks were already promoting on enwiki... Fram (talk) 08:24, 26 June 2017 (UTC)
I'll answer the questions for myself only. I did not actively participate in the thread on the mailing list in which it was decided to unblock with restrictions. Several experienced arbitrators reviewed the request and supported unblocking with a set of restrictions, and I accepted their judgment; we don't take a vote of all 15 of us on every unblock/unban request, although we're discussing whether we should. The editor was indeed asked what topics he intended to edit, and he mentioned what his current interests were. They did not relate to the areas in which he'd previously had problems, or in which he did have problems once he was unbanned. He did deny socking in 2015, but I don't think anyone relied on the denial in making their decision. Obviously the outcome of this unban was severely disappointing, and I think it is fair to state that given the failed last chance, there is no prospect this editor will be allowed to return in the future. I hope this helps. Newyorkbrad (talk) 16:52, 26 June 2017 (UTC)
- Thanks. "there is no prospect this editor will be allowed to return in the future" is what was said after the second ban as well though. Anyway, "They did not relate to the areas in which he'd previously had problems, or in which he did have problems once he was unbanned. " is patently wrong. His problems were most or all related to COI editing, whether it was about CFS, chess, basic income, or (with his socks even!) his personal monarchy. See the RfC/U that was ongoing at the time of the community ban. And I don't get why experienced arbitrators would ever allow an unban of someone who denies even the most basic facts, like his 2015 socking, even though it was CU-confirmed. Like I said, at AN / ANI discussions about granting an unban, unblock, lifting of restrictions, the two main issues are "time passed" and "realises what the previous problems were and gives good indication that they will be avoided in the future". Time passed (problems 2007-2015, is 2017 enough time passed?) is debatable, but the second golden rule of such discussions was in this case completely ignored for ... for what exactly? To allow an editor with a known anti-Wikipedia agenda and a known severe COI agenda back? Fram (talk) 17:09, 26 June 2017 (UTC)
- To be strenuously fair, the 2015 socking finding relied on a CU on another language Wikipedia comparing stale CU data stored on CUWiki to the confirmed socks. I think functionaries are inherently trustworthy in general and I have no reason to doubt the validity of the results, and like most other people I have little doubt about the 2015 socking. That being said, it's a somewhat... unusually "distant" result (as opposed to a local CU/Arb confirming with live, local data) so I can see why it would weigh less in a ban appeal process, as Opibania explicitly said. ☺ · Salvidrim! · ✉ 17:20, 26 June 2017 (UTC)
- Even as a purely behaviour based socking, it was a really strong case. But of course, if experienced arbs prefer the statement of a banned editor over that of experienced sockblocking admins... Fram (talk) 17:23, 26 June 2017 (UTC)
- @Fram: The topic-areas he mentioned as his current interests were not the ones in which he'd had problems before, or the areas in which he edited recently—or at least I didn't understand them as such. As for the denial of socking, it is probably less relevant to this case because for better or worse it wasn't the focus of the discussion, but let me put a general question out there. Let's suppose that the SPI admins and checkusers determine that an editor has been socking, with 95% confidence, and block on that basis. A year later, the editor asks to be unblocked, but adamantly denies that he or she ever socked. The editor might very well be still lying—but there's a finite chance he or she is telling the truth. Should we make it an absolute condition of unblock that the editor acknowledge the socking, even though it means that some wrongfully blocked editors will have to either stay blocked "forever" or else admit an offense they didn't commit? Or at some point does the importance of such an acknowledgement fade away? (I emphasize again that this is a hypothetical question, since in this case I know your position that the editor should have stayed blocked for lots of other reasons, but it's just food for thought.) Regards, Newyorkbrad (talk) 18:54, 26 June 2017 (UTC)
- To reiterate what I said before as well, it would seem that socking was contemplated in granting the unban request by imposing a one account restriction. I cannot speak for the Arbitrators that initially reviewed the request, but I very much doubt they would have imposed that restriction if they did not put weight in some regards to this aspect of the editor's past. Mkdw talk 19:20, 26 June 2017 (UTC)
- "You are not allowed to sock while you are unbanned" is not really a very strong restriction though. Fram (talk) 07:29, 27 June 2017 (UTC)
- To reiterate what I said before as well, it would seem that socking was contemplated in granting the unban request by imposing a one account restriction. I cannot speak for the Arbitrators that initially reviewed the request, but I very much doubt they would have imposed that restriction if they did not put weight in some regards to this aspect of the editor's past. Mkdw talk 19:20, 26 June 2017 (UTC)
- @Fram: The topic-areas he mentioned as his current interests were not the ones in which he'd had problems before, or the areas in which he edited recently—or at least I didn't understand them as such. As for the denial of socking, it is probably less relevant to this case because for better or worse it wasn't the focus of the discussion, but let me put a general question out there. Let's suppose that the SPI admins and checkusers determine that an editor has been socking, with 95% confidence, and block on that basis. A year later, the editor asks to be unblocked, but adamantly denies that he or she ever socked. The editor might very well be still lying—but there's a finite chance he or she is telling the truth. Should we make it an absolute condition of unblock that the editor acknowledge the socking, even though it means that some wrongfully blocked editors will have to either stay blocked "forever" or else admit an offense they didn't commit? Or at some point does the importance of such an acknowledgement fade away? (I emphasize again that this is a hypothetical question, since in this case I know your position that the editor should have stayed blocked for lots of other reasons, but it's just food for thought.) Regards, Newyorkbrad (talk) 18:54, 26 June 2017 (UTC)
- Even as a purely behaviour based socking, it was a really strong case. But of course, if experienced arbs prefer the statement of a banned editor over that of experienced sockblocking admins... Fram (talk) 17:23, 26 June 2017 (UTC)
- To be strenuously fair, the 2015 socking finding relied on a CU on another language Wikipedia comparing stale CU data stored on CUWiki to the confirmed socks. I think functionaries are inherently trustworthy in general and I have no reason to doubt the validity of the results, and like most other people I have little doubt about the 2015 socking. That being said, it's a somewhat... unusually "distant" result (as opposed to a local CU/Arb confirming with live, local data) so I can see why it would weigh less in a ban appeal process, as Opibania explicitly said. ☺ · Salvidrim! · ✉ 17:20, 26 June 2017 (UTC)
I still don't get why anyone would think it a good idea to ever unblock an editor with a history of problematic editing stretching back 7 or 8 years, over multiple language versions, with multiple problems, with no indication at all that they have learned anything from their problems. The reluctance by all of you to indicate "The topic-areas he mentioned as his current interests" remains, so I presume they included his micro-nation and "most beautiful" young girls where he paid good money to be able to "act" next to her, which should both have rung some major warning bells. I finally don't get why anyone's votes about this unban should remain secret: the only result is that I consider everyone who hasn't commented as someone who supported this unwise unban, but is not honest enough to own up to it. The former is a lack of judgment which is worrying but may be a one-off, the latter is indicative of a mindset which is unfit for arbcom duties.
So, the current list is:
- Pro unban, declared: Doug Weller, Drmies, Opabinia regalis
- Not active on the unban, declared: GorillaWarfare, Mkdw, Newyorkbrad
- Not made a comment, presumed to have supported the unban but not wanting to live up to it: Callanecc, Casliber, DeltaQuad, DGG, Keilana, Kelapstick, Kirill Lokshin and Ks0stm
As far as I am concerned, this can be closed as nothing more useful is likely to come from this very reluctant discussion. Fram (talk) 07:29, 27 June 2017 (UTC)
Functionary changes (July 2017)
Magioladitis 2 case request
I am troubled by how several arbitrators (Ks0stm, Mkdw, Calanecc as of time of writing) sound like they have already decided that Magioladitis is in the wrong. It makes the case seem like a show trial. If an individual arbitrator has already decided prior to seeing the evidence (i.e. evidence page) one way or the other, shouldn't he or she recuse? Banedon (talk) 08:30, 19 July 2017 (UTC)
- Hi Banedon, I think you're reading too much into my comment and there are two things it's important to point out. The first is that I didn't say I had decided to desysop, but that I found the arguments regarding possible breaches of ADMINACCT and ADMINCOND convincing (and the Magioladitis' statement didn't convince me otherwise. The second thing is that preliminary statements are part of the evidence in the case, in fact, there was a time when preliminary statements posted to the evidence page.
- The purpose of the case request page is to work out whether there is a case to answer, and, for me, whether a case might actually solve the issue. In this instance (where admin tools are concerned), ArbCom is the only venue which can remove the permission so my bar for accepting a case is lower than it otherwise might be. Having said that, there has to be a case to answer, otherwise opening a case is pointless.
- My acceptance comment was a clear indication that I believed there was an issue for the Committee to examine. At this early stage of a case my opinion isn't fully formed, but I am convinced that there is an issue that needs the Committee's attention, not that I'm certain we need to desysop. Callanecc (talk • contribs • logs) 09:08, 19 July 2017 (UTC)
Whatever decisions you will make would affect the AWB software, which I have been using for several years. Wouldn't they? I believe there are possible outcomes: AWB would still be there... but won't be updated this year, leaving AWB problematic and buggy; or AWB will be gone. If I can't use the AWB software, I have to manually disambiguate links in every page if I can't find an alternative software to do this. --George Ho (talk) 11:23, 19 July 2017 (UTC)Struck as irrelevant to the case. George Ho (talk) 17:18, 19 July 2017 (UTC)- I'm pretty sure that your convenience is not the primary reason to accept or decline a case, George. Your comment is really out of place here. Dennis Brown - 2¢ 16:16, 19 July 2017 (UTC)
- I concur that the case being accepted or not, nor how it is handled, should not be predicated on any software that a party has written. But, there's a larger issue at play here that needs to be addressed somewhere, somehow; software that has become critical to the operation of the project needs to managed by the community, not by a single contributing editor who wrote it. There are a lot of bots, for example, that do an incredible amount of work here that would make managing this project very difficult indeed if they were to stop functioning. --Hammersoft (talk) 16:18, 19 July 2017 (UTC)
- What has Magioladitis been telling you? He didn't write AWB; it was written by Bluemoose, and while Mag is one of the team who make tweaks to the software I'm sure that if he picked up his ball and went home, the other devs would get along just fine. (It might actually lead to significant improvement—it's no secret that most of the problems caused by AWB are the result of its scattergun misleadingly-named "General fixes" option, which is set as on by default and which enforces the "fixing" of assorted issues, many of which are problems solely in Magioladitis's imagination.) ‑ Iridescent 19:30, 19 July 2017 (UTC)
- Magioladitis hasn't told me anything, and so far as my rusty synapses can inform me, I've never talked with him about anything. I received information that he is an AWB developer. If he isn't, so be it, and the point I raised has nothing to do with this case. More abstractly, perhaps AWB is safe, but there are other processes that are not. --Hammersoft (talk) 19:42, 19 July 2017 (UTC)
- Honestly, I didn't know that one of AWB developers is the subject of this case. Whatever conducts Magioladitis have done as admin... I might briefly watch the case from time to time. I have no opinions about this person because I've never interacted with him before. Also, I struck my previous comment as "irrelevant" to the case. BTW, Magioladitis's first and second adminship requests are worth reading for references' sake. --George Ho (talk) 17:18, 19 July 2017 (UTC)
- I'm pretty sure that your convenience is not the primary reason to accept or decline a case, George. Your comment is really out of place here. Dennis Brown - 2¢ 16:16, 19 July 2017 (UTC)
- I haven't made any decisions yet. I expressed a familiarity with the situation because of the prior case and the subsequent community discussions regarding Magioladitis and COSMETICBOT issues. Being familiar with the issues during a request only affects my sentiment on whether a case should be heard or not. There is plenty of work still to be done involving all parties. The situation would have been inequitable if Magioladitis had not been involved in the community discussions beforehand nor been allowed to respond. In every situation I am familiar with they had been able to do so at length. I think it's dangerous to expand the request process into territory reserved for the case itself. I believe those who have expressed concerns are focused on whether Magioladitis will have a fair chance to defend themselves against several accusations made against them. That opportunity comes during the actual case. The request is only to decide whether the case should proceed (a type of dispute resolution) in which both parties involved have acknowledge (in more than one place on-wiki) ongoing drama. Broadly speaking, the general practice of using the request process as the place to hear the entire case including the defence will mean abandoning or severely lessening critical evidence and workshop phases. Mkdw talk 14:46, 19 July 2017 (UTC)
- Thanks for the responses Calanecc & Mkdw. I'm not so much concerned about whether Maglioladitis will have the chance to defend himself as I'm concerned about whether the arbitrators hearing the case have already decided. Changing someone's mind after they have formed an opinion is much harder than before. I would still like to hear from Ks0stm, since Ks0stm's accept comment is the most troubling of all. Banedon (talk) 08:35, 20 July 2017 (UTC)
Anchoring
All cases named after an editor are show trials. Even if you refuse to believe the case is irrecoverably tainted by the name of it as I laid out here, even if you choose to refute science that shows the bias is unavoidable, the case is still a show trial. This is because the "defendant" (the person named) is limited in their evidence to 1000 words and 100 diffs, per Wikipedia:Arbitration/Guide_to_arbitration#Evidence. This limitation is a hard ceiling regardless of how many people are parties to the case. Because of that, you end up with cases like this, where 61 people submitted evidence against the named party. But, the named party is still limited to only 1000 words to rebut that Himalayan mountain range sized evidence. The bottom line here is that Magioladitis is screwed. No matter what he does during this arbitration process, the table is dramatically tilted against him and there is nothing...nothing...he can do to stop it. As many people have been saying for years now, its becoming increasingly worthless to attempt to defend yourself in arbitration cases. Once you're named, you're done. The very naming and structure of cases sets up a free for all feeding frenzy with the named party as victim, with no hope for them defending themselves. I'm confident this will be followed up with comments saying this isn't the case. But, the evidence screams otherwise. --Hammersoft (talk) 15:03, 19 July 2017 (UTC)
- Regarding evidence length, Magioladitis can request more space if he needs it. We almost always grant such requests, particularly when there are a lot of statements to respond to. GorillaWarfare (talk) 17:14, 19 July 2017 (UTC)
- Understood, but can you explain how one person is to respond to 61 others filing evidence against them? The system is broken. I don't have a suggestion on how to fix this aspect of it, but broken it is. --Hammersoft (talk) 17:48, 19 July 2017 (UTC)
- The system isn't "broken", you just don't like the system, that's a very different thing. To the extent that an ArbCom case is something like a trial, remember that the vast majority of trials bring the evidence of multiple parties to bear on a single party. It's part and parcel of the justice system in many forms of government, and I don't see why it shouldn't be used here as well. If the problem being reported is about a single user, then the case should be named after that user, and the evidence should be about the behavior of that user -- for and against, and everyone who has something relevant to say can contribute.
In any case, Hammersoft, you've already been told by a number of Arbs that your complaints are a valid topic of discussion but not here. Please take it to the proper venue, as you've been instructed to, and stop muddying the waters here.Beyond My Ken (talk) 18:14, 19 July 2017 (UTC) Struck out inaccurate statements, with apologies. Beyond My Ken (talk) 21:47, 19 July 2017 (UTC)- There's no appetite to change it, so it doesn't happen. Perhaps the next committee will take a more responsible approach. The Rambling Man (talk) 19:02, 19 July 2017 (UTC)
- Beyond My Ken, I was specifically instructed to take it here by Mkdw, as seen here. I stopped raising the issue at WP:ARC as instructed. I took it here, as instructed. Now you're telling me it's not to go here, and that I wasn't told to take it here. Tell me again how the system isn't broken? :) (just a bit of humor...really). --Hammersoft (talk) 19:36, 19 July 2017 (UTC)
- One could always say that in the case filing, anyone submitting a response will automatically be considered a party to the case, allowing their contributions to be reviewed too, which might kept away drive-by people that just have an ax to grind, to speak. Alternatively, having a section of (in need of a better name) "peanut gallery" that non-parties to the case as filed can provide brief statements of support or opposition against the case, in anticipation that during the evidence phase of the case, they can provide more specific details and the "defending" party in this type of case will have more time and space to respond. Such statements should not require responses by the filing parties, and clerks should be empowered to remove inappropriate comments or make them full statements and thus party to the case (again, discouraging those that are just driving by with no purpose). --MASEM (t) 20:11, 19 July 2017 (UTC)
- @Masem: It would also keep away editors who provide evidence in support of the named party, and probably even more so. Putting the name of the case as the culprit predetermines that that editor is in the wrong, and if editors defend that party and are named in the case evidence will be used to bring them down. We all have dirty socks, and admins even more so. –Dirk Beetstra T C 21:07, 19 July 2017 (UTC)
- As I understand the ArbCom process, the case request page is to determine if there's enough at issue from a very high-level view for ArbCom to take the case or not (eg has the community exhausted its own approaches?). At that level, it should not be for people to lay out all the evidence supporting or against others' accusations; there needs to be enough evidence to justify the case, but shouldn't be a shotgun approach, which is what currently can happen (particularly in a case like this one). No, we don't want to scare editors away from providing reasonable evidence, but I believe that discouraging full statements from non-named editors where they don't really have drastically new evidence to supply, and instead simply providing the !votes about whether a case should be taken or not, would significantly help the situation that Hammersoft describes. It helps break the rebuttal cycle and does not require a "defending" party to have to reply to every statement made.
- Once a case is accepted, then is the time editors can provide much more volume of evidence, and where there is enough time and editing space to go into rebuttals. --MASEM (t) 22:55, 19 July 2017 (UTC)
- @Masem: It would also keep away editors who provide evidence in support of the named party, and probably even more so. Putting the name of the case as the culprit predetermines that that editor is in the wrong, and if editors defend that party and are named in the case evidence will be used to bring them down. We all have dirty socks, and admins even more so. –Dirk Beetstra T C 21:07, 19 July 2017 (UTC)
- One could always say that in the case filing, anyone submitting a response will automatically be considered a party to the case, allowing their contributions to be reviewed too, which might kept away drive-by people that just have an ax to grind, to speak. Alternatively, having a section of (in need of a better name) "peanut gallery" that non-parties to the case as filed can provide brief statements of support or opposition against the case, in anticipation that during the evidence phase of the case, they can provide more specific details and the "defending" party in this type of case will have more time and space to respond. Such statements should not require responses by the filing parties, and clerks should be empowered to remove inappropriate comments or make them full statements and thus party to the case (again, discouraging those that are just driving by with no purpose). --MASEM (t) 20:11, 19 July 2017 (UTC)
- The system isn't "broken", you just don't like the system, that's a very different thing. To the extent that an ArbCom case is something like a trial, remember that the vast majority of trials bring the evidence of multiple parties to bear on a single party. It's part and parcel of the justice system in many forms of government, and I don't see why it shouldn't be used here as well. If the problem being reported is about a single user, then the case should be named after that user, and the evidence should be about the behavior of that user -- for and against, and everyone who has something relevant to say can contribute.
- Understood, but can you explain how one person is to respond to 61 others filing evidence against them? The system is broken. I don't have a suggestion on how to fix this aspect of it, but broken it is. --Hammersoft (talk) 17:48, 19 July 2017 (UTC)
- Regarding evidence length, Magioladitis can request more space if he needs it. We almost always grant such requests, particularly when there are a lot of statements to respond to. GorillaWarfare (talk) 17:14, 19 July 2017 (UTC)
Masem is right, the purpose of the request phase is to convince arbcom that there's a substantive issue to be investigated, not to present detailed evidence. But as for the rest, I for sure don't want to see people voting (or even !voting) for how they think the committee should vote. That way lies madness. I have thought that requests ought to be more focused, with a separate subpage for each request (so the history is preserved in an easily accessible location) where the parties can comment on the request page itself and commentary from uninvolved editors can be added on the talk page. (When I suggested something similar awhile back, the counter-argument was that subpaging requests makes them easier to miss on watchlists. My thought is that if you're not following the dispute closely enough to know by some source that it's at arbcom, you're probably better off not getting sucked in ;)
On Hammersoft's stuff about case names (...this is still not quite the right place for it, since I assume you'd feel similarly about a section header with someone's name in it...) I'll take my own advice and repost the relevant part of my previous comment on the case request: And if you need examples of named parties who weren't sanctioned, Wikipedia:Arbitration/Requests/Case/Michael Hardy and Wikipedia:Arbitration/Requests/Case/Kww and The Rambling Man should do nicely. Conversely, the most seemingly-prejudicial case name in the recent past was "Vested Contributors" (later AE2), which of course does not name any editors. A look through some of the rejected requests of the past couple of years - I suggest in particular the recent Godsy/Legacypac request and last year's request naming Fram - should tell you just exactly how effective it is to file early for the sake of getting to frame the case. Hammersoft's analysis would be more convincing if cases were randomly assigned to the name-by-user or name-by-topic condition, but as stated it overlooks prior probability.
Opabinia regalis (talk) 04:35, 20 July 2017 (UTC)
- Masem, if everyone who comments is named as a party, any case can get detoured into unconnected areas. I made a comment on the Magioladitis proposal, noting a relevant ANI thread and some comments in it (which at least one Arbitrator found useful). M and I then had a productive discussion on my talk page, which I linked to from the case request. Now, suppose the case is accepted and your proposal was in effect. The only interactions I recall ever having had with him are already in the request, so the rest of my history is irrelevant to the topic at issue. If someone wanted to drag my editing before ArbCom, they would need either a separate case or to produce evidence directly relevant to M and persuade ArbCom that it warranted bringing me in as a case. Under your proposal, however, I would be a party and anyone with some issues with me could make a submission on me. It would be irrelevant to the case, but it would add more material to the evidence and workshop phases... multiply that by (say) 10 other editors and ArbCom would have a huge pile of irrelevancy to wade through. I doubt ArbCom would welcome that. You might argue that if I am a productive editor, I would have no fear from scrutiny, but I did have a now-blocked sock target me a while back and most established editors (and certainly admins) would have someone who might want to attack them under the cover of "evidence" to ArbCom. I applaud your motivation in seeking to improve ArbCom processes but I think this approach is flawed. EdChem (talk) 06:16, 20 July 2017 (UTC)
- My point is more getting the case request page to avoid being an evidence phase. Editors that are not directly involved with the issue, but may have seen or participated in discussions at AN/ANI/elsewhere shouldn't be laying out tons and tons of evidence, though they should be able to state, for example "I have seen the previous ANI thread and believe ArbCom should take this case". Sort of an amicus brief if we talk court proceedings. After the case is taken, then more evidence and participants can be added.
- I would agree that scare tactics ("you become party to the case if you supply evidence during the case request page") is a problem, but we also need less pre-case discussion that is better suited for an accepted case. If a non-named party is providing a ton of evidence that clearly shows them involved, they should be moved to named at the discretion of the arbcom or clerks. Moreso, clerks should be better empowered to tell non-named parties to avoid long statements at this point (maybe even restricted non-named parties to 200 words or the like, or otherwise accepted being involved/named to have more space). There's a handful of ways to avoid the case request page from becoming an active case page before ArbCom has decided. --MASEM (t) 14:24, 20 July 2017 (UTC)
- And to User:Opabinia regalis's point, about I for sure don't want to see people voting (or even !voting) for how they think the committee should vote., it's more about non-named editors giving their opinion if all necessary requirements for an ArbCom case are met: is it a behavioral issue and not content, has the community done all it can and failed to work, does it fall under existing sanctions, are there other issues involved not yet identified by the filers, etc. The !votes of non-named parties should be towards "ArbCom should accept" or "ArbCom should decline", and not the drive-by "this person has been a thorn in my side, ArbCom can deal with them" type comments. --MASEM (t) 14:29, 20 July 2017 (UTC)
- No, User:Opabinia regalis, in the case "Kww and TRM", Kww did get sanctioned (and TRM shortly after got their own case where they got resigned their advanced rights under a cloud). Michael Hardy did not get sanctions in the terms of removal of bits, they however did get told that further misuse will result in sanctions. There are enough cases where the named party did not get sanctioned on a first case, but those almost always have follow ups. Anchoring .. ??
- You have, however, completely misunderstood what the issue is .. the issue is that the focus of the case is NOT on the issue of the conflict, the focus of the case is on the named party(s). The naming of the case defines from the start what the focus is. You do NOT have a properly established procedure to determine scope of a case, you do NOT have procedure to determine what the issues at hand are. You accept, often with remarks that one is going to look at the behaviour of an editor, and from that moment anything that fits becomes the scope of the case (see the statements in the Magioladitis case, there are remarks there which are NOT about Magioladitis' bot edits or bot-like edits or COSMETICBOT, but which are completely out of that. Even if the real issue is friction within the bot-operator group that will not be what is reflected in the outcome of the case, because if someone was to provide evidence regarding bot operator X it will be countered with 'but they are not a party to this case'. And that is a rightful answer, it is unfair that I provide a lot of bot errors, miscommunication and misunderstanding by someone else than the two named editors in the case. And that is because cases focus on named editors. (and I can agree, maybe it is not only that personally named cases are wrong, maybe we should not be naming cases in the first place, or only rename them at the final decision).
- Regarding the statistical problems with Hammersoft's analysis - if case names would be neutral throughout that analysis would be fair, and that is exactly the problem, there is a question of corrolation vs. causation giving at best a feeling of impropriety, at worst the issue is a real one. But as User:The Rambling Man already said: there is no will to change that system, arbitrators have been weaseling out of this for years. But there is no reason to keep the current naming system, I have yet to see an argument why cases have to be named after the users involved other than that we have always done that. --Dirk Beetstra T C 06:09, 20 July 2017 (UTC)
- That's the point - two named parties in the Kww/TRM case, only one was sanctioned. Yes, TRM did have his own case, if by "shortly after" you mean "over a year later".
- One of the things I've noticed during my arb term is that people will believe just about anything before they'll believe that someone else has given an issue honest consideration and still disagrees. If multiple committees' worth of arbs choose not to act on your suggestion, they can't just be disagreeing with you; they must be "weaseling out of it", for reasons of... I have no idea what personal stake anyone could have in the case naming system, so I guess it must be our inherent mustelidity. (As probably a dinocaridid, I object to being associated with stinky vertebrates like that!) Opabinia regalis (talk) 06:31, 21 July 2017 (UTC)
- My personal opinion on this is that the first step to fighting bias is to be aware of it. Like, the bystander effect is extremely well established, but if one gathers a group of people, explain to them what the bystander effect is and then explicitly tell them that they're going to be measured for the effect, will you still get the same results? As long as Arbcom is aware that such a bias is possible, its effects should be moderated. Having said that I don't see any reason why this can't just be changed. Is it difficult to do? I doubt it. Is anything lost by doing it? I don't see any. In that case, why not just do it? Banedon (talk) 08:39, 20 July 2017 (UTC)
- I agree, arbitrators may very well be aware of the bias and try to counter it. On the other hand, User:Opabinia regalis mentioned that
there's also the fact that the background to this case is already familiar to most of us, since the previous case was recent and we were aware of the community restrictions.
in response to speedy accepts in ArbCom cases - giving the perception that indeed arbitrators accepted the case because it was anchored to the person? (I do note that Opabinia did not accept the case at that time). - Whether anchoring really is a problem we don't know. As User:Opabinia regalis is mentioning, it is impossible to measure. On the other hand, what would we lose if we would turn it around. Put a button on the main arbcom page, stating 'File new arbitration request', and have it, through a LUA module, create a unique (or sequentially) numbered subpage with a predetermined format. The editor fills in some data - involved editors, involved policies/guidelines, earlier discussions, earlier cases what not, saves, informs the necessary editors, and records that. As you say, I doubt we will lose anything more then the perceived impropriety. --Dirk Beetstra T C 09:33, 20 July 2017 (UTC)
- Opabinia, if you want to look at rejected cases as well, the statistical analysis fails for precisely the same reasons. My study covered three years worth of accepted cases where one or more parties were named in the title, 32 cases in all with 36 titled named parties. Every single one of the title named parties had sanctions placed against them in the remedies. You cite two cases that are not (yet; working to bring the study through to current) in my data set. 2 cases vs. 32. I dare say 32 cases is a more statistically relevant sample. And to Beetstra's point; why the resistance? What benefit does the current naming schema bring to the table that we can't live without? It's an excellent question to ask. Due to the nature of the entire data set, we will never be able to get a completely accurate statistical picture. Yet, the evidence does suggest a serious problem. Further, science has backed up the reality of this being a real problem in human psychology. Why defend the current system in the face of this? Or are we just change resistant? --Hammersoft (talk) 13:30, 20 July 2017 (UTC)
- What would we lose you ask? I'm not sure how this would work. I'm not sure I like the idea of cases being subpages with more subpages. I presume they'd be numbered or somehow anonymously identified. I don't know if this would make as many people aware of the existence of a case. I don't know if your suggestion would have any other consequences not related directly to your desire not to name cases with names. Doug Weller talk 14:13, 20 July 2017 (UTC)
- I'm very, very far removed from being a fan of change for the sake of change. I have fought very hard at RFA over proposals for reform that did not first identify a problem, which is all the rage on the project in general. People like to come up with solutions and then try to find the problem, all the while forcing the problem to fit their preconceived notions of how the solution is great and would solve the problem. This is wrong. My point here isn't about that, but rather identifying why we have the system we have now, so as to help avoid pitfalls if we changed direction. --Hammersoft (talk) 14:21, 20 July 2017 (UTC)
- @Doug Weller: You are "not sure [you] like the idea of cases being subpages with more subpages" .. fine, leave that part as it is, but generate generic titles. Have the case request being a sequential number, and when you start the case (in which case it become subpages), name it generic, or keep the same number. In that case just as many people would be aware - you have a template linking to open cases, it would just link to cases 00005 and 00008 in evidence phase, 00003 in final decision stage, and 0004 was just closed last week. If it is you that is there, and the case is named after you, what benefit do you have as the named party from having a case called 'Doug Weller', and what benefit would any 'accusing' parties have from that naming over when that case would be named Wikipedia:Arbitration Committee/Requests/Case/00001. Recognisability - we get to case 'Doug Weller 2', so we better quickly accept, we've seen him before and he is apparently still a problem? And you don't know whether there are other consequences? I still wait to hear benefits of the current naming scheme, or clear (negative) consequences of adopting another naming scheme. I do see a clear problem with the current naming scheme - for 'Magioladitis 2' one of you described it as
there's also the fact that the background to this case is already familiar to most of us ...
- in other words, the problem was anchored to the name of the party, not to the issue. No need to wait to hear more about the issue, immediately accept. --Dirk Beetstra T C 15:34, 20 July 2017 (UTC)- @Beetstra: And I see some major difficulties in remembering what cases were about what. Can you give me some analogous real world examples where it's clear this method works well? Doug Weller talk 16:13, 20 July 2017 (UTC)
- Bingo, Doug. And that is exactly what anchoring is about - you don't remember what the CASE is about, you don't know what the ISSUE was - you remember the name of the party. And that is why some arbs vote fast: 'oh, .. him again .. pff .. let's get this over with shall we'. --Dirk Beetstra T C 16:40, 20 July 2017 (UTC)
- That's a self enforcing circle Doug. Case so-and-so was about so-and-so because it was about so-and-so. --Hammersoft (talk) 17:26, 20 July 2017 (UTC)
- @Beetstra and Hammersoft: Sorry, but that doesn't answer my first point directly or address my question about analogous real world examples at all. I'm more or less off-line, at least for this, until tomorrow now. Doug Weller talk 19:04, 20 July 2017 (UTC)
- @Doug Weller: Doug, this is not a discussion about other things that work well, this is about that this does not work. the case being brought is what the OP thinks is the problem, and that is (even per the first accept votes) what you are looking at. Again, Anchoring is a thing, it is what you clearly show that you do - you want to remember the case by the name of the editor - because that was the issue that you want to remember. Now you can try to be open towards thinking about how to avoid anchoring, but don't throw the 'if you don't have a solution, then don't mention the problem'-card. --Dirk Beetstra T C 22:19, 20 July 2017 (UTC)
- @Beetstra and Hammersoft: Sorry, but that doesn't answer my first point directly or address my question about analogous real world examples at all. I'm more or less off-line, at least for this, until tomorrow now. Doug Weller talk 19:04, 20 July 2017 (UTC)
- @Beetstra: And I see some major difficulties in remembering what cases were about what. Can you give me some analogous real world examples where it's clear this method works well? Doug Weller talk 16:13, 20 July 2017 (UTC)
- @Doug, could name the case based on what it's about, e.g. this case could be "cosmetic editing 2". There's already an example of that, the War of the Pacific case. Banedon (talk) 01:18, 21 July 2017 (UTC)
- @Banedon: Sorry, that didn't ping me and I missed it. I think it's a confusing and I think it oversimplifies the issues. We don't mention it in the scope and no one looking at that name would guess what the scope actually is (which now, for instance, includes the possibility of other parties being added for reasons which have nothing directly to do with cosmetic editing). Doug Weller talk 10:42, 24 July 2017 (UTC)
- What would we lose you ask? I'm not sure how this would work. I'm not sure I like the idea of cases being subpages with more subpages. I presume they'd be numbered or somehow anonymously identified. I don't know if this would make as many people aware of the existence of a case. I don't know if your suggestion would have any other consequences not related directly to your desire not to name cases with names. Doug Weller talk 14:13, 20 July 2017 (UTC)
- I agree, arbitrators may very well be aware of the bias and try to counter it. On the other hand, User:Opabinia regalis mentioned that
- I still haven't gotten an answer to this. Anchoring means that the first assertion made is the one that "sticks". It doesn't have to do purely with how you name things. If the first assertion in any case is an assertion about an individual (which it almost always will be, and certainly must be in a case that's brought about a single individual's behavior), we haven't overcome anchoring, whether we name a case "Editor's Name" or "Fluffy bunnies are cute". The actual assertions being made are going to rather clearly act as a stronger anchor than however we name a case. I don't see how the naming of a case does any harm above and beyond the existing anchor of very specific negative accusations. How do you propose we overcome that? It makes little sense to tackle an (at best) third or fourth order effect of case names when we have a stronger effect from the initial accusation that can't be removed from the process. ~ Rob13Talk 20:16, 20 July 2017 (UTC)
- The issue arises throughout the entire case. It's not just the initial statement. Wherever the case gets listed, it gets listed as the title named party. Therefore, the case is about that person, NOT the dispute itself. Throughout the initial rfar, the evidence phase, the workshop phase, ...everyone gets anchored onto whomever the case is named after. The entire case gets polluted by the very name it exists under. So we start a case called "JoeBlow", so therefore the case is about "JoeBlow" and we're looking for evidence about "JoeBlow", and we're working in the workshop about "JoeBlow" and wow what a shock the proposed decisions are about "JoeBlow". Once you start a case under someone's name you set in place a railway that leads directly to sanctions for that person. What's very bizarre is we let filers determine the initiating name of the case, rather than parties to the case presenting their information and clerks/arbitrators determining a name (and therefore the locus of the dispute, if it's not generated generically). Therefore, the filer gets a massive "advantage" in the case in that they can get the focus of the case on that they perceive the problem as, rather than a disinterested party evaluating that. That clearly biases a case. --Hammersoft (talk) 20:46, 20 July 2017 (UTC)
- First of all, that name thing is incorrect; the filer chooses a case request name but the clerks/arbitrators regularly change it when appropriate. What's your alternative? Say we do away with names entirely and just use a number to eliminate that issue. I go to ArbCom and file a case request due to the behavior of "Joe Blow". The case is #928. In my opening statement, I explain all the behavioral issues with "Joe Blow" and ask for review of the behavior, which doesn't involve particularly bad behavior from any other sides; "Joe Blow" is just an unblockable who has been particularly disruptive lately. Others present statements about "Joe Blow". We look for evidence about "Joe Blow". We work in workshop about "Joe Blow". We get a proposed decision about "Joe Blow". I would say the origin of the railway to sanctions was "Joe Blow"'s behavior, but let's be generous any say it was my initial complaint laying out all the ways in which "Joe Blow" is being disruptive. That's the anchor, whether or not the case is just #928. ~ Rob13Talk 21:21, 20 July 2017 (UTC)
- No, User:Bu Rob13 you perfectly align us for the follow up problem, which I already addressed here today - Arbitration does NOT determine scope, arbitration does not determine what the issue is. You, e.g., report Magioladitis, asserting that there are problems with HIM (you may be right ..). ArbCom will now blindly copy that (that is what you see in some of the accept votes). Other editors will add their beef to it (what you see in some of the statements, they are addressing problems that they have with Magioladitis which are utterly outside of the issues that were in the remedies of the first case, which were outside of the bans, and not the reason of the block). The case is anchored to Magioladitis. If there are problems within the bot-operators groups (I don't know, rivalry), conflicts between the bot-approvals-group and the bot operators, differences in interpretation of the bot-related policies by different bot operators or bot approval group members, or differences between the interpretations of the bot-related polices by bot operators and the community, or maybe you yourself are at the wrong side of the communities will, which all could result in the same symptom: One bot operator running into continuous problems (and Magioladitis is the third in front of arbcom at least, I hear of operators which simply walked away - my bots have been down for months as well at some point), it all doesn't appear. We will be back here with another bot operator (and you predicted that in Magioladitis 1 .. ). You (and I don't blame you at all, you follow common practice and unclear/unwritten procedures) anchored the case to Magioladitis - and that is what is being examined now (again, see the accept votes). Magioladitis is the issue. And everything will go. Maybe there is an iffy block by him. Maybe an involved page protection by him. Maybe he did not answer sufficiently to a concern of a long-term admin. Maybe he was rude to someone. Because you brought Magioladitis here, and that is the lead that ArbCom will follow. Three accepts before he was responding, he was here before, it must be him. There is NO neutral determination of scope, issues or problem. Most is determined by the OP's opening statement. --Dirk Beetstra T C 22:19, 20 July 2017 (UTC)
- Yes, and I'm saying that's not a follow-up problem. If it's a problem (and I don't think it is, because the Committee has taken cases that have come back on the filer before), then it's clearly a problem that subsumes the naming issue. If the Committee can see past the anchor of a long-form initial complaint, they can see past the name issue. If they can't see past the long-form initial complaint, the naming doesn't make it any worse. So until you propose some total overhaul of how cases are filed for arbitration, the naming issue is higher-order. It doesn't affect things in any significant manner. ~ Rob13Talk 23:09, 20 July 2017 (UTC)
- And the Magioladitis Case Request is a perfect example of that the ArbCom does not see past the anchor. They admit it
there's also the fact that the background to this case is already familiar to most of us ...
. But again, you may very well be right that it is not a problem, but you will never know. Fact is that in all cases where there are names of parties in the name of the case, the named party. And I will ask it again: what is the problem with naming a case completely neutral. --Dirk Beetstra T C 06:23, 21 July 2017 (UTC)- "Anchoring" is not a generic synonym for "remembering things". This line of argument strikes me as very strange, that a group in theory responsible for resolving disputes between users and dealing with user-conduct problems shouldn't remember prior disputes by reference to the users involved. Opabinia regalis (talk) 06:31, 21 July 2017 (UTC)
- And the Magioladitis Case Request is a perfect example of that the ArbCom does not see past the anchor. They admit it
- Yes, and I'm saying that's not a follow-up problem. If it's a problem (and I don't think it is, because the Committee has taken cases that have come back on the filer before), then it's clearly a problem that subsumes the naming issue. If the Committee can see past the anchor of a long-form initial complaint, they can see past the name issue. If they can't see past the long-form initial complaint, the naming doesn't make it any worse. So until you propose some total overhaul of how cases are filed for arbitration, the naming issue is higher-order. It doesn't affect things in any significant manner. ~ Rob13Talk 23:09, 20 July 2017 (UTC)
- No, User:Bu Rob13 you perfectly align us for the follow up problem, which I already addressed here today - Arbitration does NOT determine scope, arbitration does not determine what the issue is. You, e.g., report Magioladitis, asserting that there are problems with HIM (you may be right ..). ArbCom will now blindly copy that (that is what you see in some of the accept votes). Other editors will add their beef to it (what you see in some of the statements, they are addressing problems that they have with Magioladitis which are utterly outside of the issues that were in the remedies of the first case, which were outside of the bans, and not the reason of the block). The case is anchored to Magioladitis. If there are problems within the bot-operators groups (I don't know, rivalry), conflicts between the bot-approvals-group and the bot operators, differences in interpretation of the bot-related policies by different bot operators or bot approval group members, or differences between the interpretations of the bot-related polices by bot operators and the community, or maybe you yourself are at the wrong side of the communities will, which all could result in the same symptom: One bot operator running into continuous problems (and Magioladitis is the third in front of arbcom at least, I hear of operators which simply walked away - my bots have been down for months as well at some point), it all doesn't appear. We will be back here with another bot operator (and you predicted that in Magioladitis 1 .. ). You (and I don't blame you at all, you follow common practice and unclear/unwritten procedures) anchored the case to Magioladitis - and that is what is being examined now (again, see the accept votes). Magioladitis is the issue. And everything will go. Maybe there is an iffy block by him. Maybe an involved page protection by him. Maybe he did not answer sufficiently to a concern of a long-term admin. Maybe he was rude to someone. Because you brought Magioladitis here, and that is the lead that ArbCom will follow. Three accepts before he was responding, he was here before, it must be him. There is NO neutral determination of scope, issues or problem. Most is determined by the OP's opening statement. --Dirk Beetstra T C 22:19, 20 July 2017 (UTC)
- First of all, that name thing is incorrect; the filer chooses a case request name but the clerks/arbitrators regularly change it when appropriate. What's your alternative? Say we do away with names entirely and just use a number to eliminate that issue. I go to ArbCom and file a case request due to the behavior of "Joe Blow". The case is #928. In my opening statement, I explain all the behavioral issues with "Joe Blow" and ask for review of the behavior, which doesn't involve particularly bad behavior from any other sides; "Joe Blow" is just an unblockable who has been particularly disruptive lately. Others present statements about "Joe Blow". We look for evidence about "Joe Blow". We work in workshop about "Joe Blow". We get a proposed decision about "Joe Blow". I would say the origin of the railway to sanctions was "Joe Blow"'s behavior, but let's be generous any say it was my initial complaint laying out all the ways in which "Joe Blow" is being disruptive. That's the anchor, whether or not the case is just #928. ~ Rob13Talk 21:21, 20 July 2017 (UTC)
- The issue arises throughout the entire case. It's not just the initial statement. Wherever the case gets listed, it gets listed as the title named party. Therefore, the case is about that person, NOT the dispute itself. Throughout the initial rfar, the evidence phase, the workshop phase, ...everyone gets anchored onto whomever the case is named after. The entire case gets polluted by the very name it exists under. So we start a case called "JoeBlow", so therefore the case is about "JoeBlow" and we're looking for evidence about "JoeBlow", and we're working in the workshop about "JoeBlow" and wow what a shock the proposed decisions are about "JoeBlow". Once you start a case under someone's name you set in place a railway that leads directly to sanctions for that person. What's very bizarre is we let filers determine the initiating name of the case, rather than parties to the case presenting their information and clerks/arbitrators determining a name (and therefore the locus of the dispute, if it's not generated generically). Therefore, the filer gets a massive "advantage" in the case in that they can get the focus of the case on that they perceive the problem as, rather than a disinterested party evaluating that. That clearly biases a case. --Hammersoft (talk) 20:46, 20 July 2017 (UTC)
- Exactly, User:Opabinia regalis - you do not remember the dispute, you remember the editor. If an editor is here first in front of you in a dispute between A and B about subject X, and then later brought to you in a dispute between A and C about subject Y, you remember the editor, not the dispute - the first case may basically be utterly irrelevant to the second case.
- And there is NOTHING to remember - when a case is brought, generally the OP will list relevant previous attempts to resolve the issues, which will obviously include recent earlier Arbcom cases. The second case request currently open for Magioladitis clearly links to the first case. Whether the first one was called 'Case 00007', 'Magioladitis', 'COSMETICBOT' or 'blue balloon', it does not matter - if findings of fact / remedies out of that case are relevant it will either already be presented to you in the request, or it will be coming to light in evidence phases or from your own records. Moreover, if 'Magioladitis 1' would have taken place before your tenure, you would not recollect it (see the fading habit of voting before all statements are in, that happened before most of your tenures). Or if Magioladitis had been a party in 5 non-username-named cases you would not remember him, this case would be Magioladitis 1, and you would not accept on the basis of that knowledge (unless it was presented to you), and if I would be brought to you now because of issues with COSMETICBOT you would not recall Magioladitis' case because .. I did not have a case yet. However, since it is only a few months ago that we had a case about COSMETICBOT/Magioladitis, it is utterly relevant to a case about COSMETICBOT/Dirk Beetstra.
- So still the question is, what is the benefit of having cases named after users themselves. --Dirk Beetstra T C 10:06, 21 July 2017 (UTC)
- The only benefit I can see, is that it can be used for hounding. A user that already has a named case against him, can easier be brought to ANI or Arb even when he was cleared at the time. Agathoclea (talk) 10:20, 21 July 2017 (UTC)
- I don't see why that wouldn't happen no matter what we named the case. And hopefully using a case where someone was cleared would be seen as hounding. Or even, who knows, bringing up the past case would be ignored and ANI or we would deal only with any current issues. Whatever we call a case, when it's filed the filing will make it clear the names of any parties the case is being brought against. Those names then become the way people think about the case, not an artificial name. And usually prior cases involving named parties are mentioned in the filing. Doug Weller talk 10:47, 21 July 2017 (UTC)
- Exactly .. the name is how people think about the case, not the issue. And indeed, prior cases are mentioned, no matter what the name is, and the filing will also make clear who the users in the case are. So no reason to name the case after the user. Pfff .. is it so difficult to give reason why cases should be named after the named users in the case? –Dirk Beetstra T C 11:38, 21 July 2017 (UTC)
- Perhaps this is an issue, but you've seriously picked the wrong case to make the poster child of it. In this case (and the previous case on the same editor), the issue was Magioladitis. The evidence was presented solely involved him and the community at large, not a conflict with any other individual editors. We name cases after named editors to identify what they're about. That is the reason that's been given repeatedly, which you've ignored. The cases were about Magioladitis. ~ Rob13Talk 14:38, 21 July 2017 (UTC)
- EVERY case with a title named party is a poster child for this. I'm sorry all the parties seem to be talking past each other on this point. It doesn't matter if Magioladitis is 10000000000000000000000000% in the wrong and the community all eternally wear halos in heaven on high. Naming a case with any individual editor in the name pollutes the entire case and forces bias into it. Just maybe...just MAYBE...Magioladitis has a good case to make. But, we will never know because the anchoring bias forces the case to be about him and only him. You're doing it yourself here. As I said before, it takes two to tango. Its extremely rare there is ever a dispute where one side is flat out, completely, utterly in the wrong and every point they have to make is utterly refutable. There are two sides to every coin. When you name a case after someone, you strongly encourage people to ignore the other side of the coin and instead focus only on the supposed wrongdoings of the name party. Psychological studies have proven this happens. Yet, it keeps getting denied here in this conversation. I remember raising this issue some years before and ArbCom insisted they were not being swayed by anchoring bias even despite being aware of the psychology studies showing they couldn't avoid being swayed. Now it's happening again. --Hammersoft (talk) 15:27, 21 July 2017 (UTC)
- Alternatively the more simple explanation is that every case named after a single editor ends up with sanctions against the single editor because that single editor has pissed off enough people enough times to get an arbcom case opened against them with significant amounts of evidence provided - and in most cases, already community or past admin sanctions against them. Your and Dirk's opinion is that naming the case after an editor has a cause and effect on the outcome of the case in some manner. Which is unlikely given the arbcom case archives. Only in death does duty end (talk) 15:35, 21 July 2017 (UTC)
- I prefer to rely on science which says otherwise. Maybe it is all about them. Who knows? Maybe there's more to the dispute than just that editor? We have considerably less chance in cases that are biased by this anchoring issue because evidence of problematic behavior by non-named parties has less of a chance of being submitted. --Hammersoft (talk) 15:43, 21 July 2017 (UTC)
- @Only in death: self fulfilling prophecy. --Dirk Beetstra T C 15:46, 21 July 2017 (UTC)
- @BU Rob13: (edit conflict) Still, it is not a reason for anchoring (and I brought another issue here, though have strong opinions on this issue). The very first case about Magioladitis was where the actual anchoring took place - there it goes wrong. Do note, that 4 of the 7, the majority, remedies in that case were NOT about Magioladitis but aimed at bot operators at large and at the community. I would argue that that case was wrongly anchored.
- This case (and the declined second request) now shows now how wrong that anchoring is - the declined case got two accept votes before Magioladitis could put down a statement, despite being about a completely, utterly different type of dispute that had nothing to with what the issues were in the first case. Yet one of the (leaning to) accept votes there says
... then the continuing issues which present themselves re Magioladitis ...
, here now it isthere's also the fact that the background to this case is already familiar to most of us ....
. This is exactly the right case to pick - just because the anchoring is so recent in memory that these cases show to the full extend what the problems are. --Dirk Beetstra T C 15:38, 21 July 2017 (UTC) - To add to Hammersoft's point Its extremely rare there is ever a dispute where one side is flat out, completely, utterly in the wrong and every point they have to make is utterly refutable. -- and when this is the situation, this is nearly always dealt with at AN/ANI/AE or other boards. But if those boards clearly can't come to a clear consensus, then that probably means the editor in question has some arguable points that should be considered fairly at ArbCom, but the anchoring predepositions the case to make that editor defensive rather than seek a middle ground. --MASEM (t) 15:55, 21 July 2017 (UTC)
- The problem in these cases, specifically when the focus is the behavior of a single editor, isn't anchoring. The problem is the ungodly amount of evidence that can be brought against one editor that can't possibly be responded to by said editor. Any reform of the process should center around limiting the evidence phase. Depending on the number of named parties there should be a limit on how many non-named parties can present evidence. Say, five or more named parties then they are the only editors that are allowed to present evidence. Less than that eached named party can have a willing non-party editor of their choice present evidence. If there's only two named parties then allow two non-party editors of their choice present evidence. After that anyone who presented evidence gets one, and only one, response to each section of presented evidence. That should be more than enough for ArbCom to determine if there's any issue requiring remedies. Each phase after that should only involve editors who presented evidence. One huge problem with ArbCom cases is how anyone can chime in at any time, at any phase no matter their involvement, particularly on talk pages. The whole point of ArbCom is that there's an issue that the community can't solve. There's no need to allow the community to keep drive-by chiming in at their leisure at any point, effectively presenting evidence after the evidence phase is supposed to be closed. Cases should be far more structured. Capeo (talk) 16:48, 23 July 2017 (UTC)
- Perhaps this is an issue, but you've seriously picked the wrong case to make the poster child of it. In this case (and the previous case on the same editor), the issue was Magioladitis. The evidence was presented solely involved him and the community at large, not a conflict with any other individual editors. We name cases after named editors to identify what they're about. That is the reason that's been given repeatedly, which you've ignored. The cases were about Magioladitis. ~ Rob13Talk 14:38, 21 July 2017 (UTC)
- Exactly .. the name is how people think about the case, not the issue. And indeed, prior cases are mentioned, no matter what the name is, and the filing will also make clear who the users in the case are. So no reason to name the case after the user. Pfff .. is it so difficult to give reason why cases should be named after the named users in the case? –Dirk Beetstra T C 11:38, 21 July 2017 (UTC)
- I don't see why that wouldn't happen no matter what we named the case. And hopefully using a case where someone was cleared would be seen as hounding. Or even, who knows, bringing up the past case would be ignored and ANI or we would deal only with any current issues. Whatever we call a case, when it's filed the filing will make it clear the names of any parties the case is being brought against. Those names then become the way people think about the case, not an artificial name. And usually prior cases involving named parties are mentioned in the filing. Doug Weller talk 10:47, 21 July 2017 (UTC)
- The only benefit I can see, is that it can be used for hounding. A user that already has a named case against him, can easier be brought to ANI or Arb even when he was cleared at the time. Agathoclea (talk) 10:20, 21 July 2017 (UTC)
- @Capeo: anchoring is part of that problem - on a case on 'issues' editors will focus their evidence on the issues, on a case on a 'case named party' (case is named after one or more of the parties), the evidence will focus on those editors (and any reall issues may be ignored). It then does make a difference whether (sometimes gargantuan amount of) evidence is aimed at an issue, or at the one editor that is giving the name to a case. --Dirk Beetstra T C 16:57, 23 July 2017 (UTC)
- That seems a false dichotomy to me. It assumes the issue is not the editor, their behavior, themselves. Particularly in admin conduct cases the issue is very much the editor. Capeo (talk) 17:22, 23 July 2017 (UTC)
- @Capeo: true, such cases exist. However, I do not see it as a reason to anchor that to the person either (it is rare, but the editor may be cleared). For cases where that is however not the case, naming those after one or two editors is just wrong. --Dirk Beetstra T C 17:32, 23 July 2017 (UTC)
- Well, I think Brad just stole my thunder below. I think you give far too much credit to anchoring as some bias that isn't easily overcome, especially around here. In my experience, as someone has presented evidence in an ArbCom case that was named after a single editor and ones that weren't but still focused on one or two editors, the insurmountable amount of diffs, that the named editors can't possibly give context to, does far more to bias the outcome than any naming convention. To the point where the "accused" eventually break and lash at out the hopelessness of it. In both those cases I agreed with the outcomes. I didn't agree with the needless piling on and needless bludgeoning. A few well crafted evidence sections, with an allowance for one response, would have arrived at the same conclusion. 50 editors, with repetitive evidence, many just saying not much more than, "this editor is bad and needs to go", is not something anyone should have to deal with. Capeo (talk) 18:03, 23 July 2017 (UTC)
- @Capeo: true, such cases exist. However, I do not see it as a reason to anchor that to the person either (it is rare, but the editor may be cleared). For cases where that is however not the case, naming those after one or two editors is just wrong. --Dirk Beetstra T C 17:32, 23 July 2017 (UTC)
- That seems a false dichotomy to me. It assumes the issue is not the editor, their behavior, themselves. Particularly in admin conduct cases the issue is very much the editor. Capeo (talk) 17:22, 23 July 2017 (UTC)
I understand the concerns about case names. However, "the discussion will focus unduly upon the user who's mentioned in the request and the title" doesn't seem to be a problem on other parts of the project. For example, lots of threads on AN or ANI are opened by User:X with a title such as "[Type of misconduct] by User:Y." The editors who comment on AN/ANI don't seem to be unduly swayed by the initial post or the title of the thread, and often will post a comment such as "User:Y's behavior isn't the problem here" or "User:X's behavior is the problem" where warranted. (Indeed, "WP:Boomerang" is almost a cliché on some of these boards.) I understand that being named in an arbitration request is more serious than being named in a garden-variety noticeboard thread ... but still, I am not sure why the arbitrators would have more trouble looking past a thread's or case's title or origin and focusing on its merits, than do the community members who comment on AN/ANI.
A more general comment here. ArbCom strives for both procedural and substantive fairness in every case, and that is true regardless of whether the caseload at a given time is large or small. But it's still important to bear in mind that overall, ArbCom plays a much smaller role on the project than it used to. Ten years ago, ArbCom was deciding 100 cases a year, five years ago it might have been 30 cases, but this year we are on pace to decide maybe five full-fledged cases, if that. I appreciate the thoughts from anyone who has suggestions for improving the fairness and processes for the cases we do decide ... but attention shouldn't necessarily focus just on ArbCom case procedures as the most important ones on the wiki, when there are (for example) a thousand threads on other noticeboards for every case that comes to arbitration. Regards to all, Newyorkbrad (talk) 17:36, 23 July 2017 (UTC)
- @Newyorkbrad: thank you for your thoughts, but this is a WP:OTHERCRAPEXISTS argument. The example the arbitration committee sets has much more impact than a thread on AN(/I). And I do agree, also there anchoring plays a role, and it certainly has an effect there as well (I could even argue that it is a tactics in some cases). It is however still not an excuse for ArbCom to keep this practice. The contrary, I would say. --Dirk Beetstra T C 18:08, 23 July 2017 (UTC)
@Capeo: Once upon a time, to reduce redundancy and improve the overall presentation of case evidence, I suggested having a designated scope for a case and co-ordinators for each major viewpoint, who would collect evidence and feedback from interested parties and put forth a concise, well-reasoned presentation. I don't believe, though, that analysis should be limited to solely those who have contributed evidence: the skills involved are different, and there can be an advantage in having someone who wasn't involved in gathering evidence analyzing it. isaacl (talk) 18:57, 23 July 2017 (UTC)
- As I understand it, this function of collecting and summarizing evidence was originally conceived as part of the clerks' duties, when the idea of clerks was new and shiny and when the caseload was ten times as high, but this proved controversial due to concerns that the summaries might miss important issues or be distorted by non-neutral summarizers, and the idea eventually fell by the wayside. Given the current low caseload, I don't think a summary/filtering step by others is likely to be all that useful, because I generally read the evidence as it comes in anyway.
- I think I was spoiled by my early experience as an arb, because the first case I worked on had excellent evidence and a useful workshop, and most are... not so much like that. Personally, I'd be skeptical of people self-selecting into a role of analyzing, but not collecting, evidence. Let's face it, gathering diffs is tedious work. It's usually not all that interesting, except for those instances where you dig through a conversation you found especially irritating and get irritated all over again. It's hard to turn your diff-digging exercise into a concise and convincing presentation of evidence. "Analyzing", though, could be awfully fun - let someone else do the boring part and then you get to
soapbox about your personal opinions without the need to show any evidence that you've, um, read the evidencehelp out the arbs by pointing out the juiciest bits of evidence against your wiki-enemyahem, I mean share your thoughtful and careful analysis. There is of course nothing stopping anyone from sharing their analysis as things are, and sometimes this is very useful - but I admit to a heightened level of skepticism when I see someone with big opinions and no skin in the game. - To go back to this "anchoring" business, it unfortunately seems to be an unfalsifiable hypothesis as far as I can tell. The claim is that cases named after parties are disproportionately likely to sanction the title party, but the statistics are inadequate to support that claim, and there are several recent counterexamples. The facts at hand are that the original case titled "Magioladitis" actually offered remedies primarily about the broader issue uncovered during the course of the case, and yet somehow this is still evidence of "anchoring" and reflects unfair influence of the case name. This thread is making baby Popper cry ;) I actually agree with the underlying point that cases are stressful and unpleasant for the parties and a good result relies heavily on the quality of the evidence, but I'm unpersuaded by the proposed solutions so far. Unfortunately I haven't yet thought of any better ideas, either. Opabinia regalis (talk) 01:40, 24 July 2017 (UTC)
- @Opabinia regalis: Of course it is a falsifiable .. just stop with it and gather data. No-one of you has given any reason why case should be named after what is perceived to be the main named party. Take away the preconception. --Dirk Beetstra T C 02:17, 24 July 2017 (UTC)
- I don't think I am alone in having difficulty following all the revisions to evidence as it comes it (the presentation of diffs by the MediaWiki software is rather poor), or finding it wearisome to read the same points made over and over again. Multiply this by all the people trying to follow a case and, as I stated in my proposal, a lot of time can be freed up for other Wikipedia tasks if the evidence can be presented more concisely. I suggested that those supporting each major viewpoint could select anyone they trusted to summarize their point of view, and all of the raw data would be visible to confirm that the summary was accurate.
- Personally, I don't believe providing evidence is the sole way to have skin in the game. There is a value in going through the presented evidence and figuring out where it is leading in the analysis of evidence section. Proposing remedies without having gone through an analysis exercise is, for me, a bigger concern. Commenters come up with these odd principles that are highly specific to the particular case as a proxy for laying out a line of reasoning in the analysis section. It would be useful to be able to go through the various links in a chain of logic to examine their validity and relevance. isaacl (talk) 03:09, 24 July 2017 (UTC)
This is a really interesting thread and there is much sense being talked here, which is refreshing. Discussions of open-ended proposals onwiki is a really difficult process as we all know. So I suggest that Arbcom appoint a small number of editors, (say 4: an Arb, 2 non Arbs and a former Arb), to look into it for them, and see if they can come up with a system that avoids the inherent bias problem and yet allows the Committee to find it easy to get to grips with. I think it'd be brilliant if you asked someone who's been at the (shall we say) rough end of a case to be involved. --Dweller (talk) Become old fashioned! 14:26, 24 July 2017 (UTC)
Responding to evidence in an arbitration case
All cases named after an editor are show trials. Even if you refuse to believe the case is irrecoverably tainted by the name of it as I laid out here, even if you choose to refute science that shows the bias is unavoidable, the case is still a show trial. This is because the "defendant" (the person named) is limited in their evidence to 1000 words and 100 diffs, per Wikipedia:Arbitration/Guide_to_arbitration#Evidence. This limitation is a hard ceiling regardless of how many people are parties to the case. Because of that, you end up with cases like this, where 61 people submitted evidence against the named party. But, the named party is still limited to only 1000 words to rebut that Himalayan mountain range sized evidence. The bottom line here is that Magioladitis is screwed. No matter what he does during this arbitration process, the table is dramatically tilted against him and there is nothing...nothing...he can do to stop it. As many people have been saying for years now, its becoming increasingly worthless to attempt to defend yourself in arbitration cases. Once you're named, you're done. The very naming and structure of cases sets up a free for all feeding frenzy with the named party as victim, with no hope for them defending themselves. I'm confident this will be followed up with comments saying this isn't the case. But, the evidence screams otherwise. --Hammersoft (talk) 15:03, 19 July 2017 (UTC)
|
Hope you don't mind me copying this down here, but I've been thinking on this a bit and would like to continue this part of the conversation. I couldn't find a great place to do it in the above section without disrupting the conversation there about case names, so I started a new section and copied the relevant bit of conversation. @Hammersoft: I'm curious if you have thoughts on how to ease the difficulty of a named party (or multiple) having to respond to statements/evidence/etc. presented by a potentially large number of people. I have generally (but not always) found that complaints by many people often cover the same concerns, and so folks can respond to the concerns rather than the individual comments, and request extensions when needed. That said, there are exceptions to that where it's clearly overwhelming, and I'd love to hear your thoughts. GorillaWarfare (talk) 01:39, 21 July 2017 (UTC)
- @GorillaWarfare: (I hope Hammersoft does not mind me commenting) in the case request phase define a scope first, define the issues that resulted in bringing the case there. Constrict the case at that time, actively discuss that. Ask questions and interact with the people bringing forward statements to define the issues/scope. When Arbitrators are confident that the issues have been identified they vote to accept or decline to look into the defined issues within the defined scope (or they decline because there are no issues to define (storm in a glass of tea), or it is too early for a case - issues do not need to be defined to decline). Basically, you accept to look at the defined issues within the defined scope. Arbitrator work is done when the case is accepted (except if they need to adapt the list of issues, widen or restrict the scope - they need a motion for that). Clerks will then define a strictly non-personal case name which fits with the defined issues (even when there is clearly an issue with one editor).
- The evidence phase then should address the issues (strictly stay within the scope) that the committee agreed upon to look at in the case. Anything outside of the arbitrator defined scope should be scrapped by the clerks (editors, parties, arbitrators and clerks can bring amendments to the initial issues/scope, but arbitrators have to accept those by motion). The evidence phase is then having different headings for the different issues. (random word/diff number choice:) Each editor is allowed 250 words and 50 diffs of 'statement', 'evidence', 'defense', or 'rebuttal' per issue (heading would be 'evidence provided by <editorname>'). Parties named in the case similarly have a word limit of 1000 and 100 diffs per issue. Parties that have been explicitly named in the issue at hand have on top of their per-issue quota (the 1000 words/100 diffs) 150 words/50 diffs rebuttal/commenting 'against' each editor who have provided evidence/defense/rebuttal within said issue (basically, you have 1000 words on that issue, and if there are 15 non-party editors providing evidence, you can rebut yourself with 15 rebuttal statements of 150 words each).
- You've got 14 days of evidence phase (you could contemplate '.. for the first 25 editors in total who have provided evidence/defense/rebuttal, and 7 days for every next 25 editors who have provided evidence/defense/rebuttal; i.e. 21 days when there are 37 participants and 28 days for the somewhat extreme 61 participants above). The evidence phase is now closed for non-named parties, but named parties have then a right on another 7 days to provide defense/rebuttal, if requested). Then the usual workshop and final decision. --Dirk Beetstra T C 09:31, 21 July 2017 (UTC)
- Beetstra makes some excellent comments. For my part, as I've long argued for in RfA reform and other areas where people come up with solutions before identifying a problem, it'd be nice to identify and clarify the problem. Once we fully understand the problem, and we're sure it is one, then we can start discussing potential solutions. I cited the TRM case for two reasons; I was aware of it and I knew from being aware of it that it was extreme. 61 is a huge number of respondents. Is that normal? Is that in six sigma range? I don't know. How much is "too much"? We don't know and haven't really asked that question. Personally, in a 1:1 scenario, it's fair. As soon as you start to drift from that, it begins to be unfair. 2:1 might not seem bad, but it is unfair. The :1 person needs to do a lot more work to defend themselves if the 2: people are not repeating what the other said. Get to 10:1 and it's a nightmare I think.
- I would like to strongly echo and agree with what Beetstra said about defining scope. I argued for this in the TRM case and this was done by ArbCom (thank you!). The scope can help limit the burden on a title named party, but obviously doesn't eliminate it as we saw with the TRM case. But, non-scope defined cases become witch hunts with respondents filling the evidence page with enormous quantities of material. Semi-randomly picking another title named case, this one from 2015; 15 respondents giving evidence at Wikipedia:Arbitration/Requests/Case/Collect and others/Evidence. Evidence spanned three years of editing. I think it's an overwhelming witch hunt. You need a legal team to defend yourself. --Hammersoft (talk) 15:16, 21 July 2017 (UTC)
- The scope is important. I was in one where someone brought up completely unrelated stuff from ~5 years / 40,000 edits back that I did during my first few months as an editor when I was young and stupid / clueless and made all kinds of mistakes. And it was allowed to influence the case as if it was germane and current. North8000 (talk) 20:58, 21 July 2017 (UTC)
- I think it would be worth considering how to improve where responses to evidence should go. Typically in cases, parties can potentially respond to evidence presented by other users in: (1) their own sections of the evidence page (where there are obvious limits on length), (2) the "analysis of evidence" section of the workshop page (which is often left unused, and looks like something that could easily be overlooked), (3) in comments sections of proposed findings of fact on the workshop page, or (4) on case talk pages (where I have seen Arbs tell parties not to put it there because it might be overlooked, and to put it on the workshop page instead). It would be better to have one centralized location for all responses to evidence, and also to give accused parties significant leeway about the length of their responses. --Tryptofish (talk) 23:22, 21 July 2017 (UTC)
- I think the evidence page should be limited to just evidence, and not responses per se. The evidence being introduced can counter someone else's evidence, but no argument for this should be made on the evidence page. The analysis section should then be used to analyze both pieces of evidence and draw conclusions. In theory this limits discussion on the evidence to "is this evidence accurate", and keeps discussion on interpreting the evidence to the workshop page. (Commenting on the findings of fact is a bit late in the game, and talk pages shouldn't be a mechanism to work around the established places for commentary.) All too often, the discussion on interpreting the evidence spreads out over multiple pages, making it hard to follow. But... almost no one does this today, in spite of the pages being structured that way (maybe originally this separation was more firmly adhered to?). It would require discipline by the participants, aided by persistent clerking. I suspect many will balk at this, but maybe I'm underestimating how many people would be interested in this approach? isaacl (talk) 16:28, 22 July 2017 (UTC)
- I have also watched / commented in cases where the lack of a well-defined scope and / or action (by clerks ?) when discussions / submissions are well outside that scope would have been helpful. I am not at all convinced that using the case request to define the scope is a good idea because it presupposes that a case will be taken, and the purpose of the request is to decide that. Adding a phase to decide on the scope adds to the length of the process. However, having a defined scope also benefits the parties in being able to skip responding to evidence that has been deemed irrelevant. Having wondered about this before, I wonder if something like this might work:
- Divide the evidence phase into two separate one-week periods.
- The first is for evidence submission as now, but with the workshop page closed.
- When ready, the drafters then open the workshop phase for a section at the top only on scope, where the drafters outline what they see the scope as and list the questions the case will answer. Discussion on the workshop talk page is allowed.
- Drafters announce when second week of evidence will start once they are comfortable that the scope is set.
- The evidence page is re-opened. Clerks will close sections from the first period that are out of scope and anything new that is added that is out of scope. Parties now know what to respond to and what to ignore, and can hat / remove responses they've already made that are out-of-scope.
- New material that is out of the defined scope can only remain as unhatted evidence with consent of arbitrators or drafters and noted in the defined scope at the top of the workshop phase.
- I realise this adds to the burden of the drafters but it would also give the parties and others a much clearer direction on where the case is heading. Obviously this idea is just a basic framework that may or may not suit the management of cases from ArbCom's perspective. If the scope definition is posted quickly and is uncontroversial, it would only add a couple of days, and it would provide an outline for the proposed decision in the questions that the case seeks to answer. It would also help with the point Tryptofish made, to help confine responses to evidence from parties to a single place, the evidence page in the second phase of additions. Thoughts? EdChem (talk) 04:45, 22 July 2017 (UTC)
- @EdChem: The problem is, that some cases already de-rail at the request part, and de-rail further at an uncontrolled evidence phase (like now, see User:North8000's comment above, see currently open case request). The case is requested, the named parties are there, and the mud-throwing already begins - anything that can be used to accuse the named parties of wrongdoings is being posted in the request phase (again, as North8000 says and in the current open request, sometimes material that is completely, utterly unrelated and dating many many years back), and it will then obviously explode in the evidence phase. If you then define the scope after a week of evidence, then that means that that evidence is standing there for a week, may be out of scope, it is going to be a huge job to get rid of it, it already sticks in the minds of people, and you run the risk that the scope is significantly widened from what actually brought the parties to the request page.
- Arbs don't need to determine issues and set a scope for declining. I think that having a clear view of what the scope is about will also help Arbs decide whether to take the case. It does not happen often, but there are full cases which in the end totally die out because there in the end is not a significant issue. Further advantage could be that if the case develops to a "issues: X performs rather rogue blocks, X protects pages while heavily editing the pages; X is rather unresponsive to comments about his administrative actions; X is often rude to IP - our scope is going to be the administrative actions of X over the last 2 years, and his response to editors, especially IPs' - that X may consider to hand in his bit and a full case is not necessary (admins do turn in their bit during a case is not uncommon).
- See it as that ArbCom in the request phase decides what are the issues at hand that are worth taking, and that then becomes the scope. --Dirk Beetstra T C 05:10, 22 July 2017 (UTC)
- QED. --Dirk Beetstra T C 19:48, 23 July 2017 (UTC)
- @EdChem:I too believe the evidence phase should be reworked, but in my mind, it would be more useful to have the first week strictly limited to named parties. The drafters then review evidence, and at their discretion, can request supplemental evidence on specific issues from interested non-parties during an optional second week. The scope and named parties should be spelled out by the drafter before the first evidence week starts, in particular boundaries ("we will only consider issues dating back to before 2017 if they can establish a clear, consistent, recurring pattern."). If the scope cannot be set off the case request, the case should be closed. Fishing expeditions, or more accurately, throwing enough mud to the wall until something sticks, is a practice that should no longer be encouraged in arbitration. MLauba (Talk) 12:30, 24 July 2017 (UTC)
- @MLauba: thanks for reading and commenting on my suggestions. I am not sure that evidence only from named parties would be a good idea. In the present case, it would limit evidence to one editor who will focus on whatever seems most problematic to him or her, yet potential desysop cases usually have evidence of multiple / repeated examples of problematic actions, and covering it all puts a lot of responsibility onto one editor. I presented evidence of competence issues in a case which were obvious to me as a chemist which the named parties might not have recognised – and ArbCom even included it in the PD, so it was useful to them. Of course, unhelpful evidence is an ongoing problem. Do you think asking clerks to more aggressively hat / collapse evidence outside the scope would be helpful and practicable? EdChem (talk) 14:05, 24 July 2017 (UTC)
- @EdChem:, you raise good points - I also like Masem's notion below to limit evidence to involved editors at the case acceptance stage, although that may possibly lead to even longer walls of texts for the requests themselves. Possibly more nuance would be needed: first separate inter-user disputes (week 1 only for parties) from desysop requests (week 1 for parties and "involved" per Masem). Second, get people who want to weigh in to signal so clearly, ideally at the request stage ("Although not involved, I am able to present technical evidence that I believe demonstrates X has a competence issue in chemistry topics"). More aggressive clerking even at the request stage, focused on helping the future drafter narrow a scope would also help. MLauba (Talk) 16:10, 24 July 2017 (UTC)
- An idea from above about anchoring is that maybe that when the case scope is defined, a list of "involved" editors is created, at which only those editors can provide formal evidence to prevent the peanut gallery effect. There would need to be a mechanism to add named parties should evidence point to someone being necessary. Other editors, not party to the case, can still comment on the talk pages, but we should make sure these are not creating uncontrolled evidence blocks. Alternatively, require those not named to the case to be heavily restricted in what they can add (like, 250 words at most). What this emphasizes is that if you believe you have a lot to input to a case, you should be considering yourself a party to the case, and not ride in after the parties have been decided and drop a ton of evidence. --MASEM (t) 14:23, 24 July 2017 (UTC)
- The problem is, Masem, that evidence is often strongly overwhelmed by 'accusing' parties. Often the defense is only on the defendant, with at most help from one or two other editors. Restricting who is of interest to the case to only named parties may result in a more skewed situation than we often already have. --Dirk Beetstra T C 16:11, 24 July 2017 (UTC)
- If we clearly established that named parties are the only ones that ArbCom should be expecting evidence from, and the case scope is determined to be "one vs many" it would be critical that ArbCom and clerks limit the named parties to those that have the most involvement in the case. Instead of a "one vs many" that should get it down to a "one vs 2-5" or somewhere around there. And arguably in the latest case over the bot cosmetic edits, its clear to see who the most involved are. In the same manner, in such cases, there's also usually one or two sympathetic supporters, who, if they participate in the case request page, should be allowed to be named to help. (as I argued a few sections above, if a situation is clearly one editor against the rest of WP, that should have been caught by AN/ANI/AE, not gotten to ArbCom. If we're at ArbCom because a manner to deal with an editor can't come to consensus there must be those that are willing to speak up for that editor). --MASEM (t) 16:22, 24 July 2017 (UTC)
- Rather than splitting hairs over "is this person is involved enough to be one of the top few named parties", I prefer my proposal (from several years ago) of having the supporters for each major viewpoint agree upon a co-ordinator to present their evidence. It can be one of the supporters, or a third-party volunteer. This should help make the presentation of evidence more concise and cohesive, saving everyone time in reading, understanding, and analyzing it. isaacl (talk) 16:41, 24 July 2017 (UTC)
- I suggest Wikipedia:WikiProject Pro Bono - where accused editors can find other editors who can help them collect evidence and rebut accusing evidence during difficult AN/I or AN discussions, or ArbCom cases . --Dirk Beetstra T C 16:58, 24 July 2017 (UTC)
- See Wikipedia:Association of Members' Advocates. Note I am not proposing that a particular group be created, though I imagine that someone who does a good job in one case may get asked to help out with another (like how someone who successfully shepherded an RfC might get asked if they'd like to take on another). isaacl (talk) 17:39, 24 July 2017 (UTC)
- That might be useful in cases where there are clearly defined sides, but I would caution when there are a spectrum of views on a subject. You can likely group a few editors into a couple of supergroups, but there's also middle grounds. Less so when we're taking cases about one or two specific editors, but these cases do exist (ala the Troubles). --MASEM (t) 17:38, 24 July 2017 (UTC)
- Well, it would be the same issue if you're trying to limit the named parties to two to five persons. Rather than trying to decide from above who gets to speak, let the spectrum of persons decide who they'd like to speak for them. They can align themselves under the presentation of their preferred co-ordinator. isaacl (talk) 17:43, 24 July 2017 (UTC)
- I suggest Wikipedia:WikiProject Pro Bono - where accused editors can find other editors who can help them collect evidence and rebut accusing evidence during difficult AN/I or AN discussions, or ArbCom cases . --Dirk Beetstra T C 16:58, 24 July 2017 (UTC)
- Rather than splitting hairs over "is this person is involved enough to be one of the top few named parties", I prefer my proposal (from several years ago) of having the supporters for each major viewpoint agree upon a co-ordinator to present their evidence. It can be one of the supporters, or a third-party volunteer. This should help make the presentation of evidence more concise and cohesive, saving everyone time in reading, understanding, and analyzing it. isaacl (talk) 16:41, 24 July 2017 (UTC)
- If we clearly established that named parties are the only ones that ArbCom should be expecting evidence from, and the case scope is determined to be "one vs many" it would be critical that ArbCom and clerks limit the named parties to those that have the most involvement in the case. Instead of a "one vs many" that should get it down to a "one vs 2-5" or somewhere around there. And arguably in the latest case over the bot cosmetic edits, its clear to see who the most involved are. In the same manner, in such cases, there's also usually one or two sympathetic supporters, who, if they participate in the case request page, should be allowed to be named to help. (as I argued a few sections above, if a situation is clearly one editor against the rest of WP, that should have been caught by AN/ANI/AE, not gotten to ArbCom. If we're at ArbCom because a manner to deal with an editor can't come to consensus there must be those that are willing to speak up for that editor). --MASEM (t) 16:22, 24 July 2017 (UTC)
- The problem is, Masem, that evidence is often strongly overwhelmed by 'accusing' parties. Often the defense is only on the defendant, with at most help from one or two other editors. Restricting who is of interest to the case to only named parties may result in a more skewed situation than we often already have. --Dirk Beetstra T C 16:11, 24 July 2017 (UTC)
Comment what additional WP:Evidence did the 61st contributer bring to the Table?..if the answer is none can we just adjust WP:Evidence such that it enables his "contribution" to be reverted. If his evidence was pertinent, does this imply a deficiency in the 'Evidence' supplied by the previous 60 again...WP:Evidence or WP:NEOTHER can be tweaked/radically rewritten. Do it quickly; get it working Bosley John Bosley (talk) 17:48, 24 July 2017 (UTC)
- Having the interested persons select co-ordinators will enable the parties to filter the contributions themselves. If someone proposes adding evidence that's already there, the co-ordinator would not include it, saving everyone time. isaacl (talk) 17:52, 24 July 2017 (UTC)
Formalize timing to accept/decline a case
In Wikipedia:Arbitration/Requests/Case#Magioladitis 2, 3 arbitrators voted to accept before all named parties left their statements (2 of the votes were cast before one of the named parties even appeared to be online). A fourth Arbitrator commented (before the editor appeared to come online). This type of behaviour has in the past led to problems, where editors were railroaded - they found a full case had been accepted before they even came online. In any case, there is no reason not to give courtesy to all named parties to write a statement (not even being very familiar with the situation, or going to accept anyway), and early voting, especially accepting, can be perceived as rude and inconsiderate. I therefore ask the committee to formalize the following:
Arbitrators will not vote to accept or decline before 6 hours have passed after the last named party has written a statement, or before 72 hours have passed after the start of the case request, whichever of the two is earlier. If named parties are added after the initial request, both periods will restart. Arbitrators can at any time comment on the case (without prejudice towards accepting or declining), or ask editors who have provided a statement for more clarification
.
- 6 hours: to give named parties a couple of hours to re-think an initial statement and possibly amend it
- 72 hours: a couple of days, so it covers the time for an editor being offline for 1 or 2 days, but not making it indefinite. I hope that arbitrators will extend this period when they have been made aware that an editor is on a longer leave. --Dirk Beetstra T C 20:26, 19 July 2017 (UTC)
- I concur with this recommendation. Per Wikipedia:Arbitration/Policy#Ratification_and_amendment, I believe this counts as a formal submission for ratification. ArbCom, please treat it as such. Some years back, ArbCom was heavily admonished for the behavior described, and most arbitrators appeared to take the advice to heart and stopped accepting cases before all parties named had had a chance to respond. ArbCom now appears to be straying from that sage advice. Formalizing this will help to prevent this ArbCom and future ArbComs from behaving in the manner demonstrated in the recent case request. --Hammersoft (talk) 20:32, 19 July 2017 (UTC)
- If you're saying this is a proposal you're submitting, you're going to need to find 98 more editors who agree with you... I think the Committee acting quickly to sanction an editor would be unacceptable. A Committee acting quickly to examine an issue, on the other hand, is commendable. ~ Rob13Talk 21:17, 19 July 2017 (UTC)
- @BU Rob: the first don’t have the courtesy to await all statements, but still have to wait for the last Arb to accept and get the majority. Plus an additional 24 hours after that. This will in hardly any cases delay any start of the case, user:Mkdw, user:Ks0stm and user:GorillaWarfare could have examined the issue at the same speed, but accepted Magioladitis 2 48 hours later and still not cause a delay. It may not matter to Magioladitis, it becomes different when you wake up next morning, find an accepted case in front of you and only have to wait to have it formally opened. Even less reason to provide counter evidence I guess. By the way, I can also see reason for the committee to act quickly to sanction, and they do that where needed. But for those cases where there is no deadline, basic courtesy?
- I hope that the arbs will take this in consideration, even without the 100 votes.. As Hammersoft says, the committee has in the past been admonished for this, and I am yet to see major downsides. –Dirk Beetstra T C 21:43, 19 July 2017 (UTC)
- Rob13, what you are stating is partially correct, but not the first step in making amendments to the policy. Per Wikipedia:Arbitration/Policy#Ratification_and_amendment, ArbCom has to approve it first. If they approve it, then 100 or more must vote support. If they fail to approve it, then 100 need to petition for it to be voted. Either way, the first step is ArbCom's approval (or lack thereof). --Hammersoft (talk) 00:21, 20 July 2017 (UTC)
- I don't believe this proposal consists of a change in policy, so fortunately the policy amendment procedure doesn't have to be applied. The Arbitration Committee can decide to change its procedures unilaterally, should the community convince it of the need. isaacl (talk) 00:29, 20 July 2017 (UTC)
- I'd rather see it as policy honestly. Most arbitrators abided by it for a few years, but now they are slipping out of the habit. --Hammersoft (talk) 00:40, 20 July 2017 (UTC)
- Formalized written procedure would be much more difficult to let fall by the wayside, and can be enforced by clerks, so I think it is sufficient and much easier to get put into practice. isaacl (talk) 00:51, 20 July 2017 (UTC)
- Formalized policy has standing with the community once passed. ArbCom's policies have no formal standing with the community, and can ignored by them (and have been). --Hammersoft (talk) 01:24, 20 July 2017 (UTC)
- Formalized written procedure would be much more difficult to let fall by the wayside, and can be enforced by clerks, so I think it is sufficient and much easier to get put into practice. isaacl (talk) 00:51, 20 July 2017 (UTC)
- I'd rather see it as policy honestly. Most arbitrators abided by it for a few years, but now they are slipping out of the habit. --Hammersoft (talk) 00:40, 20 July 2017 (UTC)
- If you're saying this is a proposal you're submitting, you're going to need to find 98 more editors who agree with you... I think the Committee acting quickly to sanction an editor would be unacceptable. A Committee acting quickly to examine an issue, on the other hand, is commendable. ~ Rob13Talk 21:17, 19 July 2017 (UTC)
- I could definitely see personally adopting an approach like this, because I do see the points folks have made above about the downsides of commenting before all named parties have responded. A formalized version would definitely require a bit more thought, both about specifics in timing and about edge cases (parties being added later into the process, or parties responding to say "I'll respond more completely later", or similar). GorillaWarfare (talk) 23:59, 19 July 2017 (UTC)
- @GorillaWarfare: Regarding your two scenarios: editors who wish to write half of their statement now and half in 2 days (still within the 72 hours) are rare edge cases. The knowledge of the editor that the Arbs will not vote until 6 hours after their statement does give editors now also the possibility to make a statement in their sandbox or even off-wiki and post it when it is quite finished instead of half now, half in two days. And I would say that having these time limits in the arbitration policy would make it more likely that those parties are identified before the accept/decline vote starts (if X requests a case on Y, then it is likely Y to say 'but I think Z is a party here as well'), which would be before the voting started. The case that parties need to be added after (a good part of) is finished is yet another reason why early voting by Arbs is to be avoided, but actually a separate issue with arbitration cases (more related to the case naming issue). --Dirk Beetstra T C 03:44, 20 July 2017 (UTC)
- The above posts are right that this doesn't require an arbpol amendment as far as I can well. But I can't work up much enthusiasm for this idea, sorry. I dislike rigid bureaucratic requirements in general, and most especially when they're written as a reaction to some specific set of circumstances. What stood out to me was the comment above that case requests shouldn't be accepted too quickly, but quick sanctions were OK. That sounds to me like hyper-focus on recent events. IMO it's the opposite; a sanction outside of a case is exceptional, while deciding to have a case is just initiating an investigative process that may well end with a finding of "there's no there there" (e.g. the Hardy case last year). Opabinia regalis (talk) 05:46, 20 July 2017 (UTC)
- @Opabinia regalis: Quick sanctions are only OK in emergency cases (emergency desysops), similar to quickly block a runaway bot.
- And I agree, too rigid procedures are maybe not the best - but after the admonishment of a couple of years ago because of railroading editors the practice did stop. Now most of ArbCom is refreshed the practice starts again. I don't like rigid procedures, but you apparently need it, you have to be reminded of what happened in the past, and lack the courtesy to the parties. --Dirk Beetstra T C 06:27, 20 July 2017 (UTC)
- There is a flaw in the proposal, in my opinion:
Arbitrators will not vote to accept or decline before 6 hours have passed after the last named party has written a statement ...
(emphasis added)
- ArbCom get many requests that (for example) amount to content disputes and are never going to get accepted, initiated by editors who appear to believe that ArbCom is an early part of dispute resolution or that many of the steps can be skipped. Preventing quick votes to decline with advice that the issue is clearly not ripe for arbitration and pointing to more appropriate approaches to dispute resolution is not a good idea. Requiring the other party or parties to comment is a needless formality that burdens the accused editors and wastes the time of uninvolved editors who helpfully post to note that the request is not going to be accepted.
- It is also worth considering that there are cases that are going to be accepted irrespective of what anyone says. Recall the Eastern European Mailing List case, which ArbCom initiated sua sponte, IIRC. Suppose it had instead been initiated by a community member once the existence of the list was disclosed, and a bunch of editors named as parties... would ArbCom have been forced to wait days for every party to comment before making the obvious statement that the content of the list warranted investigation to see if there was problematic behaviour. EdChem (talk) 06:03, 20 July 2017 (UTC)
- Although declining of cases is less of a problem, there is also no problem with waiting after all parties have commented. The case has anyway to wait until no majority of accepting arbitrators can be found. If there is no need for a case, there is certainly not a hurry (which may exist for accepted cases), so there is absolutely no reason not to wait there. And if the full scope of the issue becomes apparent the opinion of the arbitrator may change (there are many remarks along the line of 'I want to hear more, but I am inclined to accept/decline'), and sometimes arbitrators change vote during the request phase because new information comes to light.
- Regarding the Eastern European Mailing List Case - there are 'emergencies', what I thought about are the emergency desysops. Also there it would be inappropriate to wait for the named party to respond before taking action. Those should be exceptions, and need a motion from the Arbitrators. For the emergency desysops, a remedy is first put in place, but then there should be a reasonable time for the named party to comment, as per suggestion, and a determination should be made whether a full case is needed or not, and whether the remedy needs to be kept in place or not. --Dirk Beetstra T C 06:27, 20 July 2017 (UTC)
- Much of the discussion above seems to suggest that cases are being opened too rapidly - but look at the current example: there is an unopposed super-majorty of the committee voting to accept the current case, but there is no open case. The procedures include an indefinite opening hold in criteria 4. — xaosflux Talk 11:46, 20 July 2017 (UTC)
- @Xaosflux: No, nothing of this suggests that cases are opened too quickly (that has happened in the past, that is the railroading that happened to at least 1 editor). I am afraid that you misunderstood the discussion. I would actually argue the other way around, cases are opened too slowly - I would prefer that the arbs !vote within 48 hours after the last named party put down his initial statement, we are very much passed that point (and then we still have a 24 hour period before the case actually get opened). --Dirk Beetstra T C 11:52, 20 July 2017 (UTC)
- There are some case requests that are wildly inappropriate and should be declined almost at once - even some where I wouldn't expect all or any of the parties to comment. It looks as though your suggestion would prevent that. Doug Weller talk 14:15, 20 July 2017 (UTC)
- @Doug Weller: I see the point, I can imagine that for wildly inappropriate requests, or requests by relatively new users who do not know what actually should come in front of ArbCom. and I do have the biggest problem with the accepting party anyway:
- There are some case requests that are wildly inappropriate and should be declined almost at once - even some where I wouldn't expect all or any of the parties to comment. It looks as though your suggestion would prevent that. Doug Weller talk 14:15, 20 July 2017 (UTC)
- @Xaosflux: No, nothing of this suggests that cases are opened too quickly (that has happened in the past, that is the railroading that happened to at least 1 editor). I am afraid that you misunderstood the discussion. I would actually argue the other way around, cases are opened too slowly - I would prefer that the arbs !vote within 48 hours after the last named party put down his initial statement, we are very much passed that point (and then we still have a 24 hour period before the case actually get opened). --Dirk Beetstra T C 11:52, 20 July 2017 (UTC)
Arbitrators will not vote to accept before 6 hours have passed after the last named party has written a statement, or before 72 hours have passed after the start of the case request, whichever of the two is earlier. If named parties are added after the initial request, both periods will restart. Arbitrators can at any time comment on the case (without prejudice towards accepting), or ask editors who have provided a statement for more clarification
.
- Better? --Dirk Beetstra T C 15:40, 20 July 2017 (UTC)
- @Beetstra: Not really, I can imagine wanting to shut down cases within hours where they are being misused. And although I understand fully your desire not to rush to accept, I'm still uneasy about a rigid restriction. Doug Weller talk 16:16, 20 July 2017 (UTC)
- @Doug Weller: rigid restriction, sure, but as I said above, ArbCom has been admonished for this in the past, and now forgotten that. If we don't nip this in the bud, we'll just be back here in some time as apparently this courtesy is not something that comes automatically. I mean, 3 accept votes before the party put down their statement? Look at it from your side - you're there, and ArbCom does not wait for your statement because they have already made up their mind. Do you really want me to believe that I trust the committee to neutrally assess the evidence put forward by Magioladitis there? Apparently some don't care about his what he has to say before the case starts .. --Dirk Beetstra T C 16:36, 20 July 2017 (UTC) (well, that did not work :-) : @Doug Weller: --Dirk Beetstra T C 16:37, 20 July 2017 (UTC))
- @Beetstra: I thought it was two, but I'm probably misreading a time. But this is an unusual case. Not only have we had one recently, this has come just after community discussions where Magioladitis was a participant, so it's not as though we came into it cold with our first (at least recent) view of the issues being the case filing. A situation that rarely happens. I don't think we can use it as an example. Doug Weller talk 19:01, 20 July 2017 (UTC)
- GorillaWarfare was the third to accept at 19:35, Magioladits was active by that time, but did not leave a statement until 21:07, both on the 16th.
- Nice how these things link together. Because we have the #anchoring of the casename to the subject, 3 arbitrators knew that he was the problem, and hence accepted because they .. knew the problem - you thought it was convenient to have cases named after persons so you could remember later that you had a case about that person already. So why wait for a statement of the defendant, right. This is then also a rare case because we already know that the defendant is wrong, and we don't have to wait for their statement, and also do not need to give them the trust that they will get a fair case and that ArbCom will actually listen to their evidence either.
- It indeed rarely happens. Well. It was common practice until it went excessively wrong and the ArbCom got admonished for throwing an admin in front of the bus. It stopped for some time indeed, but because it was not formalized, and arbitrators hence can do whatever they want, here we are back with this unrespectfull behaviour. --Dirk Beetstra T C 22:31, 20 July 2017 (UTC)
- Beetstra, do you have a link to this previous discussion you've mentioned? It was before my time and so I'd like to review it. Mkdw talk 23:11, 20 July 2017 (UTC)
- @Beetstra: I thought it was two, but I'm probably misreading a time. But this is an unusual case. Not only have we had one recently, this has come just after community discussions where Magioladitis was a participant, so it's not as though we came into it cold with our first (at least recent) view of the issues being the case filing. A situation that rarely happens. I don't think we can use it as an example. Doug Weller talk 19:01, 20 July 2017 (UTC)
- @Doug Weller: rigid restriction, sure, but as I said above, ArbCom has been admonished for this in the past, and now forgotten that. If we don't nip this in the bud, we'll just be back here in some time as apparently this courtesy is not something that comes automatically. I mean, 3 accept votes before the party put down their statement? Look at it from your side - you're there, and ArbCom does not wait for your statement because they have already made up their mind. Do you really want me to believe that I trust the committee to neutrally assess the evidence put forward by Magioladitis there? Apparently some don't care about his what he has to say before the case starts .. --Dirk Beetstra T C 16:36, 20 July 2017 (UTC) (well, that did not work :-) : @Doug Weller: --Dirk Beetstra T C 16:37, 20 July 2017 (UTC))
- @Beetstra: Not really, I can imagine wanting to shut down cases within hours where they are being misused. And although I understand fully your desire not to rush to accept, I'm still uneasy about a rigid restriction. Doug Weller talk 16:16, 20 July 2017 (UTC)
- @Mkdw: yes, I do have, I recall to have made a table somewhere storing timelines of requests for a number of requests. Problem is, at the moment I don't seem to have access to that (the storage server throws a 503).
- In any case it does not matter, does it? Even if you define this as a one-off situtation, I don't think that this is reasonable, sensible, civil - I find it downright inconsiderate, careless, and rude. None of you have answered to the deeper issue (which shows you have issues with issues): do you understand that when you don't wait for the statement of a named party that that does not give much trust that you will take that statement into consideration (obviously: you already did not), or will later, during the full case, take the evidence of said named party into consideration (in fact, it does not give much trust that you take the statements of any other editor than the OP into account, whenever you vote to accept ...). Whether or not it is liked by (some of the) arbitrators, seen that lack of decorum and common sense it is apparently needed to codify this. --Dirk Beetstra T C 07:59, 21 July 2017 (UTC)
- @Mkdw: I can't find the data I once collected. I do recall the case mentioned here about User:EncycloPetey (who woke up to a partially accepted case, and had subsequently a motion against him without ArbCom notifying him on his talkpage). Similarly we have this tally of 9/0/1/0 (technically accepted with 15 arbs, of which 1 recuse and 2 inactive) .. The first accept vote (that is heavily referenced by other arbs) of that puts quite some weight to Gnangarra's position, who nonetheless did not put down a statement at this time. Also JHunterJ did not put down their statement yet. That is not surprising seen the next diff, where it becomes known that Gnangarra has problems with editing (statement here, by then all parties have posted).
- More recent, also in the second case requestarbitrators did not have the courtesy of waiting, and 2 arbitrators were leaning to accept. Despite the early accepts arbitrators did not see reason for a full case in the end, and the case got declined. I think that shows how valuable it is to wait until full statements are there before deciding to accept. --Dirk Beetstra T C 15:17, 21 July 2017 (UTC)
- Just food for thought: I know that ArbCom is not a court of law, and Wikipedia not a bureaucracy - but America adopted the sixth amendment in 1789, 238 years ago, and I don't think that they were the first in the world. --Dirk Beetstra T C 15:25, 21 July 2017 (UTC)
- I think there is a way to write this, so there are exceptions for "out of scope" requests, emergencies (I think emergencies need no case request), a committee member's informal or formal requests/injunctions to any user, and just lack of response. "Regular order" of things, here, is not about bureaucracy, nor "rights", as much as it is respect, courtesy and fairness. This is especially so, given the "charge" if it is to be accepted is, 'user behaves so poorly the community cannot deal with it, so binding arbitration is needed'. Seems you should have time to hear from the user, and incidentally anyone of the community that disputes the claim that it is beyond community action, and be persuaded by them. I agree with those who say, that all that is required is for the ctte to adopt the 'regular order'. - Alanscottwalker (talk) 15:54, 21 July 2017 (UTC)
As I see there are only a few types of arbitration request, and the correct way to deal with them varies, and any change needs to accommodate all of them:
- Emergency action required
- Action should be taken as soon as it is clear what needs to be done and someone with the ability to do it is available. Very often additional discussion is required to look into what happened and why it happened and whether any sanctions are required. Waiting around for an arbitrary time limit will likely make the situation worse if it is at all controversial (e.g. if it's anything like the User:Yngvadottir situation).
- Clearly nothing for arbcom to do at the present time, e.g. content disputes, very premature requests (there are some where even talk page discussion has not been attempted), etc.
- These should be speedily declined so that dispute resolution is not held up unnecessarily.
- Disputes the community has recently and/or repeatedly failed to solve.
- Most arbitrators will be familiar with these issues and even those that aren't will see quickly from a recent (often immediately preceding) thread at AN, AN/I, or similar venue that make it clear the community cannot solve the issue. Almost everyone will agree that these need a case, even if they disagree about their preferred outcome. I don't see any benefit from waiting.
- There is a clear case to answer.
- Only when the full story is known will it become clear what if any action if any needs to be taken - the answer may be that the respondent is fully acquitted or it might be they are guilty of all charges (or somewhere in between, especially if there are multiple parties and not all of them are excellent communicators), but a formal case structure is needed to find that full story. I don't see any benefit to specified wait times in these instances.
- All other requests, e.g. where it is not clear what is happening or when it is clear that more sides of the story are needed to understand whether a case will help.
- These are the only ones where a wait time is required, but I can't think of any examples where the time is not already being given.
There are four types of people named as parties to a case:
- Those central to the dispute.
- Those part of a dispute, but not centrally. In most cases this includes filers who consider themselves uninvolved.
- Those on the very edge of a dispute. They have relevant information, but are not deeply involved. Maybe they walked away after failing to resolve it at an earlier stage, or moved on to other things to avoid the drama.
- Those who have (almost) nothing to do with the dispute and shouldn't be listed as parties but have been. Most common when cases are filed by someone who isn't very familiar with how arbcom works.
If there are waiting times then they need to accommodate all of the following situations, in each case for all types of parties.
- Editors who are not active at the moment, for an unknown reason (assume good faith)
- Editors who disappear rather than respond to questions (there is a reason not to assume good faith)
- Editors who are active elsewhere but are apparently ignoring arbcom.
- Editors who are temporarily unavailable and have made that known (publicly or privately), e.g. circa 2006 I was peripherally involved in a dispute that was filed at arbcom the evening before I was going to be abroad for a week without internet access (the person filing had no way of knowing this).
- Editors who acknowledge the request but note that they will give a fuller response later.
- Editors who indicate that they will not be participating in the case by their own choice (iirc. The Rambling Man did this).
- Editors who are deliberately trying to disrupt and/or delay case (for whatever reason).
In the entirety of this thread I've not seen any attempt to address almost all of these. Thryduulf (talk) 16:37, 22 July 2017 (UTC)
- @Thryduulf: Regarding your first 5 points - I agree that there are cases of type 3 and 4, where there is maybe no benefit in waiting. However, that argument games all other possibilities. Arbitrators can always argue that they are familiar, so can vote before editors put down their statement. That is what basically happened with 'magiolatidis 2' (a declined request), two arbitrators voted early basically indicating that they knew the situation (anchoring?) and hence accepted. Although I agree that there may not be reason to wait, I see no problem to wait there either (it is not a type-1 emergency). I think that the declined second Magioladitis request is somewhere in the case 3/4 story - although 2 arbitrators accepted early, the case got declined in the end.
- The point is, I do not see any reason not to wait. Most cases do not open before 72 hours after being filed (the currently open request is open for more than 6 days) - the first arbitrators may be there within a couple of minutes, to get a full list often takes 2 days, plus 24 hours waiting time. There may not be any reason to wait, I do think that it is not a problem either.
- (I've taken the freedom to number your last points): I think I do address most of those: (1), 2, 3, (5), 6 and 7 - I state '6 hours after an editor left a statement, or 72 hours after the initiation of the request, whichever is earlier'. If an editor has not responded after 72 hours the case can be accepted (of course, that is then still to the discretion of the arbitrators, they can wait longer if they wish but not shorter). Case 1 is what I hope to catch in that AGF. Someone may not edit in the weekend, and only see it on Monday morning while the request was filed on Friday at 5 .. (72 hours may be short). 4 is an exceptional case - if someone posts that they have something else to do, and will post in 4-5 days, then I believe that it is courteous of the arbitrators to extend the 72 hours to 120 hours at least (note that this happened and was completely ignored a couple of years ago - the editor commented that they would need a couple of days with what I think was a very good reason when the tally was 2/0/0, it was 6/0/0 when they finally posted their statement 4 days later). Number 5 is a bit of a version of 4, and it depends on the wording there of the extension request there. Per WP:AGF, one could reset the 72 hours counter to that point in time. In most cases however, the last statement is well posted within 6 hours, and the case can be accepted within 12 hours after initiation of the request. That is also well within the time where most cases do gather enough accept votes. I do remain, I think it is rude not to wait for all statements, and I know editors who have expressed that feeling when they posted their statement (if there is no reason to wait for their statement, there is also no reason for them to post a defense in the evidence stage, so you can just sanction by motion and be over with it). --Dirk Beetstra T C 17:11, 22 July 2017 (UTC)
- Your last sentence seems to show a fundamental misunderstanding of the difference between the request phase, the evidence phase and the decision phase. The request phase is only about answering very basic questions - Is there are dispute? (if no, decline) Is it within Arbcom's scope? (if no, decline) Has prior dispute resolution been tried? (if no, decline unless there is a good reason why not) Is there a case to answer? (if no, decline), Is a case the best way to resolve the dispute? (if no, decline or handle by motion or some other means). That doesn't always require statements from all involved parties. Guilt or innocence is not determined at this stage, so not being able to post a defence is not relevant - that's for the evidence phase. Thryduulf (talk) 18:32, 22 July 2017 (UTC)
- @Thryduulf: No, I am not mixing those up. What I mean is, that if arbitrators do not have the decency to wait for a statement in the request stage, do not have the decency to consider my input there, what does that party then gain in the evidence phase. I can see that sometimes there is no reason to wait all statements - it can indeed be clear from the beginning, what I am talking about is basic decency. And what use is there in early accepting before all named parties have put down their statement. What do you gain? Nothing - you'll have to wait until there are enough Arbitrators who accept, and then an additional 24 hours anyway. The case that would have been Magioladitis 2 started of with 2 accept votes before Magioladitis put down his statement. After the statement that clarified the point several arbitrators described the issue as trivial and trout-worthy, and the case ended up being declined. So yes, I believe there are cases where there is no reason to wait, I also believe that there are cases where there seems to be no reason to wait, while it turns out that there are (I should further update my case-request-statistics - I keep coming with the same examples). --Dirk Beetstra T C 20:09, 22 July 2017 (UTC)
- Your last sentence seems to show a fundamental misunderstanding of the difference between the request phase, the evidence phase and the decision phase. The request phase is only about answering very basic questions - Is there are dispute? (if no, decline) Is it within Arbcom's scope? (if no, decline) Has prior dispute resolution been tried? (if no, decline unless there is a good reason why not) Is there a case to answer? (if no, decline), Is a case the best way to resolve the dispute? (if no, decline or handle by motion or some other means). That doesn't always require statements from all involved parties. Guilt or innocence is not determined at this stage, so not being able to post a defence is not relevant - that's for the evidence phase. Thryduulf (talk) 18:32, 22 July 2017 (UTC)
- The flaw in Thydulff's number 3 of his case types is it mixes up the arbitrator, as witness, and the arbitrator as arbitrator. While it is true an arbitrator may also be an incidental witness, in cases of non-recusal they need to act as much as possible with the fresh eyes of an arbitrator. This is an outline of a suggestion to deal with 'it all':
- Regular Order: Generally, arbitrators will not comment or pose questions on an arbitration request before __ hours have passed since notice of the initial filing. Arbitrators generally will not register decision on the opening of a case until ___ hours have passed. In cases deemed by the arbitrator as out-of-scope, an emergency, or in need of injunction, these timelines do not apply. In addition, any of these timelines may be informally extended, without notice of the extension. Clerking, redaction, over-sighting, administrator, or community action may occur, as appropriate, from the moment of filing. Notice of this provision will be given at the top of the request page.
- I foresee links being added. I left the times blank because I can see 6 and 18 or 12 and 24 or . . .. This would give a general time to respond (and think, even with 'I need more time' or 'I resign/retire') without allowing any user to hold anything important up but also gives flexibility. Alanscottwalker (talk) 14:37, 23 July 2017 (UTC)
- @Beetstra: I just wanted to briefly address the concerns about my voting to accept so quickly. Essentially, I did so because I had followed the developments over the couple of weeks beforehand and already knew from that that I would vote to accept a case. However, just because I've already decided to accept a case doesn't mean I've already decided what I think the outcome of said case should be, just that I think ArbCom should take a look at the issue at hand. Does that allay any of your concerns? Ks0stm (T•C•G•E) 16:00, 23 July 2017 (UTC)
- @Ks0stm: Thank you for the reply - that is in line with what Opabinia regalis said. No, it does not allay with any of my concerns, it does confirm my concerns regarding the anchoring. Questions:
- why could you not wait until all parties had put down their statement? Is there any specific reason why you accepted this fast? (what negative impact would it have on the case request if you would have waited to accept until after all named parties put their statements?)
- It appears you have made up your mind already, and already have decided that the editor needs to be looked into (diff:
More than happy to examine this under WP:ADMINCOND. Also will note that I'm particularly disgruntled with what at first glance appear to be attempts by Magioladitis to game the system
- we have not seen any evidence yet). How did you assess what are the exact issues that are at hand here, and whether WP:ADMINCOND is within the scope of the case that is needed to resolve these issues? - why should a named party have to believe that you will take the evidence presented by that named party into account, when you don't take the statement of that named party into account.
- Given that you are interested in ADMINCOND and gaming of the system by the named party - how do you expect the named party to believe that you will not show a strong confirmational bias towards any evidence that is presented to you regarding these interests, and which editors will now also supply to you since you have now in a way requested such evidence (I know this is on the edge of WP:AGF, and it is surely ironic, but parties who do have an interest in having Magioladitis sanctioned can now game the system by making sure there is ample supply of evidence regarding admin misconduct and gaming of the system).
- Thanks again. --Dirk Beetstra T C 16:32, 23 July 2017 (UTC)
- No offense, but I find this line of thinking wildly naive. ArbCom members are admins and editors. The idea that some don't, or shouldn't, watch AN or ANI and see developing situations that are going to land at their feet, and know they'll accept, is unrealistic. I'm all for a more structured phases of a case post-acceptance but I see no reason to make strictures on the accepting or declining a case. Capeo (talk) 17:11, 23 July 2017 (UTC)
- @Capeo: No offense taken, and you are completely right in that ArbCom members do edit and monitor/participate on AN and ANI (well, everywhere). It is however still not a reason as to why they have to accept a case request before all editors have put down their statements. Sometimes it doesn't matter, sometimes the statement of the last named party can make a difference even if you think you know everything before hand (I will repeat, the declined Magioladitis case, between the first and the current case got 2 accept votes before Magioladitis wrote a statement. After that a majority of arbitrators declined as the issue was too trivial or at worst trout-worthy). --Dirk Beetstra T C 17:19, 23 July 2017 (UTC)
- But you describe a scenario where the current system worked. A case that likely shouldn't have been taken wasn't. There's simply too many circumstances, many described above by Thryduulf, where hard strictures impede the process unnecessarily. Then to say, well, we can make exceptions for obvious gaming or trying to ignore a case, then you're just adding another layer of judgement as to when Arbs can disregard these strictures. Which invites more disagreement and distractions from the matter at hand. Capeo (talk) 17:35, 23 July 2017 (UTC)
- @Capeo: to me it shows that the first two arbitrators who accepted the case before the statement of all named parties did not have a complete understanding of the issues at hand. Similarly, an accepted case where all of the arbitrators accept before all named parties put a statement is also evidence that the system works. --Dirk Beetstra T C 18:15, 23 July 2017 (UTC)
- To me it shows that two Arbs were already aware of the background and thought it warranted a look. Similarly in any other situation where an Arb might accept prior to all statements. There's a fundamental difference in our views that revolves around the fact that WP, while a large community, is actually quite insular. It's near impossible for recurring conflicts to not be familiar to at least some Arbs. This isn't going before a judge who doesn't know you from Adam. To have the expectation of no prior experience with the conflict is naive. Obviously, if an Arb has had direct or sustained interactions with the named parties they should recuse. Otherwise, them having followed an issue from the outside and deeming it warrants a case is to be expected. With that I have to log off for the afternoon. Capeo (talk) 18:48, 23 July 2017 (UTC)
- @Capeo: to me it shows that the first two arbitrators who accepted the case before the statement of all named parties did not have a complete understanding of the issues at hand. Similarly, an accepted case where all of the arbitrators accept before all named parties put a statement is also evidence that the system works. --Dirk Beetstra T C 18:15, 23 July 2017 (UTC)
- But you describe a scenario where the current system worked. A case that likely shouldn't have been taken wasn't. There's simply too many circumstances, many described above by Thryduulf, where hard strictures impede the process unnecessarily. Then to say, well, we can make exceptions for obvious gaming or trying to ignore a case, then you're just adding another layer of judgement as to when Arbs can disregard these strictures. Which invites more disagreement and distractions from the matter at hand. Capeo (talk) 17:35, 23 July 2017 (UTC)
- Accepting a case is one thing. Accepting a case before the target of the case has had a chance to respond (and that's wrong too) and doing so with heavy tilting against the target before they've even had a chance to respond shows a heavily biased case. There's nothing wrong with an arbitrator being aware of what is going on on the project. There is a very serious problem when an arbitrator does not keep an open mind and instead states a case is already tilted against the target before the target's had a chance to respond and before there's been any evidence presented. No single arbitrator can know everything that is going on on the project. They should be waiting to draw any conclusions until the evidence phase is concluded, and at the very least if they have drawn conclusions to not be stating those conclusions publicly before the case has even begun, much less before the target has had a chance to respond. This should be basic common decency, whether it be to a brand new editor or as in this case to an editor who had been here for a decade and an administrator nearly as long. What ArbCom is required to do is to solve intractable disputes. They are not here to string up an editor. Especially in the acceptance phase, they should be evaluating if the dispute is intractable, whether other methods of dispute resolution have been done, and whether ArbCom could actually do something that would benefit the project. Instead, the cases are being accepted with an eye towards a specific editor and his specific behavior. As I've been saying for a while; it takes two to tango, and there's more sides to the dispute than just one. This should be basic common sense, but apparently not. I'm not calling out any single arbitrator here; this behavior is rife among both current and past arbitrators. --Hammersoft (talk) 17:51, 23 July 2017 (UTC)
- Actually, there's many times when "it takes two to tango" is an over-simplified platitude. There's no shortage of times when the fault lies only in one editor's actions as the inherent cause of conflict. And the point is getting to the cause. You also put the cart in front of the horse. Of course an Arb, if already aware of it through other venues, should mention in their acceptance why they are accepting the case. If they have suspicions, or direct experience, of admin misconduct from prior experience they should say it. Or would you rather no explanation of the grounds for which they are accepting the case and pretend they are unaware? It's far more honest to say, "I've been seeing this issues with this editor crop up over and over and I have concerns that only we can deal with this so I accept" rather than say it later or never at all. This isn't a court of law. This specific case, for instance, has gone beyond intractable, tested the patience of a large amount of editors, and has no other outlet of dispute resolution. An Arb acknowledging that in their acceptance should be no issue. Capeo (talk) 18:31, 23 July 2017 (UTC)
- If there is only one to tango - there is no reason for a case, there is no dispute - moreover, the community handles those without arbitration committee - the user is community banned. In all other cases, you can't be sure without seeing the evidence.
- Yes, I would prefer that (they pretend) they are unaware, impartial, and await statements and evidence. As I say on this page, America adopted the sixth amendment in 1789. --Dirk Beetstra T C 18:38, 23 July 2017 (UTC)
- Actually, there's many times when "it takes two to tango" is an over-simplified platitude. There's no shortage of times when the fault lies only in one editor's actions as the inherent cause of conflict. And the point is getting to the cause. You also put the cart in front of the horse. Of course an Arb, if already aware of it through other venues, should mention in their acceptance why they are accepting the case. If they have suspicions, or direct experience, of admin misconduct from prior experience they should say it. Or would you rather no explanation of the grounds for which they are accepting the case and pretend they are unaware? It's far more honest to say, "I've been seeing this issues with this editor crop up over and over and I have concerns that only we can deal with this so I accept" rather than say it later or never at all. This isn't a court of law. This specific case, for instance, has gone beyond intractable, tested the patience of a large amount of editors, and has no other outlet of dispute resolution. An Arb acknowledging that in their acceptance should be no issue. Capeo (talk) 18:31, 23 July 2017 (UTC)
- @Capeo: No offense taken, and you are completely right in that ArbCom members do edit and monitor/participate on AN and ANI (well, everywhere). It is however still not a reason as to why they have to accept a case request before all editors have put down their statements. Sometimes it doesn't matter, sometimes the statement of the last named party can make a difference even if you think you know everything before hand (I will repeat, the declined Magioladitis case, between the first and the current case got 2 accept votes before Magioladitis wrote a statement. After that a majority of arbitrators declined as the issue was too trivial or at worst trout-worthy). --Dirk Beetstra T C 17:19, 23 July 2017 (UTC)
- No offense, but I find this line of thinking wildly naive. ArbCom members are admins and editors. The idea that some don't, or shouldn't, watch AN or ANI and see developing situations that are going to land at their feet, and know they'll accept, is unrealistic. I'm all for a more structured phases of a case post-acceptance but I see no reason to make strictures on the accepting or declining a case. Capeo (talk) 17:11, 23 July 2017 (UTC)
- @Ks0stm: Thank you for the reply - that is in line with what Opabinia regalis said. No, it does not allay with any of my concerns, it does confirm my concerns regarding the anchoring. Questions:
@Beetstra: "Unaware" and "Impartial" are not synonyms, and indeed are independent of one another - all combinations are possible:
- Unaware and impartial - for example if there is a brewing dispute about Dutch botanists then I am unaware of it and almost certainly impartial regarding it.
- Unaware but not impartial - for example I'm not aware of a dispute in which user:HJ Mitchell is a central figure, but if there is one I'm not impartial regarding it.
- Aware and impartial - for example I'm aware of the ongoing disputes related to the Israel-Palestine topic area, but I am impartial regarding them.
- Aware but not impartial - I am aware of the dispute involving Magioladitis and I am not impartial regarding it (I would recuse if I were still an arbitrator). Thryduulf (talk) 19:05, 23 July 2017 (UTC)
- I know it is almost impossible to be unaware of situations before they come to ArbCom - they generally have been presented multiple times on AN(/I). That is unfortunately one of the problems with ArbCom that we have to live with. Impartiality is something different - that is something that I expect each Arbitrator to be (and I know they will recuse when they are not impartial). However if I see accept votes without waiting for the complete set of statements, then that impartiality (maybe a better word is neutrality) can certainly be questioned. --Dirk Beetstra T C 19:40, 23 July 2017 (UTC)
- Again, your conflating "there is something that needs investigating" with "the/a user named in this request is guilty", which is very far from the truth. In the same way that a magistrate remanding a case to a crown court does not mean the accused is guilty, just that there is a prima facie case to answer which needs to happen at a different venue. Thryduulf (talk) 20:23, 23 July 2017 (UTC)
- No, I am not conflating that. The problem is 'there is something that needs investigating', 'someone is guilty' and 'there is a user/are users named in this request' - 'let's investigate if the user is guilty'. And if that is your preamble - then you accept with "...the Committee needs to become involved to examine the suitability of <party> to continue holding the admin bit". If that is what you want to look at, then that is the evidence you ask to see, and that is the evidence you will get .. and you set the stage to conflate "there is something that needs investigating" and "the/a user named in this request is guilty" .. because you ask for only evidence to show that. --Dirk Beetstra T C 20:36, 23 July 2017 (UTC)
- Except that isn't how it works. The request phase is "is there something that needs to be looked at here? Please give evidence to support your position.". The evidence phase is "We're looking at this (e.g. X's suitability to continue to hold the admin bit). Please give relevant evidence about this (e.g. X's use of the admin bit and conduct as an admin)." No guilt is presupposed, evidence that shows good conduct can be just as relevant as evidence that shows bad conduct. Editors submitting irrelevant information is an issue, but it's a different problem and entirely unrelated to any arbitrary time limits. Thryduulf (talk) 23:26, 23 July 2017 (UTC)
- No, I am not conflating that. The problem is 'there is something that needs investigating', 'someone is guilty' and 'there is a user/are users named in this request' - 'let's investigate if the user is guilty'. And if that is your preamble - then you accept with "...the Committee needs to become involved to examine the suitability of <party> to continue holding the admin bit". If that is what you want to look at, then that is the evidence you ask to see, and that is the evidence you will get .. and you set the stage to conflate "there is something that needs investigating" and "the/a user named in this request is guilty" .. because you ask for only evidence to show that. --Dirk Beetstra T C 20:36, 23 July 2017 (UTC)
- Again, your conflating "there is something that needs investigating" with "the/a user named in this request is guilty", which is very far from the truth. In the same way that a magistrate remanding a case to a crown court does not mean the accused is guilty, just that there is a prima facie case to answer which needs to happen at a different venue. Thryduulf (talk) 20:23, 23 July 2017 (UTC)
- I know it is almost impossible to be unaware of situations before they come to ArbCom - they generally have been presented multiple times on AN(/I). That is unfortunately one of the problems with ArbCom that we have to live with. Impartiality is something different - that is something that I expect each Arbitrator to be (and I know they will recuse when they are not impartial). However if I see accept votes without waiting for the complete set of statements, then that impartiality (maybe a better word is neutrality) can certainly be questioned. --Dirk Beetstra T C 19:40, 23 July 2017 (UTC)
I disagree with that observation (doing the dishes every Sunday for your mother is totally irrelevant if there is ample evidence that you killed her, as is that you drive too fast - Wikipedia's system agrees with the first one, but there the second one is completely admissible), but it is not the question at hand. The question still remains - is there any reason in a regular case request why arbitrators should accept a case before all statements are in. I agree that it doesn't matter in some cases, my argument is that it does matter in others, and that you never know until you have a complete overview of the situation. --Dirk Beetstra T C 02:35, 24 July 2017 (UTC)
- I can't believe I've spent any time reading this. There is no obligation for any person identified as a party to actively participate in a case, let alone make a statement, so mandating that arbitrators wait for those statements is folly. Arbitrators should make their decision to accept when they individually feel sufficiently informed. Almost all of the statements made are simply a reiteration of exactly the battle that has brought the matter to the Arbitration Committee in the first place. Risker (talk) 02:43, 24 July 2017 (UTC)
- No, there is no obligation, but those case requests are rare where named parties do not make statements. That almost all of the statements are simple reiterations the battle is unreferenced. Case requests do exist where those statements do make a difference - you just presuppose that that is not the case for most of it and it does not matter to accept early. It also does not matter, as a basic form of decency and respect, to await all the statements. You really have to accept a case within the first hour, you can't wait a couple of hours? The case is not going to start that fast anyway? --Dirk Beetstra T C 02:57, 24 July 2017 (UTC)
Case study
@Beetstra: We now have a new case request where one of the parties is apparently ill and there is still an ANI thread. How do you think this should be handled? Keeping it opened until someone who is ill/away/whatever might in theory result in the case request being filled with attacks on that party while they can't respond. Doug Weller talk 12:52, 28 July 2017 (UTC)
- @Doug Weller: Finally - we get the point. The system is horribly broken. --Dirk Beetstra T C 13:22, 28 July 2017 (UTC)
- (e/c)Not answering for Beestra but attacks should be handled immediately by clerks (or failing them, really any admin, including arbitrators). You are also free to enter the ANI, if you think it necessary and helpful, while that might make you recuse in a particular case, that's true of all your actions elsewhere, and you can instead leave the ANI to others - if the ANI concludes that likely also informs any decision on a case one way or the other. -- Alanscottwalker (talk) 13:28, 28 July 2017 (UTC)
- (edit conflict) OK, now the real answer: In the same way that it is utterly inappropriate (not to say rude/inconsiderate/etc.) for Arbitrators to accept this case, it is utterly inappropriate for editors to pile on with statements. Is there a hurry in any direction? And it is the same for AN/I .. one of the parties is ill - either that AN/I now takes a flight in the wrong direction or it gets archived without being resolved. Do I have a solution, no .. but that is not a reason to continue such a practice. It is however a good reason to formalize it (and maybe not only for arbitrators, also for non-named parties).
- Suggestion: now that the ArbCom is aware of this, a clerk can {{hat}}/{{hab}} the case request until the last party is available (and arbs could comment 'since one of the parties is ill, I would suggest that all non-named parties and fellow arbitrators refrain from commenting/accepting/declining until the last named party is able to participate'. --Dirk Beetstra T C 13:34, 28 July 2017 (UTC)
- See also this remark from User:TheGracefulSlick. --Dirk Beetstra T C 13:37, 28 July 2017 (UTC) (ping: User:TheGracefulSlick --Dirk Beetstra T C 13:39, 28 July 2017 (UTC)).
- I saw it before I posted here. I wouldn't dream of accepting this case right now. I shall probably decline it as too soon, particularly with the principal party away at the moment. I do not think we have to wait for all parties to post to decline a case. Doug Weller talk 14:04, 28 July 2017 (UTC)
- If it is reasonably known one of the parties is not going to be able to respond in a timely manner (a week+) after the case was filed and before they had a chance to respond, this should be a matter of procedure to put the case on hold/dismiss it without a full vote, just simply to stop any potential piling-on or the like. --MASEM (t) 14:12, 28 July 2017 (UTC)
- @Doug Weller: Thank you. If you read through this thread, I have also considered that declining a case could be done (when done with consideration). I indeed think that in most cases there is no harm in declining. That doesn't solve the AN/I though ..
- Just to note, a situation similar to this has occurred in a past request for arbitration, and there 2 accept votes were cast before the situation became clear, and 4 accept votes were cast while waiting for the last statement. --Dirk Beetstra T C 14:20, 28 July 2017 (UTC)
- @Masem: - I think I say something like this above - If this becomes clear, the counting should start from the moment the last named party comes back online - I do agree that we should add 'Clerks should put the case on hold while said editor has not responded yet'. You might get a pile-on of negative statements (as is usual on case requests), which may have an effect on the final decision of the arbs.
- By the way, whether we are waiting for collecting decline votes, the case request should still be closed for non-parties to comment. --Dirk Beetstra T C 14:20, 28 July 2017 (UTC)
- I do wonder what effect it would have for clerks to prevent/remove any uninvolved editor from commenting on a case request until at least an initial statement from all named parties are put into play. I can see some issues, but just thinking aloud to see if this would make sense. --MASEM (t) 14:23, 28 July 2017 (UTC)
- @Masem: I was starting to ponder the same thought. 1) case request is initiated, 2) all named parties get a chance to put their statement (at this time Arbs can only decline or request info), 3) after all named parties others are allowed to put a statement, and Arbs are allowed to accept. It is a bit of an extension of what I initially suggested (minus the decline). And if Arbs need more info from non-parties .. they obviously have to wait. --Dirk Beetstra T C 14:27, 28 July 2017 (UTC)
- I should add, that for all editors who are actively editing on Wiki that the time for them to post a statement should be restricted to the initiation point of the request. Say we have one filing party and two other named parties. The two other named parties get XX time to comment. One of the two parties comments they are ill, then only THAT party gets YY time after they resume editing to post a statement. (now think about that this can be used to game the system - though that de facto means that that editor has left Wikipedia (they cannot edit until they make a statement, if they edit without making a statement, ArbCom can accept when XX time has passed). --Dirk Beetstra T C 15:22, 28 July 2017 (UTC)
- It might be helpful at that stage (after case was filed but before all parties had a chance to response) for a clerk or two to review any noted ANI/AN or other dispute resolution steps to confirm that there has been some attempt at the community resolving the matter before the arbcom filing. The one area that I perceive as helpful from uninvolved admins is the quick identification that not all community options for resolution have been attempted, allowing an unnecessary case to be close post haste without a lot of drama. If we do prevent uninvolved admins, then clerks should do some checking (though I don't expect them to be as up to speed, so they mainly can make a rather rough assessment if there's clearly been community discussion before the filing). --MASEM (t) 15:26, 28 July 2017 (UTC)
- There is no hurry though .. The filing party should already have listed that, and if they didn't the arbitrators should either ask whether (and which) other dispute resolution was performed, and/or decline based on that. It does not need statements of not-named users. --Dirk Beetstra T C 19:02, 28 July 2017 (UTC)
- It might be helpful at that stage (after case was filed but before all parties had a chance to response) for a clerk or two to review any noted ANI/AN or other dispute resolution steps to confirm that there has been some attempt at the community resolving the matter before the arbcom filing. The one area that I perceive as helpful from uninvolved admins is the quick identification that not all community options for resolution have been attempted, allowing an unnecessary case to be close post haste without a lot of drama. If we do prevent uninvolved admins, then clerks should do some checking (though I don't expect them to be as up to speed, so they mainly can make a rather rough assessment if there's clearly been community discussion before the filing). --MASEM (t) 15:26, 28 July 2017 (UTC)
- I should add, that for all editors who are actively editing on Wiki that the time for them to post a statement should be restricted to the initiation point of the request. Say we have one filing party and two other named parties. The two other named parties get XX time to comment. One of the two parties comments they are ill, then only THAT party gets YY time after they resume editing to post a statement. (now think about that this can be used to game the system - though that de facto means that that editor has left Wikipedia (they cannot edit until they make a statement, if they edit without making a statement, ArbCom can accept when XX time has passed). --Dirk Beetstra T C 15:22, 28 July 2017 (UTC)
Meta discussion on CU policy
An RfC has been raised for the CU policy, which would affect the en.wp policy. Meta:Meta:Requests for comment/Clarification to CU policy WormTT(talk) 11:07, 4 August 2017 (UTC)
I believe that I am owed an apology.
It has been nearly a year since the following happened: [ Wikipedia talk:Arbitration/Requests/Case/Michael Hardy/Proposed decision#Guy Macon's section ]. This week I was asked by another editor to comment on a matter that may end up before Arbcom, and I declined, even though I have evidence in the form of diffs that I believe has been overlooked.
The reason I no longer have anything to do with Arbcom is because I was wronged. Several arbitrators made a false accusation against me, and I have never received an apology or any response to my repeated requests that the arbitrators post evidence supporting their claims. Yes, the proposed sanction failed in this particular case, but I see no evidence that the arbitrators who originally voted for the sanction have ever shown any understanding that what they did was wrong.
So I decided to post -- once -- this request in an attempt to reconcile these differences rather than simply ignoring Arbcom from now on. This may have been a mistake -- we will see when we see what the reaction to this is -- but I felt that it was worth a try.
I am not alone in my opinion that attempting to impose Arbcom sanctions without any attempts to resolve the situation at any other dispute forum was wrong. See Wikipedia talk:Arbitration/Requests/Case/Michael Hardy/Proposed decision#Tryptofish's section, Probation section, NeilN's section, and Tarage's Section.
I do not have a lot of confidence that an apology will be forthcoming, and I have no intention of debating the issue further. Just be aware that a productive editor with 11 years experience and a clean record is still deeply hurt by something Arbcom did, is actively avoiding having anything to do with Arbcom, and feels that he is owed an apology. --Guy Macon (talk) 17:07, 11 August 2017 (UTC)
- Support. But I think the likelihood of an apology is about the same as the likelihood of me growing wings. Nonetheless, the Hardy case was a disappointing situation of ArbCom getting a lot of stuff wrong at first (which is not only understandable, but nearly inevitable in complicated cases that have to be understood all at once), followed by a very uneven pattern of some Arbs reconsidering their mistakes (which is the way things ought to work) and others circling the wagons (which may also be understandable, but which is nonetheless wrong). As I told Guy at the time, I had my own bad experience with wagon circling during the GMO case. Guess what ArbCom: people make mistakes, and the world does not end if one corrects one's mistakes. Now, having said all of that, Guy, I got over the GMO thing and put it behind me, and I hope that you will come to do likewise with what happened to you. It's only a website. --Tryptofish (talk) 19:08, 11 August 2017 (UTC)
- I commented extensively (probably too extensively) on that PD, as you may recall, Guy Macon. Individual arbitrators may admit to error now and then, but ArbCom as an institution offering an apology is virtually unknown, as far as I am aware. I remember at the end of my contribution, arbitrator OR was arguing that closing the case was desirable as nothing more would be achieved, and the fact that this left you feeling hurt and harmed was probably seen as collateral damage. You have a better chance asking individual arbitrators on their talk pages, I suspect, rather than here, and I would be astonished to see any collective statement from the committee, despite the MH case being awash with reasons to express regret. EdChem (talk) 15:00, 12 August 2017 (UTC)
- Very bad idea. I tried that when I had a legitimate request that Arbcom follow their own published rules[4] (At the time Arbcom was allowing Hardy to post 2331 words and to keep adding to that but limiting me to 500 words) and one on the arbs used that as ammunition to belittle my concern and to use the fact that I dared to contact individual arbitrators on their talk pages to justify attempting to sanction me.[5] --Guy Macon (talk) 15:41, 12 August 2017 (UTC)
- Comment A general comment: there needs to be more willingness on the part of those "in power," especially administrators, to admit error. There is a tendency to double down instead. If Arbcom was wrong then they should apologize, but that should be a requirement for administrators in order for them to retain their status. Figureofnine (talk • contribs) 15:23, 12 August 2017 (UTC)
- Join the club. I had a private apology from a ex-Arbcom member for what happened to me, but there'll never be a public explanation of what happened and why. The Wikimedia community has too much jockeying for position, and deals done behind closed doors, to ever be able to call itself transparent. The good news is that if this type of unaccountability carries on long enough, those that feel hard done by will vastly outnumber anyone else with an interest in our movement, which I guess means that if I live long enough even I can have a turn on Arbcom if I feel like it's worth the hassle. --Fæ (talk) 15:54, 12 August 2017 (UTC)
Personnel changes in the CheckUser team
- Welcome back to the team, Deskana! —DoRD (talk) 21:18, 14 August 2017 (UTC)
- --Jezebel's Ponyobons mots 22:33, 14 August 2017 (UTC)
- Thanks! I'm glad to be back. :-) --(ʞɿɐʇ) ɐuɐʞsǝp 23:19, 14 August 2017 (UTC)
Mdann52 reappointed an arbitration clerk
Response in a timely manner
There is discussion regarding a lack of response from ArbCom at User talk:Δ#Public Appeal. Please respond in a timely manner over there. Thank you. ···日本穣 · 投稿 · Talk to Nihonjoe · Join WP Japan! 20:32, 4 August 2017 (UTC)
- FWIW, we're just generally a little behind on email, not specifically this one. (Mid summer is a good time to enjoy your not-in-front-of-a-screen hobbies...) If you emailed arbcom recently, please be patient :) Opabinia regalis (talk) 06:56, 5 August 2017 (UTC)
- @Opabinia regalis: Oh, I have no doubt about that, based on the responses over on that topic I linked. I was just making sure ArbCom was aware of that discussion so they could respond if so desired. I can't remember the last time I emailed ArbCom about anything (if I ever have emailed them). I don't have a dog in this "fight". ···日本穣 · 投稿 · Talk to Nihonjoe · Join WP Japan! 05:43, 6 August 2017 (UTC)
A related aside
Greetings Committee and friends. Please indulge this related discussion which emerges directly aside the above mentioned:
Upon reviewing arbitration policy, and procedures, it occurred to me that the committee is remiss in their transparency obligations regarding ban appeals heard by private correspondence. To demonstrate, the correspondence mentioned above is an in progress ban appeal in accordance with the instructions given at Arbitration Committee/Ban appeals yet no such open appeal is shown where instructions indicate that it ought be.
The community is inherently interested in matters before the committee, including matters heard in private. And we are rightfully due a transparent accounting. I believe that should include a public disclosure that private discussion is underway, an invitation for friends to opine, and the "publicly made" statement of "detailed rationale", which policy requires for "decisions [made] related to cases".
I am not aware that the committee has ever fulfilled its charge of transparency regarding private deliberations, and challenge her members to repair their conduct in this regard. Thank you.--John Cline (talk) 14:49, 6 August 2017 (UTC)
- Thanks for your interest. Many of the appeals that ArbCom receives are clearly not going to be granted, including ones received from long-term abuse cases, blatant vandals, and the like. It would serve little purpose to require an ongoing public log of these. In addition, in some cases we discuss matters, including certain unblock/unban requests, off-wiki for user-privacy reasons and will not be able to provide a full public rationale for our decisions in some such matters. It's always been recognized that this is at odds with full-blown wiki transparency, but it's also always been part of the reason why we have an elected committee whose responsibilities include handling matters that should not be publicized on one of the most prominent websites in the world.
- That being said, I took a recent case in which a granted appeal turned out badly, as a reminder that we can profitably consult the community when we are considering granting an appeal and deeply private information is not involved. For example, that is what is happening in the Δ case discussed above. Newyorkbrad (talk) 21:32, 6 August 2017 (UTC)
- Thinking out loud here, perhaps the Committee could decide that when a case is not one of those that are clearly not going to be granted, a very simple notice that "an appeal from [username] has been received" could be posted, without giving any further details, and other users seeing the notice could then email the Committee. Then, when a decision has been made, another brief notice could be posted, saying either that the request was granted or was denied. This would provide a public record of start and finish dates without impeding privacy, and provide a way to receive community input. I think the Committee can use discretion about when such notice is or is not useful for a given case. --Tryptofish (talk) 22:13, 6 August 2017 (UTC)
- Something for consideration is the scope of such a proposal and its implications for the right to privacy. ArbCom was set up specifically to handle private matters. There has always been a group that has been uncomfortable with this idea and sought to undo some of the protections of such a process. Transparency and privacy are sometimes mutually exclusive of one another and the check and balance for misconduct is that the positions are term appointments as selected by the community. An RFC would need to be held on this issue because as I understand it, would be rewriting a signification fundamental mandate of ArbCom.
- In regards to the proposal specifically, I'm not convinced such a requirement for reporting is actually outlined in our policies and guidelines when it comes to private correspondence, but rather the legislative opposite. It does not make sense that these would need to be disclosed and reported to the community. Doing so would undermine a significant rationale for privacy. It does not advance transparency if the logs are about private items. Instead, it will instigate public questions about a private process. Questions will be asked and likely answers cannot not be given. What purpose would a public log serve other than to permanently implicate anyone on a record with anything more than negative implications (as we see with declined unblock appeals)? Why stain editors with this further [potential] embarrassment?
- There are certainly appeals that could use more community involvement and more notices, but that's different than what's been proposed here. Since this was nested under Δ's appeal, I would also contend that two weeks is not very long for an ArbCom appeal given the consistent queue. Transparency was not core issue nor was this editor being ignored (they had already received an email stating their appeal was being reviewed). Mkdw talk 02:21, 7 August 2017 (UTC)
- There are two major cases where I see a public log being potentially useful: when there is useful input to be gleaned from the community that we might not otherwise have, and when we find folks trying to appeal blocks on their talk pages, UTRS, IRC, etc. while simultaneously appealing them to the Committee.
- I agree, though, that there are significant privacy concerns. Publicly logging all appeals puts the appellant and the Committee in an awkward position where appeals may be declined but we can't divulge why. In my experience, when not given specifics on a case, people tend to assume blocks/bans/etc. were placed (or unblocks were denied) for very severe reasons. I worry that we might decline an appeal for a fairly benign reason (for example, they were caught socking four months ago and we asked them to wait six) and the logged declined appeal would lead people to believe they were harassing folks, running a giant sockfarm, etc... GorillaWarfare (talk) 02:42, 7 August 2017 (UTC)
- Thinking out loud here, perhaps the Committee could decide that when a case is not one of those that are clearly not going to be granted, a very simple notice that "an appeal from [username] has been received" could be posted, without giving any further details, and other users seeing the notice could then email the Committee. Then, when a decision has been made, another brief notice could be posted, saying either that the request was granted or was denied. This would provide a public record of start and finish dates without impeding privacy, and provide a way to receive community input. I think the Committee can use discretion about when such notice is or is not useful for a given case. --Tryptofish (talk) 22:13, 6 August 2017 (UTC)
- Since I've been on the Committee, I don't believe we've ever kept a public list of folks currently appealing bans. I'm curious about your statement that the ban appeal instructions say that notice of an appeal should be posted on the clarification and amendment page -- where are you seeing that? GorillaWarfare (talk) 00:01, 7 August 2017 (UTC)
- I too would be very interested in knowing which part of the policy is not being adhered as it's a very serious claim if correctly being interpreted and applied under its intents. Clarification on this matter will be helpful. For example, the Wikipedia:Arbitration/Policy#Transparency and confidentiality linked above does not provide validity for this claim. Mkdw talk 01:44, 7 August 2017 (UTC)
- The instructions I spoke of are within Wikipedia:Arbitration/Requests/Clarification and Amendment/Header where it says:
I see no other way for a banned user to "appeal for the removal of sanctions" except by email using [email protected] and no way to "use the Clarification and Amendment section for [such an appeal]" except some discernible form of actual use. I hope this sufficiently answers GorillaWarfare's question, and the things piqued in Mkdw as well. Thank you.--John Cline (talk) 04:01, 7 August 2017 (UTC)Use this section to request clarification or amendment of a closed Arbitration Committee case or decision. ... Requests for amendment are used to: ... appeal for the removal of sanctions (including bans). (my emphasis)
- Bans may include topic bans, interaction bans, or other (such as bans on automated edits and so forth). The instructions expressly provide that appeals by blocked and "banned" users should be done by email (and not using ARCA). Furthermore, several sections of Arb policy (including the instruction links listed in the ARCA heading for blocked and banned users) contemplate privacy and confidentiality: Wikipedia:Arbitration Committee#Communications and privacy and Wikipedia:Arbitration/Policy#Transparency_and_confidentiality. The instructions listed above do not require ArbCom to log these incidents publicly. Doing so would be a significant expansion of ArbCom policy and reduce the central aspect of privacy that surrounds ArbCom and its purpose. Mkdw talk 04:19, 7 August 2017 (UTC)
- Don't have much to add that wouldn't echo Mkdw's response above. GorillaWarfare (talk) 04:24, 7 August 2017 (UTC)
- Not to mention it would add to our turnover time. To whom it may concern, yes, I wish we would act more speedily, but we get a lot of email, we are a group of people, no one gets paid to be in charge and run the office--which would be an ungrateful task anyway even if it were a paid position. We do our best, but sometimes these things take a lot more time than you think, sometimes people we question are slow in replying, sometimes it takes us a long time to figure out who and what we are questioning. Since last week or so we have been discussing, internally and with our spiritual mentors, ways in which we can speed things up but it's not that simple. I myself have been the laziest arb on record for a month or two and I'm trying to pick up my share of the load--but what some of you don't know (there is no reason that you should) is that some of the requests we get just go on forever and ever, with never-ending and ill-pertaining jeremiads. Typically the first Arb that answers then gets a boatload of subsequent requests for information, all of which take time and all of which typically go nowhere. Usually we shaft Doug Weller with those jobs but I think he wizened up to our tactics. In short, we're trying, and I know we can do better. But the transparency thing, what you are proposing here is not a good idea nor is it based on policy: essentially this would turn us into another AN, it seems to me, or it would turn every request into an arb case. Drmies (talk) 15:54, 7 August 2017 (UTC)
- Thank you. I'd like to point out that my original posting was not written as a proposal at all. It was a statement of things I had observed; colored by my interpretation, culminating with a challenge to the Committee to self-repair. The trending opinion is that no repairs are in need though it rests upon a fantastical reading of Wikipedia:Arbitration/Policy#Transparency and confidentiality. Pardon my candor but I am befuddled that anyone could read that paragraph and conclude that it called for no transparency at all.--John Cline (talk) 20:36, 7 August 2017 (UTC)
- We are aware that your original post was not a proposal. We are simply stating that from an application point of the literal the wording of the policy, it does not support your original position. We are stating that a proposal would be required first to move the policy to such a place (something we would not be supportive of without the broad consensus of the community). Additionally, the argument has not been made that Wikipedia:Arbitration/Policy#Transparency and confidentiality calls for no transparency. That is not what has been said; we are pointing out that the ARCA heading instructions and the Arb policy conflict with your original position.
- For the purposes of this conversation and resolution, one point should be discussed at a time. Mkdw talk 23:13, 7 August 2017 (UTC)
- For the purposes of this conversation, one point is meant for discussion: either the committee is remiss in their transparency obligations regarding ban appeals heard by private correspondence; in accordance with Wikipedia:Arbitration/Policy#Transparency and confidentiality, or they are not? If they are, a number of things could clearly be done by existent policy, if they are not, then I am wrong, and ridicule is not misplaced. Thank you.--John Cline (talk) 04:20, 9 August 2017 (UTC)
- Thank you. I'd like to point out that my original posting was not written as a proposal at all. It was a statement of things I had observed; colored by my interpretation, culminating with a challenge to the Committee to self-repair. The trending opinion is that no repairs are in need though it rests upon a fantastical reading of Wikipedia:Arbitration/Policy#Transparency and confidentiality. Pardon my candor but I am befuddled that anyone could read that paragraph and conclude that it called for no transparency at all.--John Cline (talk) 20:36, 7 August 2017 (UTC)
- Not to mention it would add to our turnover time. To whom it may concern, yes, I wish we would act more speedily, but we get a lot of email, we are a group of people, no one gets paid to be in charge and run the office--which would be an ungrateful task anyway even if it were a paid position. We do our best, but sometimes these things take a lot more time than you think, sometimes people we question are slow in replying, sometimes it takes us a long time to figure out who and what we are questioning. Since last week or so we have been discussing, internally and with our spiritual mentors, ways in which we can speed things up but it's not that simple. I myself have been the laziest arb on record for a month or two and I'm trying to pick up my share of the load--but what some of you don't know (there is no reason that you should) is that some of the requests we get just go on forever and ever, with never-ending and ill-pertaining jeremiads. Typically the first Arb that answers then gets a boatload of subsequent requests for information, all of which take time and all of which typically go nowhere. Usually we shaft Doug Weller with those jobs but I think he wizened up to our tactics. In short, we're trying, and I know we can do better. But the transparency thing, what you are proposing here is not a good idea nor is it based on policy: essentially this would turn us into another AN, it seems to me, or it would turn every request into an arb case. Drmies (talk) 15:54, 7 August 2017 (UTC)
- The instructions I spoke of are within Wikipedia:Arbitration/Requests/Clarification and Amendment/Header where it says:
- I too would be very interested in knowing which part of the policy is not being adhered as it's a very serious claim if correctly being interpreted and applied under its intents. Clarification on this matter will be helpful. For example, the Wikipedia:Arbitration/Policy#Transparency and confidentiality linked above does not provide validity for this claim. Mkdw talk 01:44, 7 August 2017 (UTC)
- Historical note: I am afraid Mkdw is slightly incorrect in comments made above. Arbcom was initially created in 2004 to take over the responsibilities of Jimbo Wales in managing dispute resolution, including blocking and banning of longterm editors. The tools that relate to privacy (checkuser and oversight) were not even available then. The original oversight extension was instituted in 2006, checkuser in late 2006 - both about two years after Arbcom was created. The use of those tools was limited to technically knowledgeable arbitrators and bureaucrats (i.e., people who had already received strong trust support from the community), and they did not normally share info with the rest of the arbitrators. In both cases, at least part of the reason for the creation of the tools was to reduce workload on the very limited number of developers who had access to either oversight edits or to see the data that is normally reviewed by checkusers. I am sure people will notice the pattern - Arbcom has historically been the repository of responsibilities with single or few points of failure that directly involve concerns of the community - including but not limited to private or nonpublic personal information, for which Arbcom's actual responsibility is the appointment of checkusers and oversighters. There are very rare cases where other situations need to be segregated to that extent (I can think of some issues that related to users who were minors, usually before the WMF had even the precursor of the Support and Safety team), but a lot of it requires almost no responsibility on the part of Arbcom. Indeed, only on English Wikipedia does Arbcom have any such "responsibilities", with perhaps one other project having arbcom-appointed CU and OS. Risker (talk) 01:07, 10 August 2017 (UTC)
- Thanks Risker. No comments on oversight and checkuser since I hadn't brought those up, but what I meant "is set up" and not "was set up". My experience with ArbCom is only current. I was also probably being too casual and colloquial with "always" meaning during my tenure on Wikipedia (some ten years only). Presently, ArbCom's role, in regards to privacy, extends a lot further than appointing functionaries. While no editors hold any legal responsibilities on the English Wikipedia, there are community expectations associated with ArbCom. It has been my experience through our increasing interactions with WMF Legal and SuSa, that a fair amount of latitude is granted and often recommended we handle ourselves in alignment with these community imposed responsibilities. Regards, Mkdw talk 04:16, 10 August 2017 (UTC)
- I know this might go against the grain a bit, but considering how a recent unbanning exploded in the lap of ArbCom, have you ever considered posting a non-binding Request for Comment post at WP:AN or WP:ANI for select unban requests? After you have done your basic research and when you aren't quite sure. Unbinding is the key word here. You might find more information than you can dig up yourselves, and in cases like this, get a feel for the community you represent. For complicated cases on the fence, I can't help but think this would assist you, give some clarity, and reduce the chances of mistakes being made. One week should be enough. Dennis Brown - 2¢ 14:58, 11 August 2017 (UTC)
- @Dennis Brown: That's an interesting suggestion. We definitely do not want any similar explosions. Doug Weller talk 18:28, 11 August 2017 (UTC)
- That has been done at times in the past, and I believe it's what's planned again regarding the Δ (Betacommand) request. Regards, Newyorkbrad (talk) 18:31, 11 August 2017 (UTC)
- That sounds like a good plan. Most of the time it isn't an issue, and you can't always predict when it will be an issue, but I do think you are more likely to get a positive response from the community in these high profile cases when the community is allowed the chance to opine. As a whole, the community is pretty forgiving, so I think that favors the banned in some ways. Dennis Brown - 2¢ 23:55, 13 August 2017 (UTC)
- In many cases such a non-binding discussion would work, but in every case where the committee's decision goes the opposite way to the apparent community consensus you have the potential for a huge drama. For example, if publicly-available evidence shows editor A to be an innocent party, but private evidence (that is impossible to share publicly) shows that they were guilty of harassing another editor, the arbs will not unblock regardless of what the community thinks. Now imagine that the private evidence only comes to light late on in the public discussion when the overwhelming majority of commenters are in favour of an unblock. Now imagine if it is an editor who has been controversial in the past. No matter how often arbitrators say that a decision is based on non-public information there are some, sometimes very noisy, people who refuse to believe that the decision was anything other than abuse of process/a personal vendetta/a conspiracy/etc. I am not saying this non-binding discussion should not happen, just that if it does it needs to be framed in such a way that the chance of drama is minimised - for starters by making it explicitly and repeatedly clear right from the start that it is non-binding and explicitly saying exactly what that means. Thryduulf (talk) 15:32, 19 August 2017 (UTC)
- That has been done at times in the past, and I believe it's what's planned again regarding the Δ (Betacommand) request. Regards, Newyorkbrad (talk) 18:31, 11 August 2017 (UTC)
- @Dennis Brown: That's an interesting suggestion. We definitely do not want any similar explosions. Doug Weller talk 18:28, 11 August 2017 (UTC)
Kostas20142 appointed trainee clerk
Soap unbanned
- As a bit of a personal comment on this. My decision to support the unban was primarily because I believed that Soap adequately explained, in his appeal, why the previous behaviour occurred and the steps he's taken since then which would prevent a similar issue occurring again. Callanecc (talk • contribs • logs) 09:48, 19 August 2017 (UTC)
- Which is excellent news. I knew Soap just needed a break from Wikipedia. Glad to see him back. Kurtis (talk) 04:32, 28 August 2017 (UTC)
- Miniapolis the wikilinks to the two "community decisions" don't work. Ben · Salvidrim! ✉ 22:23, 4 September 2017 (UTC)
- Thanks, Salvidrim!; I've fixed them. GorillaWarfare (talk) 22:47, 4 September 2017 (UTC)
Enforcement log cleanup
I bring this up every 6 to 12 months, and it's still an open issue. It was decided several years ago to move the enforcement log info from the case pages to the then-new chronological pages specifically for enforcement logs, yet this has not been completed. The presence of old E-log data, stopping in 2014 or so, at pages like WP:ARBATC is highly prejudicial to anyone logged there. What it looks like is "these bad apples got sanctioned, and when they did all trouble in this area stopped", when this is virtually never true. I've directly experienced problems because of this (repeatedly), and this is something like the fifth time I've requested this cleanup in one venue or another. It also means that logs people are actually using every day as the logs are incomplete, which can lead to various kinds of mistakes.
If you're just never going to fix it, please make me a trainee clerk, and I'll fix it, and do some other clerk stuff. I'm very process-y on WP, and worked for 9 years in a nonprofit law firm, so I have the mentality for it. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 20:47, 5 September 2017 (UTC)
2017 checkuser and oversight candidates
The Rambling Man prohibition amendment
- Original announcement
- Please, if you're going to spam the place, at least do it accurately, Brad "recused" and did not "abstain", so please be accurate. The Rambling Man (talk) 20:44, 18 September 2017 (UTC)
- The subsection is titled "Abstain/recuse" in the original motion. AlexEng(TALK) 20:53, 18 September 2017 (UTC)
- At the risk of placing too much weight on this insignificant point, while the section name was recused/abstain, that was to identify where the arb should place their response and Brad clearly marked his response as "recuse". I agree with TRM that it would be more accurate to identify his position as "recuse". If you want to retain the section headings you could make it:
- recuse/abstain: Newyorkbrad (recuse)
- Or something similar.--S Philbrick(Talk) 21:22, 18 September 2017 (UTC)
- If we're going to be pedantic here, the word "abstain" should just be removed entirely. One doesn't declare abstention, it's just what happens when you don't participate. NYB participated to the extent required to declare his recusal. The 7 Arbs not mentioned at all are the ones who abstained. (Don't get me started on Robert's Rules straw polls that count abstentions) Ivanvector (Talk/Edits) 21:40, 18 September 2017 (UTC)
- In ArbCom practice a distinction is drawn between abstention and not voting. One is also usually drawn between abstention and recusal. I was recused on this motion. Any ambiguity in the motion voting categories is not the fault of the Clerk who posted here; I will try to ensure the template is clearer going forward. Newyorkbrad (talk) 02:27, 19 September 2017 (UTC)
- If we're going to be pedantic here, the word "abstain" should just be removed entirely. One doesn't declare abstention, it's just what happens when you don't participate. NYB participated to the extent required to declare his recusal. The 7 Arbs not mentioned at all are the ones who abstained. (Don't get me started on Robert's Rules straw polls that count abstentions) Ivanvector (Talk/Edits) 21:40, 18 September 2017 (UTC)
- The subsection is titled "Abstain/recuse" in the original motion. AlexEng(TALK) 20:53, 18 September 2017 (UTC)
- Please, if you're going to spam the place, at least do it accurately, Brad "recused" and did not "abstain", so please be accurate. The Rambling Man (talk) 20:44, 18 September 2017 (UTC)
- Just to be clear on the ammendment, because of the changes TRM is now encouraged to insult or belittle other editors all he wants, so long as he doesn't engage in speculations about their motivations or their competence? Am I reading that correctly?--Jayron32 11:17, 19 September 2017 (UTC)
- No. If you read the whole debate you'd understand. The Rambling Man (talk) 11:42, 19 September 2017 (UTC)
- Thanks for that reminder. It was long, but it does clarify the problems with the prior wording. Thank you. --Jayron32 13:19, 19 September 2017 (UTC)
- No. If you read the whole debate you'd understand. The Rambling Man (talk) 11:42, 19 September 2017 (UTC)
Catflap08 and Hijiri88: Motion (September 2017)
2017 Checkuser and Oversight appointments: Candidates appointed
- Hey, congratulations to all our new functionaries! Mz7 (talk) 06:54, 7 October 2017 (UTC)
- Pretty crazy. I thought I'd get a polite form letter that thanked me for my contributions and suggested I reapply next year. Well, I signed the confidentiality form. That was actually a kind of complicated process. NinjaRobotPirate (talk) 07:45, 7 October 2017 (UTC)
- I'm leaving my computer for an all-day trip, but will check back later to find out next steps.--S Philbrick(Talk) 12:24, 7 October 2017 (UTC)
- Congratulations to the new functionaries, and welcome back, xeno! —DoRD (talk) 16:07, 7 October 2017 (UTC)
- What DoRD said! Congratulations to all! Katietalk 17:09, 7 October 2017 (UTC)
- Congrats to all! Everybody take the next 24 hours off, and then the checkusers should report to be shackled to WP:SPI until that nasty backlog disappears. Cake will be served. Beyond My Ken (talk) —Preceding undated comment added 21:47, 7 October 2017 (UTC)
- Yes, congratulations... you fools...! Hahahahaha! :-) Boing! said Zebedee (talk) 21:58, 7 October 2017 (UTC)
- My condolences to all of you who will now have a lot of work to do. :-) --Tryptofish (talk) 22:23, 7 October 2017 (UTC)
- My thanks to all of the candidates, including those who are now going through the appointment process. It is important to bring new members into the team on a regular basis, and these are fine choices. Look forward to working with you. Risker (talk) 23:05, 7 October 2017 (UTC)
- Just a note to thank the committee for their efforts and I will dig in starting tomorrow or so. –xenotalk 14:50, 10 October 2017 (UTC)
Arthur Rubin case closed
This has been hashed out again and again. We got nowhere then, we'll get nowhere now. Mdann52 (talk) 21:52, 17 October 2017 (UTC) |
---|
The following discussion has been closed. Please do not modify it. |
Alright, this was a discussion. Not all of them have satisfactory conclusions. Dennis, I understand where you're coming from, but I beg to differ on the "chilling" part: The Rambling Man knows well what he can and cannot do within the limits of his topic ban, a well-earned behavioral guideline which was probably broken twice here already. My endorsement of Mdann52's hatting is therefore also preventative (besides the fact that Mdann is our clerk and therefore has a mandate--please don't use the term "censorship" lightly). If ArbCom indeed sucks so badly, then no doubt we'll all get voted out (or not voted in, those of us who are running again), and I understand that TRM is drafting his own scorecard--which of course will also have to adhere to the guidelines, unless he runs it on Reddit. Drmies (talk) 00:41, 18 October 2017 (UTC) |
Self-nominations for the 2017 ArbCom elections are now open
Self-nominations for the 2017 English Wikipedia Arbitration Committee elections are now open. The nomination period runs from Sunday 00:00, 12 November (UTC) until Tuesday 23:59, 21 November 2017 (UTC). Editors interested in running should review the eligibility criteria listed at the top of Wikipedia:Arbitration Committee Elections December 2017/Candidates, then create a candidate page by following the instructions there. Mz7 (talk) 06:48, 12 November 2017 (UTC)
Motion: Sexology
Temporary checkuser permission for election scrutineers
- Doesn't this need to be posted at m:SRP? --Rschen7754 05:29, 27 November 2017 (UTC)
- Done. Callanecc (talk • contribs • logs) 08:52, 27 November 2017 (UTC)
- @Callanecc: Before those are processed, it might be worth thinking whether the "temporary" user right feature (new from last year) should be used here. If so, you'll need to specify the expiry time for the stewards. ~ Rob13Talk 08:56, 27 November 2017 (UTC)
- It would be up to the stewards whether they'd consider it for CU/OS rights. Apart from that, it would also mean another motion and vote by the Committee. In practice, it doesn't make that much difference, as, when they are finished, stewards just tend to remove their own temporary CU right. I'd also argue that this way, they are more likely to remove the CU right when they finish rather than waiting for it to expire by itself (if that is actually a concern). Callanecc (talk • contribs • logs) 09:02, 27 November 2017 (UTC)
- We do use that feature for temporary CU/OS access, such as on ptwiki where functionaries are confirmed each year. However, the status quo of granting and removing works here, given the uncertain timeframe of the result checking. -- Ajraddatz (talk) 10:22, 27 November 2017 (UTC)
- As I noted in proposing the motion, the right should be lifted once the election results are certified. We can't put a specific expiration date since we don't know how long the scrutineering will take. That being said, this identical motion has been adopted for each of the past several years and as far as I know there's never been a problem. Newyorkbrad (talk) 21:07, 27 November 2017 (UTC)
- We do use that feature for temporary CU/OS access, such as on ptwiki where functionaries are confirmed each year. However, the status quo of granting and removing works here, given the uncertain timeframe of the result checking. -- Ajraddatz (talk) 10:22, 27 November 2017 (UTC)
- It would be up to the stewards whether they'd consider it for CU/OS rights. Apart from that, it would also mean another motion and vote by the Committee. In practice, it doesn't make that much difference, as, when they are finished, stewards just tend to remove their own temporary CU right. I'd also argue that this way, they are more likely to remove the CU right when they finish rather than waiting for it to expire by itself (if that is actually a concern). Callanecc (talk • contribs • logs) 09:02, 27 November 2017 (UTC)
- @Callanecc: Before those are processed, it might be worth thinking whether the "temporary" user right feature (new from last year) should be used here. If so, you'll need to specify the expiry time for the stewards. ~ Rob13Talk 08:56, 27 November 2017 (UTC)
- Done. Callanecc (talk • contribs • logs) 08:52, 27 November 2017 (UTC)