Wikipedia talk:Arbitration Committee/Noticeboard/Archive 29

Archive 25Archive 27Archive 28Archive 29Archive 30Archive 31Archive 35
Original announcement
That's a good idea, Salvidrim!, I'll make the edit. Liz Read! Talk! 21:32, 26 June 2015 (UTC)
Hehe, sorry for giving you extra work. ;) ☺ · Salvidrim! ·  21:34, 26 June 2015 (UTC)
No problem, Salvidrim! This was my first time solo closing a case so I'm looking for feedback on this one. Liz Read! Talk! 21:41, 26 June 2015 (UTC)
Could you somehow copy the relevant text at Wikipedia:Arbitration/Requests/Case/Technical 13/Proposed decision#Motion to close to the main case page, so the resolution the case is on there? It seems pretty weird for all the headings from "Principles" onward to be empty in a closed case, and they should probably be removed altogether as I'm not sure if they're applicable here. Graham87 12:29, 27 June 2015 (UTC)
As an aside, I was very sorry to see how this case concluded. I had work closely with Technical 13 on a couple of projects and found his assistance helpful. That said we haven't been involved recently, and I had no knowledge of or involvement with some of the issues that led to the case so did not contribute to the case.
Back on topic, like Graham87, I was surprised to go to the proposed decision page and see a sea of boilerplate not yet filled out. When I scrolled to the top and saw the motion to close it became clearer, but it was still a bit disconcerting to see those sections below. However, rather than delete them much which might raise different questions, I would propose collapsing in them with a notation something to the effect that these sections were not applicable for this case see the motion to close.--S Philbrick(Talk) 13:34, 27 June 2015 (UTC)
Sphilbrick, I like the idea of collapsing the unused sections. I agree they'd be confusing in a case that had no PD and was closed by motion or suspended. Seraphimblade Talk to me 15:31, 27 June 2015 (UTC)
I'd be happy with that.Doug Weller (talk) 15:43, 27 June 2015 (UTC)
It was suggested on the Clerks list to close this case like Wikipedia:Arbitration/Requests/Case/SchuminWeb so I have collapsed the unused sections on the Main Case page. Liz Read! Talk! 15:58, 27 June 2015 (UTC)
I did the same on the proposed decision page.[1] In my opinion, this is the sort of uncontroversial and helpful action that the clerks should be doing as a matter of course. --Guy Macon (talk) 18:41, 27 June 2015 (UTC)
There are discussions on clerks-l about what to do in cases like this, in case anyone's wondering. - Penwhale | dance in the air and follow his steps 16:22, 28 June 2015 (UTC)
  • Since this is an arbcom block, could an arb decide what they want to do with T13's user page? I'm generally not a fan of uninvolved editors "helping out" by replacing a user page with a {{indefblocked/banned/whatever}} banner when the blocking admin or arb hasn't felt the need to place it (which was done here). Actually, I'm not a fan of the banners in general, as I think they are often viewed by the banned editor as unfair, and therefore a reason sometimes to come back somehow to fight its placement. Replacing the user page as it was, or deleting it, would IMHO make it slightly more likely that things die down. --Floquenbeam (talk) 15:40, 27 June 2015 (UTC)
    • I don't see that as really our call, to be honest. If you think the standard means of dealing with user pages in such cases is unsatisfactory, that's really the community's choice to deprecate or change the practice. You could start a discussion on it, of course, but I don't see it as something we'd do by fiat. Seraphimblade Talk to me 03:16, 28 June 2015 (UTC)
      • @Floquenbeam: This is a perennial issue. The general practice is to add a "banned" tag at the top of the userpage, but implementation of this can be inconsistent; and, as Seriph. said, we leave it to others. However, somebody should clearly overhaul the banned tag itself. It is shouty and overly prominent. If the tag was toned down, adding it would look less like grave-dancing, and you might solve the problem of what to do with banned users' userpages because it'd become less of a big deal. AGK [•] 14:43, 28 June 2015 (UTC)
        • I would suggest having two versions of the "ban tag" – a "less shouty" more matter-of-fact one for situations like these, and the current "more shouty" one for those that have received punitive bans. --IJBall (contribstalk) 15:48, 28 June 2015 (UTC)
          • I don't think that would be a good idea. There is not supposed to be such a thing as "punitive bans", and creating a template that supposes the existence of them would make this problem worse, not better… AGK [•] 16:02, 28 June 2015 (UTC)
            • I understand. But when someone has gotten a site ban at either ANI or ArbCom for serious offenses (e.g. trolling, harassment, pervasive socking, and such), it's probably a good idea that editors be made fully aware of that, even if it's subtle. While bans aren't supposed to be "punitive", some bans are a far more serious matter than others (such as this case with T13, which is basically a "technical" ban rather than a "serious offense" one), and as a general concept it's probably a better idea that the community know that (if for no other reason it may help editors to better flesh out socks, in those sorts of cases; etc.)... --IJBall (contribstalk) 16:11, 28 June 2015 (UTC)
              • I think we (as a community) obsess too much over how a banned user should be tagged that we get distracted from what the purpose of Wikipedia is, to be a bit blunt. --Rschen7754 16:17, 28 June 2015 (UTC)
                • True. But if you hang out at ANI too much (as I do), you do start to see how persistent some of this project's worst disruptors can be (and how much effort some of regular editors have to go through, just to try to stop them, so anything that can assist those editors in their task is all to the good, to my thinking). --IJBall (contribstalk) 16:38, 28 June 2015 (UTC)
  • What the...? I admit, I'm confused by this – T13 asked for a indef block, and the committee upped it to an indef ban?... I feel like I've missed something here. Could somebody super-briefly summarize this result for me? TIA. --IJBall (contribstalk) 05:47, 28 June 2015 (UTC)
  • The committee was starting to hear a full case. T13 asked for an indef-block due to retiring. Instead of suspending the case until T13's uncertain return, ArbCom opted to close the case immediately by motion, by converting the existing indef-block in a siteban and ask that it be appealed to them should T13 ever wish to return. The only difference between a self-requested indef-block and an ArbCom siteban is that the former can be undone by any admin, while the latter can only be appealed to ArbCom -- in short, ArbCom didn't feel it necessary to hear a full case in asbentia but wanted to make sure T13 would go through them before returning, if ever.  · Salvidrim! ·  06:09, 28 June 2015 (UTC)
  • OK, that makes perfect sense – ArbCom just wanted to be the ones that would formally allow T13 to return to editing, rather than just a fellow Admin. I totally get it now. Thanks Salvidrim!(!) --IJBall (contribstalk) 06:15, 28 June 2015 (UTC)
  • I voted for this in principle on the mailing list, but did not get round to the on-wiki page on time. So, noting my support after the fact. AGK [•] 14:45, 28 June 2015 (UTC)

Promotion of clerk L235

Original announcement

I'm pleased to see this. We disagreed on one clerk action in Wikipedia talk:Arbitration/Requests/Case/Sockpuppet investigation block, but I'm generally in favour of more clerk oversight of relevance and decorum, and L235 seems to be able to manage that with tact and grace. --Anthonyhcole (talk · contribs · email)

Temporary injunction regarding the restoration of Scott's admin privileges

Original announcement
  • I start by noting aside from his other issues, including off-wiki commentary, that Scott's User page has this statement: "In August 2014 I gave up being a regular Wikipedia editor on the basis of my belief that this project is fundamentally broken." So, I find the possibility of this WP:DIVA regaining admin tools to be a concern. This should be a no-brainer for ArbCom, and I urge a merciful and speedy rejection of Scott's request for the extra admin buttons. Thanks. Jusdafax 14:47, 1 July 2015 (UTC)
    If someone is a WP:DIVA who resigns adminship in a huff, but has done nothing wrong, should they be denied resysopping? Hawkeye7 (talk) 23:42, 2 July 2015 (UTC)
    Given that the "Diva" page begins A Wikipedia diva is a long-time user who believes he or she is more important than other editors, and is also predicated around an editor having left in response to a dispute, whatever answer you get to that question will be interesting, but not relevant to my case.  — Scott talk 23:50, 2 July 2015 (UTC)
Agreed! But that brings up another point: if there is no dispute, then the matter is beyond ArbCom's remit. Hawkeye7 (talk) 21:35, 3 July 2015 (UTC)
  • As Scott did not resign under a cloud and has never been a party of an ArbCom case, so far as I am aware, ArbCom is clearly taking "Ignore All Rules" to a new extreme here. On what policy basis do they have a right to intervene here? This is a matter for the Bureaucrats not for the Discipline Committee and the secret, poison-pen testimony sent to them by personal enemies... Carrite (talk) 14:53, 1 July 2015 (UTC)
  • Well WP:IAR is policy, as is Wikipedia is not a bureaucracy. Scott and his defenders have gone to great lengths to wrap themselves in every Wikipedia policy except those two fundamental ones. When I attempted to engage him in a dialogue about his off-wiki attacks against me, he took the Wikipedia version of "the Fifth" and declined to answer. (Oh except for correcting my grammar in describing his abuse.) Yes, Wikipedia is broken. One of the ways it is broken is the Scotts of Wikipedia, Wikipedia admins who feel that it is OK to go off-wiki to behave like Mr. Hyde while they act like Dr. Jekyl on-wiki. There are a number of ways of characterizing that kind of behavior, but I think that arbcom should make one of those words "unacceptable." Coretheapple (talk) 15:04, 1 July 2015 (UTC)
"IAR is policy" + "Wikipedia is not a bureaucracy" = a neat rationale for lawless clique rule as opposed to the rule of law. We all have differing visions of Wikipedia's ideal future, I suppose. Carrite (talk) 15:50, 1 July 2015 (UTC)
Avoiding being under "the rule of law" is pretty much the purpose of those policies. Regardless you have been given more specific policy rational below. Chillum 15:52, 1 July 2015 (UTC)
In my opinion the rule of law and the protection of rights is a good thing. Cliques ruling by exertion of raw power is a bad thing. Your mileage may vary. Carrite (talk) 15:56, 1 July 2015 (UTC)
This is a publication ran by a not-for-profit organization. There is no "right" to be an admin. Any argument you make should not be derived from an expectation of rights on someone else's website. Chillum 16:01, 1 July 2015 (UTC)
This is an encyclopedia administered by an autonomous community hosted on servers maintained by a not-for-profit organization. ArbCom is absolutely overextending their mandate here. See ya next elections, guys. Carrite (talk) 22:24, 1 July 2015 (UTC)
Policy basis for intervention is this section from WP:ARBPOL: "The Committee may take notice of conduct outside its jurisdiction (en-WP) when making decisions about conduct on the English Wikipedia if such outside conduct impacts or has the potential to impact adversely upon the English Wikipedia or its editors." -- Euryalus (talk) 15:22, 1 July 2015 (UTC)
That reference is obviously to ArbCom with respect to cases it takes; it is not carte blanche for the committee to intervene into non-ArbCom processes across Wikipedia. ArbCom does not have universal veto power over everything at En-WP. This strikes me as a power grab. Carrite (talk) 15:46, 1 July 2015 (UTC)
(ec)I may be wrong but I don't think it is that unusual for arbcom to make decisions without a formal case. Chillum 15:49, 1 July 2015 (UTC)
Exacty. See, for instance, WP:LEVEL2. Salvio Let's talk about it! 16:03, 1 July 2015 (UTC)
As far as I'm concerned, it's a commonsense interpretation of the level-II desysop procedure which allows us to act in these cases; of course, we could wait for the 'crats to flip the switch and desysop afterwards, but in my opinion that would be pointless. Salvio Let's talk about it! 15:48, 1 July 2015 (UTC)
Desyssop for what?!?' Scott has never abused tools. He did not exit under a cloud. This is not a community reconfirmation hearing or a Arbcom case in which there is no formal decision to accept, no opportunity for testimony. It is a couple enemies of Wikipediocracy using the poison pen and Arbcom unilaterally extending its power because it thinks it can. Carrite (talk) 15:53, 1 July 2015 (UTC)
We are not unilaterally extending our powers. Desysopping admins who have abused their tools or conducted themselves unbecomingly is and has always been within our province. Salvio Let's talk about it! 16:03, 1 July 2015 (UTC)
Salvio giuliano I do not know enough about the facts the case to know how Arbcom will decide. However, if they decide that Scott should not have his admin bits, I respectfully disagree that the committee should simply direct the bureaucrats to not return them. It may seem pointless to have the bureaucrats restore the bits and then Arbcom removed them but I think that is the right thing to do if the decision is that Scott should not be an admin. While I think it was quite appropriate for the bureaucrats to hold off restoration pending the investigation, if the result is that the case is closed and the bits are not restored, they will be forever some confusion about the sequence of events. For example, if someone asked a few years in the future if the bureaucrats have ever declined to restore the bits to an admin who did not resign under a cloud the answer would have to be "well, yes, sort of, arguably no, it's sort of complicated". Who needs that? If the bits are restored by the bureaucrats and then removed by Arbcom, the answer will be clear.--S Philbrick(Talk) 16:44, 1 July 2015 (UTC)

On the bureaucrats' noticeboard, I suggested that Scott's request for resysopping be placed on hold temporarily while he was discussing with ArbCom, and Scott readily agreed. In an effort to avoid unnecessary tension and drama—or perhaps more realistically, at least to postpone any dispute at least until we know what is being disputed—I suggest that this action be considered non-controversial simply on that basis.

The merits of Scott's request to resume his administrator tools are not the issue here, but for what it is worth, the fact that either a current or a former administrator has criticized Wikipedia is not, by itself, a sufficient basis for desysopping or for declining resysopping. Of course, if a hypothetical once and would-be future admin were to reveal an intent to use the tools in a fashion inconsistent with the best interests of the project, or it transpired that he or she had misbehaved in some desysopworthy fashion, that would be a different matter. Newyorkbrad (talk) 15:54, 1 July 2015 (UTC)

Taking time to review this request is obviously a sensible precaution. There shouldn't be anything controversial about arbcom's action here. --B (talk) 16:06, 1 July 2015 (UTC)
The committee was foolish to take this on. Whatever decision they make, they will further alienate either their critics here or at other sites. American Pharaoh (talk) 16:21, 1 July 2015 (UTC)
I doubt that the committee's decision is based on what people on "other sites" have to say about them. While the committee frequently makes errors, I don't believe they did in this case. It's certainly reasonable, given the nature of the objections, to investigate before granting the request. --B (talk) 16:28, 1 July 2015 (UTC)
The nature of the objections is exactly why the committee should not have become involved. It simply reinforces the perception that there are two sides engaged in a battle to the death. There are not. Recall that even the former husband of Lila Tretikov ran a Wikipedia criticism site. American Pharaoh (talk) 14:19, 2 July 2015 (UTC)
Regardless of the source of the objections, if allegations of impropriety have been made to ArbCom, accompanied with what is stated to be evidence - which in this case I'm yet to be shown - it's only right that the committee should take time to investigate. I would want them to take any submission I made to them seriously; the door swings both ways.  — Scott talk 16:14, 2 July 2015 (UTC)

After the many requests that I've seen come up, with plenty of opposition (and for far better reasons than those presented in this case) and the response being "welp, they didn't give them up under a cloud, restore bit", this is extremely interesting. Arkon (talk) 16:19, 1 July 2015 (UTC)

I don't really pay attention to BN so I don't know what other similar cases there might have been there. Is it possible that in these cases the committee didn't intervene because nobody brought it to their attention? I was somewhat disturbed from reading the BN thread that the bureaucrats seem to have been prepared to, as you say, handle this situation as "welp, they didn't give them up under a cloud, restore bit" before the committee asked them not to. Even if you don't resign under a cloud, if you're coming back under a cloud, our policies shouldn't be a suicide pact that we're going to take action that is not in the best interest of the project. --B (talk) 16:28, 1 July 2015 (UTC)
We have taken action in this case because we have received private communication regarding the case. After initial investigation of this, there are questions we want Scott's answers to, and this could not be done within the timeframe the 'crats were prepared to wait without a formal motion (and they have acted perfectly properly here). The policy that allows us to do this has been explained below. Why we do not act in all cases is that this is (I think, I haven't checked) only the second occasion since I became an arbitrator in January that the committee has been in possession of relevant information. If you (or anyone) would like controversial resysop requests to be handled differently then you need to get a consensus to change the WP:RESYSOP policy (and neither this page nor the 'crats noticeboard is the correct venue for that). Thryduulf (talk) 23:29, 1 July 2015 (UTC)
It is not a "case." If you want to open a case, that's within your purview after pro forma restoration of tools, assuming there is cause. Poison pen secret communications and a new veto power are not the way things are done. It is an appalling precedent, for starters; it is even more appalling that ArbCom seems oblivious to the terrible precedent they are setting. You've got an open case, a real case, that is coming up fast on two months since acceptance and the proposed decision options aren't even up. Work on that, that's your job. This is not your job, "approving" retoolings has never been your job. Carrite (talk) 00:21, 3 July 2015 (UTC)
No-one has a problem with ArbCom taking their time here. However, there are three possible outcomes here. (1) Scott has his permissions restore (2) Scott does not have his permissions restored, with a valid reason (3) Scott does not have his permissions restored, "but we can't tell you why". (1) and (2) are acceptable, (3) is not. Black Kite (talk) 00:31, 3 July 2015 (UTC)
I especially like your phrase "coming back under a cloud" which is perfectly apt. Binksternet (talk) 16:45, 1 July 2015 (UTC)
It's a good idea I think. It's just an odd time to implement such a thing, when there has been plenty of chance to do so in the past when faced with what I would consider a greater "cloud". Arkon (talk) 16:51, 1 July 2015 (UTC)
+1. Really nothing more to say than that, as these people unilaterally set new precedents on a daily basis, it seems. Carrite (talk) 22:20, 1 July 2015 (UTC)
  • I assume some here have access to the WPO thread but most don't. I was reading, refreshed, it disappeared into their hidden section. I'm not sure what was there, but I did find it unusual that all of a sudden, the thread on Scott getting back the admin bit disappears during a refresh. I'm assuming that thread and what I didn't see has to do with why we are here, and I'm assuming the Arbs will clear it up quickly. Rare occurrences call for creative processes and as long as both parties are satisfied with the delay (and Scott has said he is okay with it), then everyone just needs to relax and wait a day or two before jumping to conclusions. Even the Crat at the switch was essentially encouraging Arb take a look. Dennis Brown - 22:29, 1 July 2015 (UTC)
    Actually it was moved from the must-be-logged-in area to the public area, Dennis Brown, since people from here had been asking to see it. You'll see it's current location here. --SB_Johnny | talk23:04, 1 July 2015 (UTC)
    Then it was moved back, because I've never had an account there to log into. Dennis Brown - 23:09, 1 July 2015 (UTC)

I suspect ArbCom (and the bureaucrats) are hoping that this sort of thing will be a rare occurrence. It would be a bit tiresome if every potentially controversial resysop request goes through this sort of process. While people should write to ArbCom if they have genuine concerns, giving this more time than it deserves may encourage people to write to ArbCom on flimsier grounds and that would waste ArbCom's time as well as the time of others. What I would like to see here is a clear statement from ArbCom why they end up taking any action they take (or indeed no action). i.e. that Scott is under no cloud at all, or that there is a reason for (effectively) desysopping (and that this was discussed with Scott and he was given a chance to challenge it in a full case if needed). I would also hope that they are asking why this was brought up now, and not earlier. Finally, if the complaints are frivolous, those making the complaints should be admonished in some way to discourage similar conduct in the future. Carcharoth (talk) 06:40, 2 July 2015 (UTC)

I think that a couple of the comments on the Bureaucrats' noticeboard are worth reproducing here:
"Without comment on this particular request (I haven't had the time to review it in detail), but it seems that objections to procedural resysoppings have become more common lately, which will in turn make administrators less likely to give up their tools in situations when it may be prudent for whatever reason (be it account security, personal reasons, needing a break, etc.).
The community may wish to implement a lightweight desysopping process, but shoehorning it into the resysopping process as a "gotcha" is not the way to go."[2]
and
"We've established a policy that if an admin voluntarily gives up tools not under a cloud they simply have to ask for them back. We do not get to say "now that you don't have the tools and you want them back is a few questions I'd like answered and if you don't answer them satisfactorily you might not get the tools back.
I get that there might be special circumstances in this situation and I'll let ARBCOM investigate those off-camera. However, should there be a declination to return the tools, it had better be for an extremely strong reason and not simply that someone is unhappy with something Scott has done and now have a good chance to keep him from getting back the tools. On a related point, I do want to thank Scott for agreeing to a delay."[3]
--Guy Macon (talk) 07:44, 2 July 2015 (UTC)

As a Crat, I have some concerns along the lines mentioned by Guy Macon above, balanced by Carcharoth's usual wisdom. I'd ask Arbcom to read Carcharoth's comments carefully. --Dweller (talk) 13:47, 2 July 2015 (UTC)

I love that members of the Arbitration Committee so often repeat the "ArbCom is not a court" meme while simultaneously embracing legal phrases such as temporary injunction. I strongly agree with Black Kite, Carrite, and Carcharoth here. This is way out of bounds and a terrible precedent. --MZMcBride (talk) 17:50, 2 July 2015 (UTC)

Scott is okay with it so far, but I think it has gone on long enough Arb needs to explain the general gist of why they did the injunction, to more than just Scott, since the community as a whole has an interest. Dennis Brown - 20:31, 2 July 2015 (UTC)
We have explained everything we can currently publicly explain already. When we make a decision we will probably be able to explain more. Thryduulf (talk) 21:45, 2 July 2015 (UTC)
@Thryduulf: You did receive my email? I've not had any acknowledgement in the last 23 hours.  — Scott talk 22:06, 2 July 2015 (UTC)
Pinging Thryduulf. FYI, Scott, pings don't work unless a new signature (technically a new link to your own userpage) is added to the message in the same edit. L235 (t / c / ping in reply) 22:49, 2 July 2015 (UTC)
I wonder how it is I never knew that. Thank you.  — Scott talk 23:00, 2 July 2015 (UTC)

I'm wondering if they'll exercise this much oversight when some of those who dropped the tools and "retired" to avoid consequences for their actions appear and ask for the tools back? Intothatdarkness 21:30, 2 July 2015 (UTC)

If somebody resigns their tools to avoid consequences of their actions they are termed as having resigned "under a cloud". The WP:RESYSOP policy requires that in such circumstances you can only regain the tools by successfully passing a new RFA so ArbCom oversight is almost never required. Thryduulf (talk) 21:45, 2 July 2015 (UTC)
There is no policy provision for "ArbCom oversight." You are making stuff up. Carrite (talk) 00:26, 3 July 2015 (UTC)
  • Noting that there is some precedent here, in a more clear cut case last year, an individual dropped their tools whilst arbcom was discussing a block, meaning the arbs considered it under a cloud. The community decided that there was no cloud declared at the time, and re-instated the tools (which were subsequently removed later in the year). In this case, as I was on Arbcom at the time, I'm confident there was not even a potential cloud when Scott left.
    I'm not saying that Arbcom can't prevent this resysop. I haven't seen the evidence that Scott has been acting poorly on Wikipediocracy, if such evidence exists, I agree that he should not have the admin bit. I do, however, believe the decision should hold up against scrutiny - therefore should be modelled on Level II procedures - importantly with the understanding that if the committee does suggest that return of the user-rights is not appropriate, a case may be required. WormTT(talk) 07:42, 3 July 2015 (UTC)

I have read far too much -- if ArbCom finds that a person who did not leave "under a cloud" should be singled out for special treatment, then ArbCom should be willing to start a case about the person, and not rely on Star Chamber-like proceedings in camera. The belief that "secret covenants secretly arrived at" are desirable on Wikipedia should be an issue to be addressed by the Community, and not by arrogation of powers by a committee.

We already have a tainted RfA where a former Arbitrator referred apparently to information from the committee as a reason to bar a person being re-instated as an admin ( I assume the nominator was unaware of the history of Rich Farmbrough and the full details of the case, otherwise they would not have proceeded with a nomination and a misleading statement that could potentially harm their reputation in future RfA nominations) and one or more sitting Arbitrators or clerks have opined on that RfA ( I have no confidence in the nominee's judgment, No answers to questions, extremely extensive block log, currently subject to an ArbCom-imposed restriction and two community-imposed sanctions, loss of community trust, personal lack of trust in his judgement, likely net negative, etc. appear to contribute to using an ex officio status to govern a community process, alas.)

If ArbCom really wants to de-admin anyone, they should "man up" Person Up and open a case. Collect (talk) 13:26, 3 July 2015 (UTC)

@Collect: what we have been doing is establishing whether there is a justification for either a case or a straight desysopping. When case requests are posted publicly it is not unusual for it to take several days for arbitrators to make a determination of whether a case is required or not - it is no different when the information is presented to us privately. We have been in communication with Scott and he has answered our questions, and we are just now coming towards then end of our discussions based on those responses. Thryduulf (talk) 19:28, 3 July 2015 (UTC)
Linguistic debate is off-topic here. Thryduulf (talk) 00:33, 4 July 2015 (UTC)
The following discussion has been closed. Please do not modify it.
Surely there is a better expression than that. GorillaWarfare (talk) 19:53, 3 July 2015 (UTC)
I am moderately literate in English and I did not find a totally gender-neutral equivalent at all. Else I would certainly have used it. Collect (talk) 19:59, 3 July 2015 (UTC)
Understandable; it's a shame that the phrase has become so ubiquitous. For the future, perhaps "grow a spine", "take responsibility", or "get it together" would convey your meaning just as well. GorillaWarfare (talk) 20:06, 3 July 2015 (UTC)
"Grow a spine" specifically is singular in nature - while "man up" can, and often does, refer to a group. "Get it together" is quite off the mark, alas.
Ben Zimmer in The New York Times specifically refers to the meaning as including: "“Do the right thing; be a mensch,” to use the Yiddishism for an honorable or upright person. ... In a nice mash-up of idioms, Rabbi Daniel Polish has interpreted the Torah story of Joseph and his brothers as a parable of — what else? — mensching up."[4] [5] The On-Line Slang Dictionary assigns no "gendered" meaning to the verb, and no equivalent words.
MacMillan: to start being brave and dealing with a difficult situation.
CED: used to tell someone that they should deal with something more bravely None of the major sources imply in any way that it is a gender-problematic term.
J.P. in The Economist[6] comes straight out to say "Person up" and "people up" are absurd. So where dictionaries offer no remotely similar term for the precise meaning, and the "gender neutral" version is derided, I suggest I used a perfectly proper term.
The Guardian notes that a noted feminist - Harriet Harman uses the term - saying "Couldn't think of alternative!". Which is the same position I am in. It has the ability to be used of a group ("grow a spine" fails) and "get it (one's act) together" as an idiom means "To become serious, organized, worthwhile, etc. " which is nowhere near the idiomatic usage given here.
OLSD has "to get one's current state in order." So I am quite sorry - but I am using Harman's wording. Collect (talk) 20:31, 3 July 2015 (UTC)
Interesting. When I google "man up" phrase (to avoid results for what was apparently a film, TV show, etc.), seven of the ten results ([7], [8], [9], [10], [11], [12], [13]) discuss how it is a gendered term; the other three ([14], [15], [16]) offer definitions without discussing whether it's gendered. It's always interesting when people insist on continuing to use language that others take issue with (particularly when the people who take issue with it are among those it's targeting). GorillaWarfare (talk) 21:01, 3 July 2015 (UTC)
You appear to read far different intonation into sources than I would (since I gave many of those precise sources).
I cited Zimmer's NYT column at length above.
I would not use "everdayfeminism.com" as a source for any claims at all - it has an editorial position, but is not a "reliable source" at all for English usage.
The MacMillan source, as I earlier stated, does not back "person up" or the like at all. In fact it says "Somehow, telling someone to person up just doesn't have the same ring to it, precisely because the meaning of the expression is tied up in the stereotypical qualities of masculinity and everything that the word man implies – being tough or brave in the face of adversity, etc. We therefore seem to have come full circle, using a word infused with gender-specific stereotypes in a gender-neutral way.
I had, in fact, already cited Wiktionary, and your use does not back your claim, alas.
Stackexchange directly backs my position: Many of these kinds of phrases will cause offense in exactly those who are looking for it - my wife will occasionally tell one of my daughters to man up - it's an informal, short and pithy phrase and if you're offended, well, erm, man up. There are some genuinely offensive phrases in the list but this is more than likely going to backfire on them, cf. the Streisand Effect was one response on the page you cite here.
"Goodmenproject.com" does not appear to be a "reliable source" either - its own self-description "“The Good Men Project is a glimpse of what enlightened masculinity might look like in the 21st century,” " specifically removes itself from being a reliable source about actual current English word usage.
And to leap from "the source says nothing at all abut gender in the term" to an implication that therefore the term is considered gendered is a very vast leap indeed.
As I said before thrice - I would have used a completely non-gender term if one were readily available. Ms. Harman could not find one, nor can I. Misrepresenting the sources I already gave, is, alas, a rather weak form of argument on the issue, such as it is. Cheers. Collect (talk) 21:28, 3 July 2015 (UTC)
Better yet - I shall simply agree to disagree with you on the "genderness" of the phrase herein discussed at length. Cheers. Collect (talk) 21:35, 3 July 2015 (UTC)
The difference here is that this isn't a Wikipedia article. I mentioned that being told to "man up" strikes me as unnecessarily gendered, asked you not to use it, and offered a few alternatives. Insisting that it's not gendered is basically saying "I don't care that this makes you uncomfortable, and I care more about using this phrase than I care about the people who find it bothersome." Anyhow, I'll stop arguing the point in the interests of not derailing this conversation further, but I would ask that you think about it. GorillaWarfare (talk) 21:41, 3 July 2015 (UTC)
I held back from commenting earlier but this is really more than I can bear. Collect: Stop it. Whatever it is you hope to achieve by mounting an impassioned defense of your phrase (which is, by the way, gendered, and inappropriate to use in a project that aims to be inclusive of all genders), the only thing you're actually achieving is making yourself look like a sea lion.  — Scott talk 21:44, 3 July 2015 (UTC)
I already did. I do, however, demur that "everydayfeminism.com" is an authority on current English usage. And since four times now I have said I would have used a different phrase if I could find one I suggest that the stick is not being held by me. Cheers as always. And I find snark to be rather weak for discussions. Collect (talk) 21:50, 3 July 2015 (UTC)
There was no snark in what I said. You are behaving absurdly for no obvious reason. If you're intelligent enough to contribute to Wikipedia, then I'm sure you'll be able to find an appropriate way to express yourself in future.  — Scott talk 21:55, 3 July 2015 (UTC)
I think Collect and GW should just move the discussion to Wikipediocracy, as it seems a more fitting milieu for it. --SB_Johnny | talk22:03, 3 July 2015 (UTC)
Actually, discussion of linguistics, word choices and the like are what Wiktionary's Tea Room is expressly for. Thryduulf (talk) 00:33, 4 July 2015 (UTC)

With respect to Wikipediocracy moving threads

There has been some commentary intimating that Wikipediocracy's moderation team (of which I am not a part) is behaving nefariously by moving threads from publicly-accessible areas to registered-accounts-only areas. Here is the explanation: public threads are visible to search engines (Google), private threads are not. If a thread starts to get overly personal it is sometimes moved away from Google visibility to minimize collateral damage, if you will. This is a good thing, yes? Anyone can register a WPO account to see this material if they so desire, or not. There is also a backstage area in which the inner circle of WPO administrators communicate with one another, I understand, akin to ArbCom-l, which is a restricted mailing list. I do not have access to this area but believe it exists, as it reasonably should. Carrite (talk) 00:33, 3 July 2015 (UTC)

Tim's correct on all points. There's a private area where really problematic posts can be moved if the moderation team needs to decide what should be done with them (the forum equivalent of RevDel), but public threads are never more than a forum registration away from any interested reader.  — Scott talk 01:10, 3 July 2015 (UTC)


If Scott is sceptical about Wikipedia on rational grounds (not sure if that's a case, though), then there is no reason to believe that he can't be a good Admin. No system is perfect, but often within a group participating within any system you tend to get people who are "ideological believers" who tend to overlook or argue away faults of the system. It are typically people from that group who tend to become the leaders of the system, they are then in a position to exert more influence over who gets into powerful positions. So, ArbCom needs to be vary careful when considering Scott's position on the basis of "political correctness". Count Iblis (talk) 18:59, 4 July 2015 (UTC)

Procedural stuff

@Carcharoth, Dweller, and Worm That Turned: Absent clear policy to the contrary, in my opinion the only legitimate procedural route is: (a) resysopping by the bureaucrats and then (b) desysopping by the Committee (if and only if reasonable grounds exist for doing so, under say Level II procedures). This is because the injunction is not an end in itself as it is not a binding final decision. It only puts things on hold temporarily. So, while the bureaucrats cannot ignore the injunction, they can certainly ask about progress and decide when a reasonable time for investigation has elapsed. This, it seems to me, is consistent with the policy requirement that "if there is a currently open community discussion concerning this issue, re-granting will wait until the discussion is closed". It is important for bureaucrats to have this discretion as a bulwark against inadvertent overreach.  Roger Davies talk 06:19, 5 July 2015 (UTC)

Matches my interpretation, certainly WormTT(talk) 07:58, 5 July 2015 (UTC)
I agree with Roger. Fortunately we have now rescinded the injunction. Doug Weller (talk) 12:45, 5 July 2015 (UTC)
Agreed, and though it's a bit of procedurally bureacratic, this would probably be the cleanest way to have the userrights reflect the events. –xenotalk 12:56, 5 July 2015 (UTC)

Rescinding the temporary injunction regarding the restoration of Scott's admin privileges

Original announcement
Noting that the floor at WP:BN has been re-opened for at least 24 hours to allow for community comments and bureaucrat review. –xenotalk 13:01, 5 July 2015 (UTC)
  • So... why were Thryduulf, Courcelles, DGG, and DeltaQuad opposed to this motion? The question of whether restoration of permissions is subject to ArbCom intervention is an important one, and a summary of what concerns were raised seems fair to disclose, given the advanced permissions. I don't know why keeping the community in the dark is beneficial here. 50.45.233.119 (talk) 19:45, 5 July 2015 (UTC)
    • I opposed because Scott's reply did not satisfy me. Procedurally, policy allowed us two options: 1. desysop, 2. do not desysop. Had we chosen option 1 we would have had the option to impose conditions on how and when he could seek to gain the tools again. I would have been very surprised if the how had been anything other than RFA as ArbCom directly regranting adminship fell out of favour a long time ago (something I agree with). If imposed, the when would either be a minimum time before he could apply (e.g. 6 months) or "at any time" - there would be no obligation on him to apply at any given time of course. Thryduulf (talk) 22:15, 5 July 2015 (UTC)
    • Similarly to Thryduulf. The purpose of procedure is to get a reasonable result, not follow it for its own sake. I thought it better for the community to make the decision. DGG ( talk ) 22:39, 5 July 2015 (UTC)
    • Like the above, really. I did not consider Scott's reply to be in any way satisfying to the relevant concerns, and that such concerns were sufficent that a fresh RFA was necessary. Courcelles (talk) 22:49, 5 July 2015 (UTC)
    • Scott's reply, while addressing some of the issues, did not satisfy the concern that the a part committee had, or that I had. Also their was more opposition to restoring the bit than is publicly noted (in my opinion), only because of 'procedural/policy' grounds, which I of course disagreed with. I was fine with a fresh RfA without a time limit also. -- DQ (ʞlɐʇ) 05:42, 6 July 2015 (UTC)

Motion passed in AE arbitration case granting amnesty and rescinding previous temporary injunction

Original announcement

Since this amnesty seems to remove any basis for imposing sanctions on individual editors, perhaps it would be helpful to clarify why this case remains open. Looie496 (talk) 13:31, 5 July 2015 (UTC)

I gather it remains open to clarify whether declining to impose a sanction is to be considered an administrative action per the AE-handling guidelines, and other unanswered procedural questions. –xenotalk 13:34, 5 July 2015 (UTC)
(edit conflict)To clarify the relevant policy, which is something I'd rather not do by motion, because a full case gives us the opportunity to hear from the community. Salvio Let's talk about it! 13:35, 5 July 2015 (UTC)
Exactly. Doug Weller (talk) 13:42, 5 July 2015 (UTC)
I think this is still badly in need of clarification. What exactly is the case supposed to accomplish? What sort of evidence will be accepted? Are you open to evidence regarding the actions that led to this case? What sort of proposals are appropriate for the workshop? You might consider passing another motion to establish the new parameters of the case. Looie496 (talk) 14:17, 5 July 2015 (UTC)
  • There needs to be clarification, but that clarification would be best served not in a factual vacuum - I hope the committee will still discuss the factual background in relation to policy application (Amnesty is fine, here, but I do think it's bad being hypothetical). The whole structure of the various policies that have been discussed and clearly played various roles here, needs some elucidation [17] in this case context and not as a hypothetical exercise:
    • AE, which comes from where in policy?
    • DS, which seems to be the policy it comes under?
    • Banning Policy and the relations to Blocking and unblocking?
    • The correct procedures and considerations in discussions, for a proper local consensus, and if local consensus is not the proper consideration, where and how does it play any proper role?
    • Where, exactly various discussions should occur?
    • Who has standing to raise them? Who has standing to act, ivote, and declare, per Uninvolved policy?
    • With respect to Arbitration and Enforcement, how and where to address huge WP:POLCON-like issues, like what arose here?
    • Proper behavior in commenting, because we can't really reach clarity without that? Alanscottwalker (talk) 15:12, 5 July 2015 (UTC)
  • Please close the case, and initiate a discussion at the Arbcom procedures page which is the page that authorizes administrator actions in relation to AE. The crux of the issue is whether administrator decisions are to be treated as the equivalent of administrator actions. Certainly that is the case almost everywhere else on the project; closing deletion discussions being a good example, where the appeal process is identical whether it is an admin decision (i.e., closed as keep) or an admin action (closing as delete, and then deleting). I will note this is not a new problem for Arbcom, as I remember it arising a few times during my tenure, usually when an administrator applied a sanction despite the fact that the majority of administrators reviewing a request had posted their position that a sanction was not appropriate. As I recall, there was concern that not supporting admins who applied sanctions made it less likely that admins would do so; however, the reverse is also true, in that admins whose decision not to apply sanctions would suffer a "chill" and stop participating. There is also the issue of double jeopardy, where a report to the AE noticeboard is closed as "no sanction" and another administrator then sanctions the user for the same activities (essentially what happened in this case). I'll note in passing that the procedures page does not appear to be properly organized with respect to its AE procedures, but that's a clerical action I will leave to someone else to manage. Risker (talk) 19:53, 5 July 2015 (UTC)
    There is no such thing as a valid close in less than 24 hours, because it denies people in other time zones a chance to read it. Hawkeye7 (talk) 21:40, 7 July 2015 (UTC)
There are lots of perfectly valid closes within 24 hours. A good half of threads on AN and ANI are closed within 24 hours; SNOW closes are applied in multiple areas, and bad-faith requests all over the project are routinely shut down practically immediately. I think just about everyone reading this page is well aware of this. Risker (talk) 22:06, 7 July 2015 (UTC)
It is fine for a bad faith request to be immediately closed. But an ANI thread that is opened at 3am my time and closed at 5am is bad faith too. Hawkeye7 (talk) 23:13, 7 July 2015 (UTC)
  • The problems involved by our procedure were brought out very clearly in the sequence of events, and the evidence had been sufficiently extensive that we could deal with the questions raised best by proceeding with the case, not starting another procedure. (On this issues involved, I am inclined to pretty much agree with Risker at this point. I am also somewhat skeptical about the general fairness of the entire DS/AE procedures.) DGG ( talk ) 22:45, 5 July 2015 (UTC)
    • DGG, I think you may have a radically different idea about what arbcom cases are about than the community, or even the arbcom policy. Arbcom can make its own procedures, and it not only doesn't need a case for that, but a case is contraindicated for that. Cases are for working out what happened and why, and determining if someone should be sanctioned for behaviour (or not). What cases aren't is the place to draft, change, or deprecate Wikipedia policies or arbcom procedures. Arbcom can't do the former, and it is an inappropriate forum for community consultation on Arbcom procedures. Those are real people who are named as parties to the case, including one of your number, who is now directly excluded from making an Arbcom decision on Arbcom procedures. Risker (talk) 00:09, 6 July 2015 (UTC)
      • After discussing it with my fellow drafters, we have concluded we will not be dismissing the case at this time. First of all, cases are the place where we generally clarify and interpret policy. That's what principles are for. Furthermore, the fact we declared an amnesty does not necessarily imply there will be no findings of fact. We have not yet determined whether any will be proposed, but we are not going to rule out that possibility at this stage. Furthermore, dismissing the case now would waste a lot of time and useful input and may end up causing a loss of momentum. Salvio Let's talk about it! 13:06, 6 July 2015 (UTC)
        • Given the fact that pretty much all of the evidence submitted before the deadline was directed to the very actions for which you have just granted amnesty, I suggest to you that the time has already been wasted. The momentum never existed, so it can't really be lost. There is no policy for you to interpret, only your own procedure, which does not need a case. The committee cannot "revise" the Admins policy to take away the intentional equivalency of administrator decisions and administrator actions; the community put that in there, and Arbcom can't take it out. Only Arbcom, however, can change its procedures to either (a) explicitly state that administrator decisions at the AE noticeboard can only be appealed on the same basis as administrator application of sanctions or (b) explicitly state that administrator decisions at the AE noticeboard can be reversed at any time if a sanction has not been placed. (That's really what we're talking about here.) This isn't a policy thing, it's a committee procedure thing, and you certainly don't need a case for that, although a community consultation wouldn't be a bad idea. It is just plain mean to keep people tied to a case where there is no intention to sanction them. Stop being bureaucratic, and fix your procedures (either to confirm that only sanctions are protected or to give equivalency to decisions). Contrary to the belief of some, very few people are actually paying attention to this case; there are fewer watchers of the pages than there are editors whom you sanctioned in the initial motion, by a significant percentage. Risker (talk) 18:40, 6 July 2015 (UTC)
          • Yeah, the whole thing looks like a big waste of time now. But I suppose that's on par with the practice of AE in general, which has become as socially gameable and ineffective as the rest of Wikipedia's toothless dispute resolution mechanisms, so I really should have expected this. Serves me right to make the mistake of spending time to compile evidence that has now, together with the substance of the case, in effect been summarily dismissed. Maybe a future Committee may show some interest in having its decisions actually enforced, but right now it looks it's time for one more round of pointless process discussions and additional layers of bureaucracy, which, well, have fun with that.  Sandstein  19:48, 6 July 2015 (UTC)
Original announcement
  • Callanecc, quick point -- in updating WP:EDR, former community sanctions converted to ArbCom ones need to be listed under ArbCom sanctions (which you have done), but should then probably also be removed from under community sanctions?  · Salvidrim! ·  03:10, 17 July 2015 (UTC)
 Done, thanks. Callanecc (talkcontribslogs) 03:34, 17 July 2015 (UTC)
  • I've added IBAN and I-BAN to our glossary. It's a good idea to reduce jargon to make language more clear, but getting people to do it is quite another thing. There's an internal language here that is almost indecipherable if you are not part of the project. "Seeing as how the VP is such a VIP, shouldn't get keep the PC on the QT? Because if it leaks to the VC then he could end up MIA and then we'd all be put on KP!" --Hammersoft (talk) 19:34, 17 July 2015 (UTC)
The perspective is that if the user is unblocked in the future after appeal, there will still be restrictions in place. Keegan (talk) 07:05, 17 July 2015 (UTC)
As Keegan says. This is actually standard practise if a user's behaviour in a particular situation warrants a siteban, as a way of keeping them from successfully appealing and then returning to that situation to stir the pot. —Jeremy v^_^v Bori! 18:30, 17 July 2015 (UTC)
It's almost never done that someone banned for a valid reason will be allowed back with no restrictions whatsoever. In the case where those restrictions weren't set at the time of the ban, though, the future Committee, many if not most of whom were not there for the original case, have to go review that case and its context, discuss what restrictions should be placed, agree upon those, and so on. It makes things a lot simpler if those of us who have already invested that effort set those restrictions and they have the option of saying "We will lift the site ban, all other restrictions stay in effect." Seraphimblade Talk to me 20:35, 17 July 2015 (UTC)
Original announcement

Technical terminology/spelling

Remedy 1 says "desysopped". Remedy 2 twice says "desysoped". The former spelling is normal; would a clerk please fix the second remedy? Nyttend (talk) 18:06, 3 August 2015 (UTC)

So, desysopped is correct, Nyttend? I corrected the main case page. I don't think this detail was included on the notices I posted. Thank you for drawing my attention to this. Liz Read! Talk! 18:24, 3 August 2015 (UTC)
I've never seen "sysoped" or "desysoped" except in recent postings, now that the term "sysop" is becoming a lot less common <elderly person's voice>than in the old days</elderly person's voice>. Older WP pages, stuff from eight or ten years ago, always uses "(de)sysopped". Thanks! Nyttend (talk) 18:39, 3 August 2015 (UTC)
(edit conflict)Technically neither term is particularly correct English (as opposed to "has had administrator privileges removed" or something analoguous). "Desyso[p/pp]ed" is strictly Wikimedia jargon. But according the normal English construction, (see Doubled consonants), whether the p is doubled or not if a matter of ENGVAR. We should definitely favor consistency by sticking to one of them, but determining which one is preferable isn't so evident. I guess the former ("desysopped") is indeed the most common usage at the present time across Wikimedia.  · Salvidrim! ·  18:43, 3 August 2015 (UTC)
I know in my own posts, I've used both spellings depending on which one is being used in that particular discussion. I've see desysopped/desysoped more often that sysopped/sysoped. That is likely because sysop is also a noun while there is no similar term to describe an editor with removed permissions. Liz Read! Talk! 18:51, 3 August 2015 (UTC)
I agree with Salvidrim! (that looks like an emphatic statement) in saying that neither is more correct than the other; that's why I just said "normal". Perhaps I should have asked for a "change" with the second, instead of a "fix"? But now we're really getting into prescriptivism, instead of descriptivism that simply says that one spelling's more common than another. Ah well; at least there's no disagreement that the use of both spellings was a problem that needed to be fixed somehow. Nyttend (talk) 22:28, 4 August 2015 (UTC)
Sorry, my bad. Thanks for fixing it, Liz. Courcelles (talk) 00:19, 5 August 2015 (UTC)
It's probably influenced by the dialectal factor of whether you spell it "travelled" or "traveled", but "[de]sysoped" would naturally occur infrequently because it looks like it rhymes with "hoped".  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  23:24, 12 August 2015 (UTC)

Process clarification

There is a hole in Remedy #2. "If [A and B] then C1; elsif [not A] then C2" does not cover "if [A and not B]". Can, and if so how, can Kww get +editfilter without first getting +admin? DMacks (talk) 19:25, 3 August 2015 (UTC)

That is a valid question that is not covered by the two possibilities outlined:
a) If he is desysopped and is resysopped, the restriction will expire.
b) If he is not desysopped, he may appeal this remedy after 12 months to ArbCom.
But if c) he is desysopped and is not resysopped, can he get the permission again? @Courcelles:, @DeltaQuad: and @Thryduulf:, should remedy #2 be amended? Liz Read! Talk! 19:36, 3 August 2015 (UTC)
My reading of this updated text was that there is explicitly no way for Kww to regain +editfilter without +admin if desysopped, which appears sensible to me. Perhaps it might've been better to formulate the idea as: "Kww loses EFM permissions. If he is desysopped and subsequently passes an RfA, this remedy expires automatically. Kww can also appeal the removal of EFM permissions to the committee after 12 months". That way both avenues remain possible, if of course that is what the committee intended.  · Salvidrim! ·  19:50, 3 August 2015 (UTC)
The way I read what I wrote is that there is no avenue to regain the EFM flag without passing RFA -- for both remedies, RFA is explicitly the appeal mechanism, not the committee. (An analogue might be how in years past the committee would desysop with provisions to regain tools via RFA or Arbcom appeal, as you're aware, this is essentially no longer done; appeals are to the community at RFA.) All that said, ARCA is open for business, and Arbcom can change its own remedies. Courcelles (talk) 00:18, 5 August 2015 (UTC)
Courcelles, there is a discussion going on about this question at Wikipedia:Arbitration/Requests/Clarification and Amendment#Clarification request: Kww and The Rambling Man and I think it would be worthwhile to crosspost your comment there. Liz Read! Talk! 13:40, 5 August 2015 (UTC)

Wikipedia:Arbitration Committee/Noticeboard/Archive 10#Wikipedia:Arbitration/Requests/Case/Kww and The Rambling Man closed ends with:

Discuss this at: Wikipedia talk:Arbitration Committee/Noticeboard#Wikipedia:Arbitration/Requests/Case/Kww and The Rambling Man

but the discussion section here is actually:

Wikipedia talk:Arbitration Committee/Noticeboard#Wikipedia:Arbitration/Requests/Case/Kww and The Rambling Man closed

DMacks (talk) 19:31, 3 August 2015 (UTC)

Done. Liz Read! Talk! 19:38, 3 August 2015 (UTC)

Finding of Fact #4 refers to Philip Seymour Hoffman and Hugh Jackman. Is there any reason that these references aren't linked? עוד מישהו Od Mishehu 20:04, 3 August 2015 (UTC)

The final decision is taken from the Proposed Decision page and the names weren't linked in that finding of fact. The relevant articles in the decision were those where there was edit-warring and they were linked. But I will go link the actors' names right now, now that you've pointed this out. Liz Read! Talk! 20:16, 3 August 2015 (UTC)

Filter 661

Finding of Fact #3 says that:

In February 2015 he created filter 661 to prevent any user from attempting to add unsourced content to any article whose title began with "List of awards and nominations received..."

There is nothing in the Finding of Facts sections which would indicate that filter 661 is inappropriate. If there's a problem with filter 661, ArbCom should explicitly state so. עוד מישהו Od Mishehu 20:04, 3 August 2015 (UTC)

They did, by placing the quote above under the heading "Kww has misused edit filters". --Guy Macon (talk) 20:14, 3 August 2015 (UTC)
I would certainly enjoy reading an explanation of how an edit filter that prevented the addition of completely unsourced tables (not single entries, but entire tables) into awards articles was an egregious violation of any policy or a betrayal of the trust the community had placed in me.—Kww(talk) 02:05, 4 August 2015 (UTC)
Stop digging, man. Lay low. Edit articles constructively and non-contentiously. Rehabilitate your reputation, which right now is crap. BMK (talk) 03:02, 4 August 2015 (UTC)
What was egregious was your conduct in this matter, you used the block tools while clearly WP:INVOLVED in the article that you blocked The Rambling Man for, you continue to insist that you only used them in an administrative capacity, yet TRM's edits were not obvious vandalism when you reverted them, and you claimed BLP as an exemption for your edit warring on that article despite the fact the man had been dead for over a year. ArbCom did the right thing here. Spaghetti07205 (talk) 16:22, 4 August 2015 (UTC) Struck WP:Sock's post. Flyer22 (talk) 15:08, 5 August 2015 (UTC)

To answer Kww, first, we must acknowledge that Finding of Fact #3 does indeed say "Kww has misused edit filters" and that Filter 661 was indeed listed as an example. So the fact that Kww creating "an edit filter that prevented the addition of completely unsourced tables (not single entries, but entire tables) into awards articles" was misuse of filters has already been established.

That being said, arbcom failed to explain exactly how edit filter 661 was an abuse. I am going to give my opinion as to why (it seems pretty clear to me), but if it is important to know why, Kww should ask arbcom to clarify.

The basic issue here is that Kww was in a content dispute (that has also been established, so there is no point in Kww repeating the reasons he thinks it wasn't a content dispute) and was edit warring (also an established fact). Now let's assume for the sake of argument that Kww was entirely on the right side in the content dispute, he is right about the edits being unsourced and that he was right to remove them. There is no "but I am right" exception to either WP:EW or WP:INVOLVED. Kww is not allowed to edit war or use admin tools, even to support policies such as WP:V. (That BLP doesn't apply is also an established fact.) In this case, Filter 661 was designed for the specific purpose of preventing editors from making the same sort of edits Kww was in a content dispute over and was in an edit war over. Ordinary editors cannot create edit filters that allow them to get their way in a content dispute, and Kww creating such an edit filter was clearly an WP:INVOLVED use of a tool that ordinarily only administrators have. Thus the conclusion that "Kww has misused edit filters". --Guy Macon (talk) 22:32, 5 August 2015 (UTC)

User:Δ, last blocked (as User:Werieth) over a year ago, is requesting an appeal. Please see User_talk:Δ#Appeal. Thank you, --Hammersoft (talk) 20:35, 12 August 2015 (UTC)

I'm a bit confused by this request. While I can't go into detail about private communication, the last time that we communicated back and forth with this editor, responses were sent and responded to in turn, confirming that they were received. If we have been contacted since then, we have not received it. Seraphimblade Talk to me 20:59, 12 August 2015 (UTC)
@Hammersoft: Please see this page's editnotice. This page is to be used for discussion of formal announcements by the Arbitration Committee; general discussion should be held at WT:AC. Would you like me to move this section there? L235 (t / c / ping in reply) 23:20, 12 August 2015 (UTC)
  • It's not exactly clear though, is it? I look at {{ArbCom navigation}} and nothing pops out about a need like this. I look at Wikipedia:Arbitration Committee/Noticeboard and nope can't do it there. Even the editnotice here is equivocal at best with no clear instruction on what to do in this sort of event. A communication needs to be made to the committee about something with which they are directly involved. Yet, there doesn't readily appear to be any place to do it. Still, the announcement about the arbitration decision for which this event is connected was posted and amply discussed here on this very board. So why is there now a problem with discussing another point about it? --Hammersoft (talk) 00:06, 13 August 2015 (UTC)
  • @Hammersoft: The thing is that back then, there was an active announcement at WP:ACN about it, so there was a discussion about that notice on this talk page. I understand that it is confusing and if there's anything that you need or if you have a suggestion, please feel free to ask me, or the clerks or the Committee. (Particularly if there's something you think we should change on our templates or notices or something.) Thanks for your work! L235 (t / c / ping in reply) 01:54, 13 August 2015 (UTC)
  • Perhaps I'm not being clear. The instructions are not clear, and there does not appear to be a more appropriate place than here. Regardless, this is meta discussion. The point was to inform the committee, and I believe this thread is doing just that. Thanks, --Hammersoft (talk) 01:56, 13 August 2015 (UTC)

Level 1 desysop of Malik Shabazz

Original announcement

When did we start talking in "Level" terms? "Adopted on 23 April 2009"; was that the first? I don't remember (and can't find through Special:Search) anything about "levels" in the last admin rampage that comes to mind, Wikipedia:Wikipedia Signpost/2007-04-23/Robdurbar. Nyttend (talk) 13:34, 18 August 2015 (UTC)

@Roger Davies: is the arbitrator with the best memory for that sort of thing. Thryduulf (talk) 13:47, 18 August 2015 (UTC)
Possibly Sept 2009? Will double check. WormTT(talk) 13:52, 18 August 2015 (UTC)
I felt at the time the terminology was created, and still feel now, that "Level 1 desysop" should be cut and replaced everywhere with "Temporary emergency desysop", and "Level 2 desysop" be replaced with "Temporary desysop". Just suggesting a different terminology, not changes in the process. "Level 1 and level 2" give off a bureaucratic vibe, and only make sense to people who already know the jargon; "temporary emergency" means something in actual English, and should be clearer to most people. If people don't like "emergency", they could use "proactive", or "rapid", or something. But "emergency" covers it pretty well, IMHO. --Floquenbeam (talk) 14:22, 18 August 2015 (UTC)
I would agree. Using "Level 1" and the like is unnecessarily bureaucratic and hides the meaning, forcing the reader to go track down what that means, particularly since these are fairly rare instances. Using the real meaning in the title (emergency desysop) requires trivial effort and actually means something to the average, ordinary editor here. Dennis Brown - 14:29, 18 August 2015 (UTC)
Could not agree more and I hope that if Arbcom does review the procedures in light of this desysop that they will take that on board. WormTT(talk) 14:32, 18 August 2015 (UTC)
I think an effort to review and clarify the language of all Arbcom procedures is long overdue.  — Scott talk 15:02, 18 August 2015 (UTC)
I concur. -- Avi (talk) 15:16, 18 August 2015 (UTC)
We've planned a look at procedures but that obviously hasn't happened yet - I just hope it does before we lose more of the experienced Arbs. I agree that the terminology needs to be clearer. If anyone hasn't found it, it's at Wikipedia:Arbitration Committee/Procedures#Removal of permissions. And with all due respect to my colleagues, I'm not clear why this was one as a Level 1 rather than a Level 2, but because of different time zones some of us weren't around to comment. Doug Weller (talk) 16:25, 19 August 2015 (UTC)
Regardless of the terminology used, this procedure was not appropriate in the slightest for this case. The requirements for both level 1 and 2 are below:
  1. (a) an account appears to be obviously compromised, or is intentionally and actively using advanced permissions to cause harm in a rapid or apparently planned fashion, or (b) multiple accounts are actively wheel-warring.
  2. (a) the account's behaviour is inconsistent with the level of trust required for its associated advanced permissions, and (b) no satisfactory explanation is forthcoming.
I strongly suggest the community request each of the supporting arbitrators for level 1 desysop to come forth with the following things: An explanation of why they believed the account was compromised when an arbitrator publicly stated that the account was probably not; Proof the "incident(s)" were planned/rapid; Proof of wheel-warring; Proof of an explanation being asked for and not given in time.
Regardless of said 5 arbitrators ability to produce satisfactory explanation for any desysop, Level 1 was certainly not appropriate here. Level 1 exists as a way to, in an uncontroversial emergency, yank the tools as a halt to any extremely disruptive activity. Not as a "we need more time to think about this so let's yank now, ask questions later". I would like for someone to name one other conduct unbecoming style case that had a desysop at the beginning - and not a "please don't use your tools until this case is over, thanks - arbcom" style warning. — Preceding unsigned comment added by 2602:306:3277:670:C9FB:E8E5:68EF:F09B (talk) 15:30, 19 August 2015 (UTC)
An IP's very first edit is on an ArbCom board? Hmmm... Erpert blah, blah, blah... 03:23, 20 August 2015 (UTC)
AGF much? So many valid, innocent explanations, so few malicious ones. — Preceding unsigned comment added by 2602:306:3277:670:E9BF:E12B:DD73:A42A (talk) 03:55, 20 August 2015 (UTC)
Please name a few of the many valid, innocent explanations, and please use your account for commenting on arbitration pages. Unless it's blocked, of course (but I presume not, since that would not be a valid innocent explanation). Bishonen | talk 14:17, 20 August 2015 (UTC).
Accidentally being logged out, not having an account, using a public computer/connection not connected to usual IP, etc... Although the second one would seem to apply to me :) — Preceding unsigned comment added by 2602:306:3277:670:fd83:edc0:249d:f414 (talk) 12:39, 20 August 2015 (UTC)
Regardless of who the IP is or how they got here, they're making points that are worth some consideration. Was the account "obviously compromised"? To the contrary, Malik has made it clear that it really was him and no one seems to doubt him on that. Was harm caused in a "rapid or apparently planned fashion"? Apparently not; the general view is that it was a single, heat-of-the-moment action. So Level 1 appears to have been improperly applied. The Level 2 criterion "the account's behaviour is inconsistent with the level of trust required for its associated advanced permissions" arguably was met, so the case hinges on the qualifier "and no satisfactory explanation is forthcoming" (notice this is and, not either/or). Maybe this is being overly bureaucratic, but here we are. Short Brigade Harvester Boris (talk) 04:31, 20 August 2015 (UTC)
Reading the words in policies and believing that doing so will influence an Arbcom decision is a fruitless endeavour, Boris.—Kww(talk) 05:12, 20 August 2015 (UTC)

I can't speak for any other arbitrator, but my reasoning behind the level I emergency desysop was that, first off, we have an administrator acting totally out of character. Even if checkuser data doesn't show a compromise, that wouldn't rule out that, for example, someone else using the computer found the admin account open or with the password saved and is using it. Checkuser wouldn't and couldn't rule that out. With that behavior having already taken place, they violate a bright-line rule on admin tool use, the use of tools while blocked. At that point, I do believe that a precautionary desysop was called for. Level I is not intended to be permanent, just a precaution until we can sort out what's actually going on, reasonably verify that the admin is in fact in control of the account, and things have calmed down. If Malik showed any indication of wanting the tools back, there's a good chance they already would have been returned. Seraphimblade Talk to me 05:37, 20 August 2015 (UTC)

Reading the IP's words and the actual policies themselves, I have to wonder whether ArbCom, or the members who made the decision, has by that inappropriate so-called "Level 1 de-sysop", caused the loss of what I consider the best admin Wikipedia has. I'd post this sentiment on the RFAR itself, but I think my section is already too long as is. Feel free to quote me, though. Softlavender (talk) 05:39, 20 August 2015 (UTC)
Things may have gone better without referring to all this "level 1" jazz and so on. I wish arbcom had just said "OK, we're going to temporarily suspend Malik's bit until we can figure out what the hell is going on here." Short Brigade Harvester Boris (talk) 13:56, 20 August 2015 (UTC)
Agree entirely with Boris. ArbCom's legalspeak gets worse and worse. Suggestion: the committee should select a member to be the "ombudsman for talking like a person". I nominate User:Doug Weller. Bishonen | talk 14:22, 20 August 2015 (UTC).
Seraphim's the only arbitrator that's actually responded to anyone asking about why level 1 was appropriate here. Yet their explanation shows that they felt it was appropriate - and furthermore why the tools have not been restored.
Is that a response to me? And is there a valid innocent reason why you don't sign your posts, 2602:306:3277:670:E9BF:E12B:DD73:A42A? It's very inconvenient. You don't otherwise sound like a new user, and that's putting it mildly. Bishonen | talk 17:40, 20 August 2015 (UTC).
Probably just another incarnation of 173.228.123.193. Clearly an experienced user, undoubtedly logging out to avoid scrutiny, but it doesn't seem to bother Arbcom to discuss issues with him. My efforts to get a checkuser run were thwarted.—Kww(talk) 18:47, 20 August 2015 (UTC)
Original announcement

Arbitration motion regarding Malik Shabazz case request

Original announcement

Procedures

Please see discussion at Wikipedia talk:Arbitration/Requests/Case/Palestine-Israel articles 3#Why do some editors get more space to make their points? --Guy Macon (talk) 20:21, 25 August 2015 (UTC)

RFC: Arbitration Committee Elections December 2015

The yearly Arbitration Committee Election request for comment is now open. All editors are invited to participate. Mike VTalk 04:18, 28 August 2015 (UTC)

Motion: Longevity

Original announcement


Coat of Many Colours

Original announcement

Just curious, is there something publicly available about this decision, or was it handled in camera? BMK (talk) 21:47, 30 August 2015 (UTC)


There is nothing public --Guerillero | Parlez Moi 00:09, 31 August 2015 (UTC)
Thank you. BMK (talk) 00:52, 31 August 2015 (UTC)

Motion: Activity

Original announcement
David Fuchs's and Ks0stm's OTRS access for oversight has been removed accordingly. Thanks for your service. Keegan (talk) 22:22, 31 August 2015 (UTC)

Arbitration motion regarding Argentine History

Original announcement

Arbitration motion regarding Christianity and Sexuality

Original announcement

Motion: AUSC Extension

Original announcement
  • Now knowing this, will they be asked as a courtesy? Otherwise I would perhaps suggest that reforms be introduced after a new AUSC is put in place. Mkdwtalk 02:40, 5 September 2015 (UTC)
  • They need to be asked before we extend their terms. I would rather scrap the subcommittee entirely, but want to do so only after we're clear on who will handle issues that the AUSC currently addresses. I'd prefer have the discussion about potentially replacing the AUSC happen without appointing a new group entirely, though of course if consensus is in favor of continuing the AUSC, I will defer to that. GorillaWarfare (talk) 02:46, 5 September 2015 (UTC)
  • Before the community can make an informed recommendation, we need to know how many requests were made to the AUSC this year, and how many of them were considered to be "in scope". Of those that were in scope, how many resulted in actions, recommendations, suggestions for improvement (either individually or to the applicable functionary group). The only one we know of publicly is that relating to a specific Arbcom case, and the AUSC's recommendation (to give a warning, essentially) was ignored by the Arbitration Committee. Risker (talk) 02:53, 5 September 2015 (UTC)
  • Yes, I have emailed AUSC asking for updated statistics to be published. However, it would help if a decision were made on the principle of the question first. No pattern applies to the frequency of complaints, nor do temporary fluctuations in the subcommittee's activity change the fundamental and theoretical question: if there should be independent scrutiny by the community of advanced permissions. AGK [•] 14:34, 5 September 2015 (UTC)
That isn't the fundamental question. The fundamental question is "is there value in having project-specific scrutiny of advanced permissions by a handful of community members". At this point, nobody has answered that question. There is already independent scrutiny through the Ombudsman Commission. Risker (talk) 05:51, 7 September 2015 (UTC)
Yes it is – it is the same thing Risker. "Independent" == community. You are not being very helpful here; please try to be.

The OC is a corporate function, and does not act for the enwiki community. AGK [•] 07:53, 10 September 2015 (UTC)

The AUSC is pretty much an entirely useless body. The only power they actually have is to screen complaints about Cu/OS use and refer the ones having merit to Arbcom proper. And there just aren't many: the only matter of any significance this year was referred to them for an opinion by Arbcom itself, then they reported back to Arbcom (See the Sockpuppet Investigations Block case for how that worked).

Calling them "community seats" these days is utter nonsense, given that Arbcom appoints them. Arbitrators hold half the seats, so there's no independence whatsoever. Arbcom itself could handle the entirety of the complaints -- and actually reduce ArbCom's workload. The entire process to select the AUSC members takes up far more man-hours than the AUSC generates, especially since Arbcom gets the complaints that have merit anyhow (Most of the meritless complaints (which are most of them) are so lacking in merit they take a minimal amount of time to consider.)

The one thing I've noticed in my four years on it, is that the Oversight-l list is a far, far more effective body for reviewing the use of the OS tool than AUSC. Matters are resolved quickly, with vigorous discussion when they arise that often settles matters in a day or two. I wrote a proposal to formally transfer the review functions of the AUSC to the oversight-l list, where any arbitrator could escalate it if it was bad enough to, in their opinion, to warrant any form of disciplinary action or decision of best practices. There's no reason a checkuser-enwp couldn't do the same thing. There is always the Ombudsman Commission as a failsafe. Absent OC action, serious complaints will always end up at Arbcom unless a completely independent body is set up with the actual authority to remove the flags. What we're doing now is endless hours of bureaucracy that accomplishes nothing. And if such a body with actual power is desired, it would have to come from the community rather than by Arbcom fiat. Courcelles (talk) 03:04, 5 September 2015 (UTC)

It is entirely useless because when the committee created it, we curtailed its power out of a misplaced power hunger or fear of delegation. The body is ineffective because it has been improperly established. The real question is, do we want CU and OS to be scrutinised by the community? If so, AUSC needs the independence and authority it currently lacks. If not, it should be abolished. Which side do you fall on, Courcelles?

The alternative you suggest is an invitation to disaster. The functionaries should not oversee themselves, and there is no evidence whatsoever that the community would support an advanced permissions oligarchy.

I suggest we divest arbitrator membership from AUSC, elect its members directly, and amend ArbCom procedure so that we always delegate final decision-making to the auditors. AGK [•] 14:31, 5 September 2015 (UTC)

  • You acknowledge that what we have is useless, though. It at least paid some lip service to community oversight when the re were direct elections, but the 2010 ArbCom removed those for functionary selection. We need an oversight authority, but right now the only ones with the experience to do so are either the existing functionaries or Arbcom itself, and Arbcom is the body with the community mandate to make decisions about access to advanced permissions. The recent SPI Block case showed the strength of the functionaries and ArbCom to deal with abuse of the tools, and the utter uselessness of the AUSC. I can recall them making two recommendations to Arbcom, and both were acted contrary to the recommendation. Could we elect a body as you describe? Probably, though I think that would require a community consensus since global policy designates Arbcom as the appointing body of CU/OS without a community preference for another method. But, no, I don't think we necessarily need some separate body with the power to discipline holders of CU/OS flags, there isn't anywhere near enough work to make it anything other than a sinecure, and where are you expecting these candidates to come from? We have a hard time getting people to stand and do the work that already exists, be it at RFA, SPI, OTRS, etc. New committees proliferate the amount of seats needing to be filled, but it doesn't increase the number of people who care and are willing to go through the election process for a very low-duty work. Such people would spend many more hours on running for the seat than doing anything once in the seat. (So, yes, I want to abolish AUSC with absolutely no formal replacement, but see this as an increase in the quality of tool oversight.) Courcelles (talk) 19:53, 5 September 2015 (UTC)
  • I acknowledge it could be better and I propose we act to make it better – not that we act to disband the body altogether. The incremental, disjointed scope and authority of the body explains its unwillingness to get involved where you would expect it (e.g. OS handbook). Would you try to get involved if you'd be told to stop overstepping the limited job ArbCom gave you? Case in point: auditors raising complaints themselves is, in itself, a workaround to its stated purpose. The only real authority we have given to AUSC is to hear complaints brought by others.

    We seem to have no trouble getting people to volunteer for the OC or ArbCom (albeit the former draws from other projects too), and in my judgement the shortage of AUSC volunteers is actually because those in the role cannot effect change – i.e. the role is unrewarding. Incidentally, I disagree with you that the current AUSC members are unfit for purpose. They are all experienced, reliable volunteers, and in any case the principle (there should be independent auditing) is just as important as the practice (whether any particular AUSC has passed findings with which you or I disagree).

    Yes, the most recent high-profile audit was led by ArbCom and the functionaries. That doesn't mean it was right to sideline the community's auditors, and it simply speaks to my point that AUSC is useless because we have sidelined it, in practice and in the very procedures defining its role. AGK [•] 13:21, 6 September 2015 (UTC)

  • (ec) As the longest-serving community member, I thought I might have some insights worth sharing. AUSC was created at a difficult time for ArbCom when a significant portion of the community had lost faith in the Committee's ability to police functionaries and respond quickly to urgent matters. I was not privy to ArbCom's deliberations at the time, but I suspect those factors figured into the creation of a 6 person AUSC with 3 non-arbitrators. AUSC's mission was not well defined, and I think it's fair to say that AUSC has not attempted to expand or aggressively define the scope of its mandate. To that end, AUSC has not employed all of the means it might have to detect abuse and has mainly relied upon functionary self-reporting and user complaints to set its agenda. To Courcelle's point that AUSC is useless, I would agree that AUSC is useless if the goal is to quickly resolve complaints with a minimum of bureaucracy. I'm not sure that is the goal though. The community members, as originally envisioned, and even now with some members who do use the tools on a day-to-day basis, serve an independent function. They are designated persons whose job it is to make sure that ArbCom does not sweep complaints under the rug, either of the misconduct of its own members or of favored functionaries. That AUSC has not had to challenge a functionary or ArbCom cover-up is not evidence of its uselessness, but rather is evidence that functionaries almost always perform their work in a forthright manner and that cultural (Oversight-L) and administrative (ArbCom) norms serve as a check on misconduct or incompetence that may occur. This role could conceivably be replaced by the OC, and I realize my perception of AUSC as an en-wiki-centered body that is responsive to en-wiki norms is colored by my participation in it, but I'll leave it to others to contemplate alternative models. MBisanz talk 14:59, 5 September 2015 (UTC)
  • On a related note, I just realized that it has been over 40 years since my seatbelts actually did what they were designed to do, and I have never used my life insurance. My question for those who say that AUSC is useless and should be abolished is this: should I cancel my life insurance and cut out my seatbelts, or should I do my best to make sure that when the time comes that I need them they will be able to do the job? --Guy Macon (talk) 12:02, 10 September 2015 (UTC)

(e/c) Part of the issue is having a body, independent of the checkusers, with expertise in the custom and practice of checkusing (that is likely to create some balance in the checkuser's action up-front - eg, it's good the ctte doesn't have to discipline, but its presence is still likely to promote self-discipline n checkusers) - such independent expertise formation, is not likely to come from anywhere else. Alanscottwalker (talk) 15:07, 5 September 2015 (UTC)

  • What AUSC was envisioned to be back in 2009 (which was an investigatory body that would do a lot of the legwork to collect sufficient evidence to remove problematic CU and OS in a manner which would be accepted by the community, and to act as a watchdog) fell apart pretty quickly. It has become clear over the years that many in the community see AUSC as one of the most certain entryways into holding checkuser and oversight permissions, whether as CU or OS independently, or as members of arbcom. That's not necessarily bad - it gives those individuals a taste of what is involved and only those who find it really interesting will follow this path, and they do have additional experience and knowledge when they put themselves forward as CU/OS/Arbcom candidates. On the other hand, the last pro-active AUSC, one that actually looked at CU/OS activity and raised questions independently, was...in 2009-10. Its membership has not in any way advocated for any changes or improvements that relate to CU or OS (e.g., it has not participated despite requests in updating the Oversighter "handbook", it does not participate in any way in setting standards, it does not review the activity logs, and it is not even AUSC members who update the activity stats). In most years, there is one or maybe two active community members, and the rest are largely inactive; I can't remember the last pro-active AUSC member who was not an arbitrator. I may have at one point supported the AUSC being an elected community only body, if it had proven its worth as an oversight body. But after the Sockpuppet case, where they could not even steel themselves to recommend tool removal from a checkuser, and Arbcom had to do the right thing and take those tools away, I do not have faith in the AUSC to actually do what the community needs it to do. That case was brought because *checkusers* didn't think the behaviour was appropriate, and because *checkusers* knew that there was a pile of nonsense in that SPI.

    A question to AGK: would he be satisfied with a community group separate of Arbcom removing tools from arbitrators? If so, on what basis can it be justified to continue automatically handing out OS and CU to arbitrators? Because that's all part of the same issue. Risker (talk) 15:39, 5 September 2015 (UTC)

  • That wouldn't be necessary. Arbitrators are already held to account in elections (or through their elected colleagues being able to recall them). The issues are not as intrinsically connected as you say, because it is only functionaries who operate without direct community supervision (or who would do if AUSC was disbanded). The nature of arbitrator CU/OS use differs very much from non-arbitrator use. AGK [•] 13:21, 6 September 2015 (UTC)
Erm, no. It is the same thing. Abuse of the tools is abuse of the tools. If a newly elected arbitrator abuses the tools, the community has no opportunity to do anything about it for two years. In fact, the policy remains that arbitrators may retain the tools after their term ends, at which time they need to meet the same activity standards as other CU/OS. There are always a few arbitrators who do a significant number of non-arbcom checks, in particular. (Nobody's all that worried about suppressions, and the oversight team has long reviewed the actions of arbitrators in this area.) So are arbitrators going to be held to the same standards as non-arbitrators? Would this hypothetical New and Improved AUSC be able to remove tools from arbitrators? Risker (talk) 14:59, 6 September 2015 (UTC)
So… it would be great if you read what I actually wrote. I'll simplify. Non-arbitrators would be held to account by elected auditors. Arbitrators by elected arbitrators. Or would it also be suboptimal that the elected auditors are only re-elected every year? What if there were 18 months between elections? 19?

Incidentally, contrary to your point there is no special extra supervision of arbitrators by the OS team. I have identical access rights and subscriptions to you, so unless there are there a hidden set of auditors this is incorrect. AGK [•] 21:38, 6 September 2015 (UTC)

I did read it. So your proposal is that arbitrators are responsible for overseeing their peers for CU and OS, and the AUSC would oversee the rest of the CU/OS. Given that arbitrators make up 30% or more of the advanced permission holders (and some of them are amongst the more active CU/OS, at that), the AUSC would be hamstrung. No, either they're responsible for monitoring everybody or nobody. I don't see any value at all in treating arbitrators as special snowflakes when it comes to the use of these permissions, and I'm quite certain the community would agree with that. My opinion is that AUSC should be abolished, but if it is retained it should be responsible for making recommendations on *all* rights holders. That (despite efforts on several occasions) the AUSC through several iterations has steadfastly ignored the requests and the existing mandate to take responsibility for collating and posting tool usage reports, to audit tool usage logs, to recommend best practices...well, I know that Arbcoms in the past have encouraged it, and I assume that the same is true over the last few years since I left the committee. It does have a warrant for action, but it has never actually used it; not since 2009-10. Risker (talk) 05:51, 7 September 2015 (UTC)
You assume incorrectly. You also are seriously misrepresenting the figures. Arbitrators are 30% of the checkuser group but do not perform anywhere near 30% of its checks (or suppressions). In fact, I did the numbers myself; they perform barely 10% of checks. AGK [•] 07:53, 10 September 2015 (UTC)
  • You're actually making one of the points against the AUSC == the new appointees ae more often than not complete newbies to the operation and practices of CU/OS. This isn't say, (I'm not liking the AUSC/functionaries to police) a police internal affairs department, where all the officers appointed have years of experience doing the actual job, most of the auditors in recent memory had to learn the job while learning to provide a token level of oversight to the processes. Courcelles (talk) 19:07, 5 September 2015 (UTC)
Well, no the analogy would be to a citizens review board not internal affairs. 'Who watches the watchers' goes back, at least, to Ancient Rome. Getting into the minutia of it is something they have at least agreed to do (and presumably have some interest in doing), and they have the position to do so -- if need be, they can raise the alarm and building in redundancies over the sensitive/technical is regularly done. Alanscottwalker (talk) 02:20, 6 September 2015 (UTC)
  • One of the things an audit subcommittee can do is perform a regular audit on the use of checkuser and oversight.
Of what would such an audit comprise?
  1. A detailed report on the assignation and use of the tools.
  2. An investigation into complaints, looking for other examples of the same behaviour.
  3. Checking the documentation of sample oversight and checkuser uses to confirm that there was no CoI, that policy and procedure were followed, and that correct logging of reasons was employed.
This would increase confidence in functionaries. It would also mean that people would be happier raising abuses, knowing that they would be taken seriously. For example I saw a minor (and probably unintentional) checkuser abuse recently, I didn't bother mentioning it, because previous response has been a combination of circling the wagons, and WP:IDIDN'THEARTHAT. If I could trust AUSC to respond if the functionary didn't then I would think these minor abuses worth dealing with. In my experience minor abuse leads to major abuse after a while.
In addition there is scope for improving the checkuser tool to improve transparency and accountability, as I outlined previously. Recommendations from the AUSC would carry some weight with WMF.
All the best: Rich Farmbrough, 19:23, 6 September 2015 (UTC).

Part of the problem is that per the m:CU and m:OS global policies, stewards will only accept removal requests from ArbCom or a vote from the community. I do not believe they would accept an AUSC request as AUSC does not fall into either of those categories. --Rschen7754 22:27, 6 September 2015 (UTC)

That would seem like the stewards making up rules. m:CU states that "ArbCom can decide on the removal of access". Can being presumably distinguished from must. And AUSC derives its authority, on enwiki, from ArbCom; under the arbitration policy ArbCom may delegate its authority to another body, which from then acts on behalf of the committee itself. The stewards would presumably not wish to override local policy decisions, especially without a clear basis in global policy.

Should the WMF/Meta expressly assert otherwise, ArbCom can always rubber stamp the removal of permissions. AGK [•] 22:39, 6 September 2015 (UTC)

On the contrary, this would be making up rules as well; since AUSC or other groups not representative of the rest of the community are not explicitly given the ability to remove CU or OS, the stewards would decline such a request. Stewards also do not recognize the arbitration policy, and tend to err on the side of being conservative especially with policies involving privacy (i.e. not allowing for the regranting of resigned CU/OS permissions unless explicitly authorized by local ArbCom or local policy). --Rschen7754 22:48, 6 September 2015 (UTC)
Bottom line, there is no provision at all in global policy (which cannot be overridden by local policy) for Arbcom to delegate removal of permissions (or, for that matter, granting of permissions) to any other group excepting the community as a whole. AUSC, no matter what configuration is chosen, is not the community, even if it is made up solely of community members. That is, incidentally, why it is set up the way it is. Global CU and OS policy has not had any significant changes since the AUSC was created. Not even the Ombudsman Commission can require that someone have CU or OS permissions removed; they make a recommendation to the Board of Trustees. Arbcom has always had the option of simply rubber-stamping the AUSC's recommendations, and to my knowledge that's pretty much what it has done, most of the time - except in the most recent case, where it overruled the AUSC and opted for significantly more serious sanctions. (I agreed with Arbcom's decision, which is no surprise since I brought the case.) So you are pretty much left with either disbanding the AUSC or leaving it as status quo (i.e., without direct power but only recommending action to the Arbitration Committee). Of the two, I'd rather see it disbanded. Instead of running for AUSC, those candidates can be considered for CU or OS directly. Risker (talk) 05:51, 7 September 2015 (UTC)
The AUSC was created because a significant portion of the community had lost faith in the Committee's ability to police functionaries; that is why it included appointees. A series of bad decisions taken on political grounds undermined faith in both ArbCom and the way it is constituted. At the same time ArbCom claimed that it was far too busy with CU work to perform the role that it was actually elected to do. Now arbs are claiming that there is actually very little work in it and they could do it. In the minds of many, rolling back this first tentative step at reform by disbanding AUSC will only confirm what they think they already know: that ArbCom is at best incompetent and at worst corrupt. Hawkeye7 (talk) 10:37, 7 September 2015 (UTC)

"and asks for community input on how they would like to see the Subcommittee function in the future". I presume you're asking for said feedback here, and not at some other venue? If I'm wrong, that venue should be publicised in the announcement. --Dweller (talk) 12:04, 7 September 2015 (UTC)

Yes, please --Guerillero | Parlez Moi 17:13, 7 September 2015 (UTC)

I'm dropping back from inactivity to make this comment, but I'm still inactive so might not be able to respond. Sorry for the length! When I nominated to join AUSC as a community member I thought it would be a body which was able to set its own agenda and standard as separate from ArbCom, however the past year has shown me that almost the opposite is true. AUSC as a body is dependent on ArbCom who (IMHO) have shown little interest in providing muscle to AUSC's investigations (referring here to inactivity primarily) and to enforcing a high standard of functionary activity (except SPI block case). Given that I don't see a continuing purpose for the AUSC unless there are major reforms, primarily that it does not have arbitrators sitting on it (even if members are still appointed by ArbCom) and that it's decisions do not require ArbCom to decide what action to take but rather are rubber stamped by ArbCom (only if ArbCom still needs to remain involved, per global/WMF policy). Here are some points to support that:

  • Risker mentions above "the last pro-active AUSC" was years ago, I disagree to some extent. Since 2014 I've been looking fairly regularly at functionary actions but as I've previously mentioned (after SPI block case I think) it's impossible to check a large enough number of them to have an effect, therefore the sub-committee has to rely on reports from functionaries and the community.
  • My feeling, from the last year or so on AUSC, is that there is very little appetite among the community members and both sets of Committee members to act in a proactive manner by more rigorously setting and enforcing a standard. Rather AUSC tends to see things as within policy in good faith (or because the CU/OS thought it was) unless they are blatantly not, such as the WMF staffer inappropriately using OS (which isn't IMHO the primary purpose of AUSC).
  • As was seen in the SPI block case AUSC tends to stick to a middle ground (although when the PD was posted it did seem pretty clear that we didn't have all the facts). That case also showed that ArbCom was running their own investigation anyway so the AUSC investigation was largely a waste of time.
  • There have been a few times in the past year where I've thought a functionary's action was too far, outside best practice or outside of policy, or that the action taken by AUSC should have been stronger but I was in a minority of the sub-committee (see next dot point) so no action could be taken.
  • It's also unclear how much evidence (and what type) is required to remove the tools or to warn or remind or whatever. That's because there isn't much precedent and it's not up to AUSC to determine how much evidence is necessary. Added to this, its unclear how much weight AUSC recommendations are given by ArbCom. If the authority to remove CU/OS rested with AUSC that would be easier to determine. Currently we have a situation where the body which does the investigation has to guess at what the Committee of that year sees as acceptable and what they don't.
  • If AUSC believes there is tool abuse by an arbitrator it is almost impossible for the sub-committee to do anything concrete about it since we'd (AUSC) would need to report it to the Committee the arb is on and there really isn't a defined way for the Committee to remove the CU/OS from a sitting arb. So AUSC just concretes itself into the middle ground where they avoid making a decision (which is a fairly regular occurrence).
  • I guess there is also an inbuilt preference not to do things AUSC can't do by itself. For example, AUSC can't remove the tools so they don't recommend it, instead they recommend a warning as they can do that themselves without needing an ArbCom vote (although that is debatable).
  • Throughout the past year as a community member I've suggested that AUSC release statements of best practice (a suite of them following the SPI block case investigation) be released. That hasn't happened primarily because it was put off by some AUSC members and it was unclear who had the authority to release them - that majority of AUSC members, all six AUSC members, ArbCom, consensus of the functionaries or the community (as it would be creating policy).
  • One example of AUSC not being able to do anything is something which should be simple. ArbCom procedure is crystal clear on required functionary activity levels however of the four (I think, maybe more) reports of functionaries not meeting activity requirements only once was (as far as I know) action taken (and that was the form of an initial email to the inactive functionaries). I gave up working out who was inactive and reporting it to ArbCom since nothing happened, and when I asked if AUSC could make the two attempts at contact it was met with silence.

I know I've made the same point more than once but hopefully I've found some different examples of what I'm talking about. From my point of view I can't see a purpose for AUSC in its current form and membership (including me, as we are all used to this version of the AUSC) as it is toothless and tends not to act (possibly because it can't do anything itself). To link to a previous comment, I primarily agree with Courcelles' position above.Callanecc (talkcontribslogs) 08:13, 8 September 2015 (UTC)

@Callanecc: Sadly, I think the only way to force ArbCom to enforce its own inactivity policy consistently is by asking them publicly or on a list like functionaries-en about specific functionaries not meeting the policy. Which is sad, because it does mean the airing of such matters in public which could be embarrassing for the functionaries in question. --Rschen7754 21:55, 13 September 2015 (UTC)
@Rschen7754: I am working on my second batch of inactive functionaries for this year. Things like this take time --Guerillero | Parlez Moi 22:57, 13 September 2015 (UTC)
It probably shouldn't -- couldn't a bot assemble the necessary data for committee review? I thought I saw reference to a beta wiki the other, but I can't find it. Anyway, you could probably get a bot developer to make you one if they had access to a test wiki. NE Ent 23:18, 13 September 2015 (UTC)
@NE Ent: The data is here --Guerillero | Parlez Moi 23:28, 13 September 2015 (UTC)
Not following -- if the stats are available what's the time sink? NE Ent 23:34, 13 September 2015 (UTC)
@NE Ent: Sending the warning emails, waiting for replies, drafting a motion, getting over half of the arbs to support the motion --Guerillero | Parlez Moi 23:51, 13 September 2015 (UTC)
IMHO the warning emails should be canned and the rights should just be removed after 6 months (to allow for borderline cases), especially since ArbCom can just regrant the rights at will anyway. --Rschen7754 23:54, 13 September 2015 (UTC)
(edit conflict)Sounds like bureaucrat work more than committee work -- just pass a motion directing the bureaucrats to remove the permissions for inactivity exceeding the specified time. They already track that stuff for admin inactivity desysops. NE Ent 23:56, 13 September 2015 (UTC)
CU and OS are the right of the stewards to grant and take away. Due to some past friction, I would be very surprised if they would act on a carte blanche motion to pull bits after 6 months due to our exemptions for arbs, auditors, and OC members and the English Wikipedia's moodiness to things that happen on meta. --Guerillero | Parlez Moi 00:01, 14 September 2015 (UTC)
  • Looks to me like inactive CU/OS (and Stewards, OTRS users, etc.) are going to be removed across all Wikimedia projects anyways if they fail to sign the WMF's new confidentiality agreement by the end of the year... (link)  · Salvidrim! ·  22:44, 18 September 2015 (UTC)

First of all, a gentle but firm reminder that Checkuser (CU) efficacy / accuracy is an assumption, not an empirically tested fact. Please see Wikipedia_talk:CheckUser/Archive_4#Checkuser_accuracy for prior (attempt at) discussion. This is not to say CU should not be used, but that its use should be accompanied by humility and an open mind.

  • The notion that an individual must be a current or former en-wiki CU to be competent to function on an audit committee is a unwarranted wiki-insider viewpoint. Without going into great detail -- because I don't want a lame "enabling socks" block -- the protocols by which web browers and web servers communicate are in no way secret; wikimedia itself open source software, and there were Request for Comments long before there was a Wikipedia (1969 it says there.)
  • The mere existence of an audit committee is important to both the CU operation and the community; audits are supposed to boring and uneventful. My local government often stations visible police on highways during holidays to reduce traffic violations; their mere presence has an effect even if few arrests are made.
  • Part of distrust issue is likely because WMF rules restrict CU access to folks who have passed Rfa or equivalent, so audit committee members are seen as wiki-insiders.
  • So what we is a mostly "outsider" group of auditors. On the other hand, a committee made up entirely of editors with no CU or arbcom responsibility could easily devolve into a unnecessarily critical second guessing, Monday morning quarterback panel.

So my specific recommendation is to have the community directly elect 2/3 of the panel and have committee members make up 1/3 of it. I'd also suggest sufficient numbers (e.g. 10 and 5) so that individual concerns could be reviewed by randomly assigned audit subcommittees of available members (e.g. three member panels of two elected one arbitrator). ("Available" determined by self-declaration of the individual during any given time period.) If a audit subcommittee finds possible misuse of the CU tool, they could then refer to the arbitration committee for resolution. NE Ent 21:47, 26 September 2015 (UTC)

Template:ACAmotion for deletion

This template has been nominated for deletion. As this template appears to be used by the ArbCom, I decided to notify at this noticeboard. --TL22 (talk) 13:42, 27 September 2015 (UTC)

I just deleted it --Guerillero | Parlez Moi 20:19, 27 September 2015 (UTC)

Indefinite ban of Tarc

Original announcement

So, no details of this "off Wiki motion" will be disclosed, so that no public scrutiny is possible? I have always felt that Wikipedia business should stay on Wikipedia, rather than decisions being made behind closed doors. ~Amatulić (talk) 23:59, 18 September 2015 (UTC)

Tarc said very bad things off-wiki about individual wikipedians. Repeating them here would further victimize them. If you're interested in specifics, contact Tarc. He may be happy to continue his harassment. 177.154.145.99 (talk) 00:19, 19 September 2015 (UTC)
(edit conflict)Sometimes that's just not possible/not the best scenario for those involved. Discussing the motion on-wiki would likely require the arbs to discuss the nature of the incident. I can only imagine that someone who was subject to such gross harassment wouldn't want to have it aired out in the open and accompanied with community commentary. They'd have to relive the incident all over again in public and bring it to the attention of others who wouldn't even have known about it. I appreciate transparency as much as the next person, but there's a valid reason why we've appointed select community members to handle privacy concerns off-wiki. Mike VTalk 00:37, 19 September 2015 (UTC)

Personally, I don't doubt that this was a reasonable action, but how is the community supposed to verify that? "Off-wiki motion, ban, everyone move along now" is not a healthy solution for the community. It has the feel of a secret trial followed by a mysterious disappearance. An approach like that should be used only when there are serious privacy-related concerns. Perhaps that is the case here, but if so, it should be stated. Of course total transparency is not always possible, but we have to strike some sort of balance so that things aren't totally opaque. Everyking (talk) 00:51, 19 September 2015 (UTC)

@Everyking: To answer your question, in my opinion, this is one of those times where we are taking privacy concerns over transparency. -- Amanda (aka DQ) 03:24, 19 September 2015 (UTC)
As long as at least two or three trusted individuals (as in "we trusted them enough to elect them") agree that the action is needed and that we should not know the details, I am fine with the lack of transparency. Just don't let it ever get to the point where one person makes the decision without any review. That tends to end badly. --Guy Macon (talk) 04:23, 19 September 2015 (UTC)
@Guy Macon:; By my records, this matter was open for discussion for just over two weeks. Any arbitrator who wished to disagree with this action had sufficient time to do so. (none did) (Two weeks isn't some great achievement in timeliness, but I raise it to say this wasn't rushed through under the proverbial cover of darkness.) Courcelles (talk) 05:52, 19 September 2015 (UTC)
That was my impression as well, but it's good to have somebody actually check. I have never seen a shred of evidence of arbs, admins, crats or anyone else doing things "under the proverbial cover of darkness". That's why I accept statements that amount to "we decided to do X, and we can't reveal all the evidence we considered while making that decision." I trust arbs, checkusers, etc. to be open where they can, and to make sure that a few different sets of eyes are on the situation where they can't. --Guy Macon (talk) 16:21, 19 September 2015 (UTC)
At the very least, shouldn't this be spelled out when the decision is announced? I feel like the ArbCom is far too secretive in general, even in ordinary matters where privacy isn't an issue. If the ArbCom practiced transparency in general, I think there would be a lot more confidence in the ArbCom when it said, every once in a while, that the details on this one have to be kept confidential. Everyking (talk) 07:44, 19 September 2015 (UTC)

I have no idea what happened here, so I'm just speaking of general principles. In the case decision it states "Furthermore, only issues directly affecting the English Wikipedia can be considered and resolved." But here the Committee is explicitly dealing with issues off-Wikipedia. I know it has before in cases of outing, of course, but in the GG case the distinct impression was that the many, many off-wiki antics of GG were to be disregarded when deciding what happens on-wiki. I regularly see off-wiki behavior from Wikipedia editors directed at other Wikipedia editors on Reddit, Twitter, and the various -chans. Can the Committee provide administrators attempting to police this topic area (which is, unfortunately, only me most of the time) some guidance about what off-wiki behavior we can and cannot take into account? Gamaliel (talk) 01:40, 19 September 2015 (UTC)

@Gamaliel: I don't know either the GG or LB cases, so I can't comment on those, but as I see it (individually, not the committee) when behavior affects someones ability (aka not option, like by outing them, especially to peers of Wikipedia) to edit Wikipedia, that's enforceable grounds. But there also must be a clear (and my committee members will very likely vary on the level required) connection between their account and the off-wiki venue. My standard for this is beyond reasonable doubt, others are higher. I encourage other Arbs to comment with their position here. -- Amanda (aka DQ) 03:37, 19 September 2015 (UTC)
There should be zero tolerance for outing, of course. There is some off-wiki chatter that this matter was a case of outing, and if it were wouldn't question the matter, because it's very clear what is and is not acceptable. But the announcement here specifically says "harassment", which is pretty broad and can encompass a lot, hence my confusion about what to do when I see behavior that can fall under that umbrella. Gamaliel (talk) 03:43, 19 September 2015 (UTC)
Yep, and I can understand that. I won't comment on the specific case here, but I thank you for asking the question nonetheless. -- Amanda (aka DQ) 04:06, 19 September 2015 (UTC)

Odd that off wiki harassment is considered here, and not in the case of, say, Lightbreather. PeterTheFourth (talk) 03:00, 19 September 2015 (UTC)

At the risk of stating the obvious, off-wiki harassment was extensively considered in the Lightbreather case, by both Arbcom and the functionaries group. As a result of this consideration, one editor was banned for engaging in off-wiki harassment of Lightbreather. Accusations were also raised against a second editor, but despite exhaustive discussion there was no consensus on a finding against them. -- Euryalus (talk) 08:33, 19 September 2015 (UTC)
@Euryalus: Final decision: Lightbreather sitebanned, topic banned, account restricted, subject to 1RR, 'reverse topic banned', and interaction banned. And also, to prove the arbitration committee isn't totally useless, they've asked other people to go ahead and write some stuff about off-wiki harassment. How nice of them! PeterTheFourth (talk) 11:33, 19 September 2015 (UTC)
@Euryalus: -- Pshhh.I remember the GG case, where time after time it was shown users with the same Reddit names harassing, outing and threatening(jobs and more nefarious threats), but ArbCom would only state they were 'powerless' to connect the accounts and couldn't punish the GG trolls for off-Wiki harassment. Many of those accounts are still active. Ironic for ArbComs credibility here. So one thing ArbCom has been consistent on is that it will punish long term editors who react to the harassment, (except in cases where the claimed harassment are themselves GG supporters). It's good to see Salvio being consistent in their inconsistent decisions. Surprised about some others here. But I've come to accept that ArbCom is the place where otherwise good Wikipedians go to lose their moral compasses. Dave Dial (talk) 17:17, 19 September 2015 (UTC)
What's sad is the relentless on-wiki harassment that Masem brought to AE to get stopped and the resolution was to push him out and do nothing about the harassment he endured for nearly a year. --DHeyward (talk) 18:50, 19 September 2015 (UTC)
Historically offwiki actions are taken into account when the subject is already in the admin/arbom bad books. They tend to be ignored when performed by users with advanced permissions (admins, arb members, previous arb members etc). Its fairly biased when you look at previous cases where offwiki evidence is supplied. It also varies by arbcom, some take a harsher stance, some more lenient, but ultimately you get away with it depending on who you know. Only in death does duty end (talk) 07:09, 19 September 2015 (UTC)

I'm with DD2K here. It's laughable to see Tarc banned for this after the committee sat on their ass in the face of offsite harassment in the same topic area. Hats off. Protonk (talk) 03:26, 20 September 2015 (UTC)

I find it curious to see complaints about us not taking action when we cannot prove a link between accounts, and then in the same breath complaints about us taking action when we can. If off-wiki harassment of editors is taking place and we get to know about it (people do seem to sometimes forget we are not omnipresent) we will investigate it to the best of our ability (collectively we spent many tens of hours investigating the harassment in the Lightbreather case for example). If we can prove the harasser is a specific Wikipedia editor then we will take action against them, as we have done here and as we did in the Lightbreather case. If there is not sufficient evidence to prove that the harasser is a specific Wikipedia editor then there is nothing we can do. Thryduulf (talk) 10:24, 20 September 2015 (UTC)
"'There is nothing we can do. So, off-site" harassment" is only discouraged if the responsible party acknowledges authorship, and ARBCOM cannot or will not protect or assist Wikipedia volunteers if that harassment, however vile, is performed irresponsibly or has the cover of deniability -- if, for example, the perpetrator says "my dog did it!" or "it was another fellow using the same name!"
Does the eagerness of ARBCOM to see harassment against a powerful and well-connected white male editor, while (alas! So sad!) having been unable to protect gay victims and female victims who are haven't been their supporters and WikiFriends, disturb the members of the committee? Do you think a greater volume of irresponsible anonymous editing directed at extorting opponents' silence or aquiescence, will improve Wikipedia and the net? How will you handle the flood of professionally-directed harassment from political and business organizations, through deniable intermediaries, this policy encourages? Do you suggest that all of us who aren't WikiFriends with ARBCOM should lower our profiles? MarkBernstein (talk) 14:00, 20 September 2015 (UTC)
I find it curious that ArbCom would indef a long time editor for a tweet from their Twitter account, that stated the same information that can be found, and could be found for a long time, on (redacted). As if that is beyond all scope of decorum, even though not one word about it was typed on Wiki. Yet you have multitudes of 'new' accounts, anon IPs, and the joker who supports this type of behavior above my comment here, making real life threats of harm. Against families and jobs. If members of ArbCom can honestly believe they are being 'fair', then you have lost all measure of that word. A false dichotomy based on false equivalency is a logical fallacy. Not understanding that should disqualify any member from service. I guess Wikipedia can run CU checks for paid editing, but is 'powerless' against real life threats of harm against women and Feminists. Boys will be boys, ¯\_(ツ)_/¯ Dave Dial (talk) 15:04, 20 September 2015 (UTC)
It was brought to WP. The person being attacked asked for help. The result was the editor asking for help took a voluntary break while the editor that brought ridicule to him continues to edit. There is no doubt they are connected. More bans should be forthcoming. not less. --DHeyward (talk) 15:14, 20 September 2015 (UTC)
If this is the reason the Committee blocked - and if this is what happened they absolutely should have blocked - then why not simply say "outing"? Is there more to the issue than that? If it was simply outing, did Tarc get the same opportunity that Phil Sandifer got to delete the offending material? Gamaliel (talk) 15:19, 20 September 2015 (UTC)
@DD2K: Who is the joker above your comment that is making real life threats of harm? Brustopher (talk) 22:35, 20 September 2015 (UTC)
@Brustopher:I believe that DD2K's sentence should be parsed so that "making threats of harm" modifies "this type of behavior", not "the joker".
The editors post was removed, and I stated "supports this type of behavior', not that they engage in that type of behavior. In any case, since I've been threatened with a block like Tarc received for discussing this here, I guess I'll disengage. Members of ArbCom don't like having their decision making abilities questioned. Dave Dial (talk) 23:06, 20 September 2015 (UTC)
Please see Joe job e.g. /twitter.com/DaveDial2k NE Ent 00:14, 21 September 2015 (UTC)
In past cases handled by Arbcom joe jobs were absurdly unlikely. One offsite account existed for years before the wikipedia account in question started being active. In another case I can't find a single person outside of arbcom who discussed the issue offsite who thinks there's any chance of it being a joe job. Nonetheless I think taking this case in isolation, Arbcom did the right thing here.Brustopher (talk) 10:26, 21 September 2015 (UTC)
From what I know, that(Joe Job) doesn't describe what has transpired in this case. But then again, I don't know everything and ArbCom has been less than transparent here. In any case, why is it a good idea for you to make a fake Twitter account using my name? What point does that get across? Dave Dial (talk) 16:05, 21 September 2015 (UTC)
I believe he was demonstrating how easy it is to create an off-wiki account with an identical name to a Wikipedia. There is no connection between User:DD2K and /twitter.com/DaveDial2k, whence the bar for linking accounts is high because if it weren't all wikibattles would be held offsite in false flag accounts with much "Look what he said about me here!!" at ANI and ArbCom. Pretty much the proof is either an on-wiki "This is my account, real name, twitter handle, email" etc. Tarc has been deceptive before but comes clean when his deception has run its course/accomplished its goal (that's my experience, anyway). I suspect ArbCom asked him or had other very strong links to connect him and there is "beyond a reasonable doubt." I don't think his integrity would permit him to deny something that is true so that leaves either silence or confirmation. My reasoned guess is he proudly said it was him and that he did nothing wrong - which is his prerogative and is a view shared by others on WP. His course is charted on his principles which is admirable but not necessarily in line with the project. I think the burden the committee faced was more of "What do we do?" rather than "Is WP Tarc the same as Twitter Tarc?" --DHeyward (talk) 18:41, 21 September 2015 (UTC)
What was the nature of the harassment? Against whom? Of course there's a way to describe things without violating anybody's privacy. If arbcom thinks otherwise, then that needs explaining too. Thanks, - Wikidemon (talk) 18:15, 21 September 2015 (UTC)
A form of doxing/outing/IDing a specific wikipedia editor. The details beyond that are unnecessary (and I don't know the details anyway). Sensitive topics need not be repeated here and repeating them doesn't change the decision. --DHeyward (talk) 18:41, 21 September 2015 (UTC)
The Committee has blocked for outing before, but they've also clearly identified that as the reason they've blocked. Here, if outing was the reason, I'm baffled that they just won't come out and say that. I understand they want to protect the victim and not highlight the source of that information so everyone can look it up themselves, but you can identify it as outing without including those details or even the name of the target. Calling it harassment just opens up a can of worms into our laps. Gamaliel (talk) 20:04, 21 September 2015 (UTC)
I looked it up in about ten seconds on google, but there is a lack of clarity. This is about Gamergate, isn't it? It appears that somebody posted off-Wiki comments about off-wiki activities of a Wikipedian with whom Tarc has clashed, naming that person by their Wikipedia alias. These pastimes are not illegal or particularly scandalous, but they are personal, and of the sort that some would keep private. The posts in question were made by an account in Tarc's name that, apparently, was spoofing Tarc and not operated by him. Tarc, when asked about it, endorsed the truth of the posts, thereby indirectly committing a doxxing. Is that correct? I think arbcom could summarize this (or the actual version of events, in case I have it wrong) in about this level of detail without any upset. Secret deliberations and refusing to explain will probably have the opposite of the intended result, due to the Streisand effect. Now there is a big mystery that concerns people. It's not about changing the decision or searching for undue drama, it's about how much faith to place in arbcom and Wikipedia process. - Wikidemon (talk) 20:37, 21 September 2015 (UTC)
The behaviour of the Committee remains disturbing in the extreme. The Committee was elected, and has reaffirmed again and again that they make decisions based not on what is right, but on a political calculus of what most likely gets them re-elected. But this current committee makes nearly all of its decisions in camera, acts as judge, jury and executioner, and refuses to provide even the most basic rationale. ("Continued serious breaches of policy" Seriously?) There is a genuine fear of the Committee which operates without any form of oversight, transparency or accountability. Any of us could be blocked at any time for no reason at all. Hawkeye7 (talk) 20:57, 21 September 2015 (UTC)
We do not make a majority of our decision in camera. Since 1 Jan 2015 we have past 48 motions/decision. Of those, 7 involved staffing changes, 1 was a statement from the committee, 2 were desyops within policy, and 4 motions. That is 14 of the 48 or 29%. If you exclude the staffing changes, that ratio drops to only 15%. --In actu (Guerillero) | My Talk 18:04, 23 September 2015 (UTC)
I'm pretty sure that whatever ArbCom did was actually well-justified, and not something that that should result in excessive suspicion. But the absence of transparency predictably leads to suspicion. The last time I referred to User:Tryptofish/Draft B for ArbCom, a member of the Committee lambasted me for making a perennial proposal, but it seems to me to remain something that the Committee ought to engage with in some form. --Tryptofish (talk) 21:02, 21 September 2015 (UTC)
  • And the tiny ArbCom car pulls up, all the arbs scramble out, tripping over their own big shoes and each other's, arriving at an absurdly tangled result. Given the hands-off treatment of the reptilian editor whose culpability in the Lightbreather matter everybody but ArbCom and Wikimedia staff seems to be convinced of -- despite his obvious if clumsy efforts to out me, including the same sort of behavior that was condemned when Lightbreather did it, which ArbCom brushed off after no more than token examination -- it's hard to see this as anything more than a heresy trial and celebratory witch-burning. (If Arbcom is going to be the Congregation for the Doctrine of the WikiFaith, let's call it by its proper name.) Once again, ArbCom is reaffirming unspoken pillar 6: All animals-editors are equal, but some of you are less equal than others. The Big Bad Wolfowitz (aka Hullaballoo) (talk) 14:25, 25 September 2015 (UTC)
Given the committee’s apparent stance -- overriding the Phil Sandifer precedent, declining to indicate why outing is "harassment", really declining substantive explanation at all -- it may be time for the community to simply take stock and address the pressing questions this seemingly ill-advised and undoubtedly ill-explained motion leaves you. For example, what is harassment? In particular, do we have higher standards for the protection of powerful and influential Wikipedians, and lower standards when the target is a pain in the neck? Is the fact that the victim is a jerk justification? And is there a distinction to be made between harassment and satire -- for example, is calling Arbcom’s Gamergate decision “Infamous” harassment? Second, it is clear that editing Wikipedia with one's real name, or with a user name associated with accounts elsewhere, can incur lasting and permanent consequences. Children need to be protected from danger; should we not adopt a policy that minors may not edit using their real names, or account names used elsewhere? Finally, it is also clear that the committee will take no action against off-wiki harassment intended to extort an on-wiki result if the off-wiki harasser cannot be proven beyond a reasonable doubt to be directed by a Wikipedian -- a standard which cannot possibly be met and which is therefore effectively moot. Shall the community endorse an essay indicating that outside harassment to influence Wikipedia is neither supported nor opposed by Wikipedia? It is desirable that new editors understand what they are doing. MarkBernstein (talk) 14:48, 25 September 2015 (UTC)
MarkBernstein raises some interesting questions that deserve following up on. For instance, is Mark's retweeting of this harassment? Perhaps we ought to ask DHeyward? -Starke Hathaway (talk) 15:28, 25 September 2015 (UTC)
I'm assuming you believe the answer to this question to be yes? Wouldn't that put you in the same boat now given you have shared it too? Surely the appropriate thing to do in such a situation would be to email DHeyward privately instead of making a public spectacle of the issue? Brustopher (talk) 16:14, 25 September 2015 (UTC)
"it is also clear that the committee will take no action against off-wiki harassment intended to extort an on-wiki result if the off-wiki harasser cannot be proven beyond a reasonable doubt to be directed by a Wikipedian" - I don't see how this is a bad thing. Is the committee really being asked to sanction people where there is a reasonable doubt that the person is innocent? And where such sanctions might have real life consequences for the person sanctioned? Because that's what the situation was with the LB issue. I'm not going to vote to sanction someone for sexual harassment (especially if their real life identity is known) if there's a reasonable doubt, and I'm disappointed that anyone thinks we should. Doug Weller (talk) 16:23, 25 September 2015 (UTC)
I think you're missing something here. A lot of the people who think there ought to have been a sanction (dare I say most) don't think there was any reasonable doubt whatsoever regarding who was behind it. Brustopher (talk) 16:28, 25 September 2015 (UTC)
Your definition of 'reasonable doubt' and the 'man in the street's definition appear to be wildly different given the evidence available in the LB case. I have seen people arrested and charged for less in the UK (admittedly this is because they can then get a court order to make the web hosts and ISPs give up pesky things like IP logs etc as firm evidence). The case with LB is not that there was reasonable doubt, its that Arbcom appear to be applying a varying level of doubt depending on who the parties are, that often appears to be stricter than what would be required by law enforcement. Wikipedia is a private server, it is no one's *right* to be here, there are no freedoms being abused if someone is shown the door. See the WMF's SanFranBans(tm) with regards to certain problematic editors as an example. Only in death does duty end (talk) 16:34, 25 September 2015 (UTC)
@Bustropher and Only in death: Having seen all the evidence in the LB case, that presented by both sides and that revealed by our non-public investigations, and having examined all the links and assumptions therein, there was so much doubt in my mind that the person alleged to have been responsible was the perpetrator that I am not certain that even a standard of "on the balance of probabilities" would have resulted in a conviction (for want of a better term). What the UK police do or do not do is largely irrelevant as we do not have powers of subpoena, arrest, interview under caution, etc. All we can do is ask sites for logs, and if they say no there is nothing we can do about that. Do remember also that we did ban user:Two kinds of pork during the case as there was strong evidence, well beyond reasonable doubt, that they were responsible for at least some of the off-wiki harassment LB was subjected to - this is seemingly forgotten or overlooked by many of those who would like us to have an agenda. Tarc was banned because there was sufficiently strong evidence to tie them to off-wiki harassment of another editor. Thryduulf (talk) 17:51, 25 September 2015 (UTC)
@Thryduulf: See Capeo's remarks below. I suspect Arbcom either did not see all the evidence or willfully refused to look at some of it. Unless you are willing to engage in an open discussion of said evidence; what you saw, what was presented and rejected by yourself and the other arbs, your comments on this really have no credibility. Its not possible to trust someone who judges and condemns another based on secretly presented evidence. Only in death does duty end (talk) 19:28, 25 September 2015 (UTC)
Without knowing which evidence Capeo is referring to (they are rather vague and admittedly unsure of the details) it is impossible for me to say whether we saw that evidence or not. I am not a member of WO and so cannot search that site, but I do know that we did consider all the evidence that we were aware of, which would include anything sent to us by email. The reason that some evidence was presented in secret is because doing otherwise would itself be harassment (including doxxing in some cases) of Lightbreather and/or other living people - I for one will not engage (potentially) illegal activity just to settle someone's curiosity. Thryduulf (talk) 20:04, 25 September 2015 (UTC)
Did ArbCom take into account that the accused harasser had also targeted Lightbreather on Wikipedia [detail redacted to avoid any accidental outing]? Because there's no indication that you did. The Big Bad Wolfowitz (aka Hullaballoo) (talk) 21:08, 25 September 2015 (UTC)
Yes. When I say we took everything into account, I mean that in the literal sense: we took everything into account. There is no point trying to single out any particular thing because the answer will be "yes" to all of them. Thryduulf (talk) 22:23, 25 September 2015 (UTC)
@Thryduulf: , the only more specific I can be is that it was (I believe indeffed) user Vigilant that claimed he tried to get ArbCom to look at some supposed indisputable evidence to no avail. As LB herself was part of that thread I'd assume she would have at least forwarded this evidence herself anyway. I was more curious if there was a policy of simply dismissing evidence if the source was, well, questionable. Not just questionable by whom acquired it but by the means it was acquired. Many means of gathering info on people on the Internet I would think fall afoul of the WMF privacy policies ArbCom is bound by. Paradoxically ArbCom apparently demands close to that level of private info to connect off wiki accounts to on wiki accounts. Or maybe not? As tweets seem to be enough in some cases? I'm just confused because you can literally just stop by the cesspools of the GG related subreddits and see the planning of harrassment, doxing and general disruption by people making little to no effort to hide their on-wiki identities. I'm just having a hard time gauging the required level of proof to be able to act on it. Capeo (talk) 22:36, 25 September 2015 (UTC)
@Capeo: yes that evidence was sent to us, and yes it was considered. We do not have the time to read all the GG related subreddits. If evidence is presented to us of Wikipedia editors engaging in harassment of other Wikipedia editors we will obviously take a look to see if there is anything we can do, but obviously we cannot guarantee that that will be anything. We have no ability what so ever to police what goes on on reddit - you need to complain to the relevant people there if people are breaking that sites's terms of use (which I certainly hope planning harassment qualifies for, but I have never looked to see). Thryduulf (talk) 10:40, 26 September 2015 (UTC)

@Doug Weller: The consequence of this policy, however, appears to be that no degree of harassment or extortion off-wiki will be sanctionable on-wiki unless the perpetrator claims credit and demands to be punished. Open blackmail off-wiki -- “stop opposing us on the _____ page or we’ll get you fired and send an interesting dossier to your spouse” -- will in practice only be recognized by Arbcom if the perpetrator incriminates himself. Targets, of course, must turn the other cheek and accept whatever blows they are dealt, as any on-wiki incivility will of course be punished whatever the context. As @Only in death: reminds us, you are entitled regulate your private club site however you like, but I’m curious about how you envision this working in the coming years. The world now knows, for example, that an open campaign to secure the ban of five named opponents can easily succeed, and it knows that distributing nude photos of Wikipedians won’t be sanctioned if one says, “someone else did it.” With this motion we learn the importance of throwaway, deniable Twitter and reddit accounts when harassing Wikipedians, if harassment there was. It appears that this will become standard operating procedure in contentious areas; is that the Wikipedia you want? I’m puzzled.

My other perplexities -- whether the definition of harassment depends as heavily on one’s influence as it seems, whether precedents like Sandifer only apply to Wikipedians with the correct politics, whether satire and criticism can be read as harassment when directed at powerful and influential editors but not when directed at unpopular and troublesome editors, how admins are to recognize harassment -- all remain. MarkBernstein (talk) 17:10, 25 September 2015 (UTC)

Having worked on OTRS in the past I know that there are situations where maintaining privacy is crucial for a number of purposes. These include not only keeping private information secret, but also in avoiding exacerbating an already difficult situation. Did it occur to anybody here that the reason they are not talking about it is because there is a situation that would be made worse if they did?
I think we all need to accept that not everything that is done by arbcom is public, they have the discretion to decide what information should be public and what should be private. It has been said that a significant number of arbiters have seen the case if there was any funny business going on I doubt that many people with their own agendas would keep it secret, we have had adequate oversight. This discussion will serve little purpose other than creating rumours through supposition. HighInBC (was Chillum) 17:13, 25 September 2015 (UTC)
In principle, @HighInBC:’s comments are a useful reminder, but they are not pertinent here. I and others are not seeking more information about the events that led to this motion, but rather inquiring over whether the committee understands the consequences of the result and how we should proceed in this radically-transformed Wiki environment. MarkBernstein (talk) 17:23, 25 September 2015 (UTC)
It would help if you would get your facts right. In the LB case, nude photos of LB were not distributed - photos of various women, none of them LB so far as I know and certainly none of them said to be LB by her, were posted to a porn site and labelled "Lightbreather". Still bad, but not at all the same thing. And only those who have seen all the correspondence and details can legitimately state that there was no reasonable doubt. The "man in the street" is often convinced simply by an arrest. And I and others did not even consider anyone's "freedom" being abused. I'll also note that there was not just a lack of reasonable doubt and the concern that the real world consequences to an individual and family might have been serious if a sanction was placed, but there was actually reason to think that it was not the person being accused. Doug Weller (talk) 17:53, 25 September 2015 (UTC)
I'm curious about something. During the LB case a lot was going on over at WO. A certain member over there, extremely adept at doxxing, claimed he had indisputable evidence as to the harasser who posted the images. I didn't see it as I'm not a member but some of the far more level headed and less shrill members who saw the evidence came to the same conclusion. The doxxer has said that he contacted committee members and offered the evidence but that he either didn't hear back or was refused. I forget the specifics. If this is true is there any particular reason Arbcom wouldn't consider such evidence if offered?Capeo (talk) 18:15, 25 September 2015 (UTC)
@Doug Weller: I'm trying to be reasonably concise; "nude pictures" is a bit shorter than "nude pictures of women that were distributed to porn sites and that claimed to be pictures of LB, put were actually pictures of someone else." But OK. The problem remains: if a Wikipedian picks up a second computer, creates a fresh Twitter account and a fresh Reddit account, and uses them to extort the silence of Wikipedians, that's apparently not sanctionable. If a Wikipedian named FooBar uses their account FooBar1993 on some other service to harass Wikipedians, that’s only sanctionable if FooBar[1993] confesses. If a Wikipedian picks up a phone and says "Hey, bro, why dontcha post __________ about _________ for the lulz -- that'll shut the ___ up!", that’s not sanctionable unless both conspirators confess. The target still gets outed or fired if they don’t comply; of course, most targets will just lower their profile and leave Wikipedia and hope that will satisfy their tormentor. I'm genuinely puzzled: how do you see this working?
Similarly, I understand that my question about precedents and my question about equal protection may be awkward for you, but here once more we have Arbcom declining to make even a token gesture of concern for your volunteers, or to clarify what you want volunteers and admins to do. MarkBernstein (talk) 18:39, 25 September 2015 (UTC)
Mark, some of the people who have been involved (I am not necessarily saying they are WPedians) are in my opinion (and I think that of my colleagues also, and as I think should be obvious to anyone who has been following this) quite capable of pretending to be someone else for the sake of trying to get them in trouble. The inevitably resulting flap about whether or not they have been correctly identified would give rise to cries of injustice, to an extent that might far exceed the present. All arbs saw the evidence; some thought the person well enough identified; some not--how could we possibly be expected to proceed in something like this without consensus? To help you realize this, suppose, Mark, it had been you they were pretending to be. DGG ( talk ) 05:10, 26 September 2015 (UTC)

The mistake here was not the Arbcom 2015 decision announced here, but rather Arbcom 2013's failure to site-ban Tarc in response to Wikipedia:Arbitration/Requests/Case/Manning_naming_dispute#Disruptive_participation_by_Tarc; it was depressingly forseeable that failure to address Tarc's obvious unsuitability for editing Wikipedia would merely kick the can down the road. As to the current "discussion," I'll note:

  • It's not a reasonable criticism of the committee that some of the Internet's 4,700,000,000 pages were not read by them in resolving this case.
  • A site ban is not a tragedy; the primary value of Wikipedia is its an Encyclopedia, and readers can still read it regardless of their status as editors.
  • Reflecting on the contrast between Gamergate controversy with, say European migrant crisis should give one an idea of how totally unimportant the who said what about whom, who slept with whom, blah, blah, blah Gamergate nonsense really is. Obviously, to the extent real life stalking / harassment / threats has come out of it, it's important, but it's important for local law enforcement, not the committee; AN / ANI / AC et. al. are primarily designed for the resolution of content editorial disputes, we're not running a justice system here (see WP:NOJUSTICE).
  • The committee and community needs to be forward thinking -- accepting easily faked, unverifiable off-wiki evidence in this case -- regardless of how likely we think it to be true in the current instance -- lays the groundwork for its use a tactic in future conflicts.
  • Attempting to further justify a private evidence enforcement action on-wiki is inherently an impossible Between Scylla and Charybdis task, and attempts to do so will not satisfy perpetual wiki-critics but rather give them another toehold to respond to. (In case that's not clear I'm suggesting committee members cease further commentary.) NE Ent 12:33, 26 September 2015 (UTC)

Harassment and Concerted Action

@DGG: @Doug Weller: I fear you've been assuming my questions are rhetorical, and aimed at reversing Tarc or relitigating Lightbreather or questioning the evidence. They’re not.

Put yourself in your opponents’ shoes. Suppose that your job was to slant a Wikipedia page, or set of pages, for or against (say) a political candidate. You”re not one of the petulant children or moonbats who sometimes wind up at Arbcom: you're skilled and talented and you're a hard worker. You've got a team to help you. You’ve got a budget.

I submit that, in light of Gender Gap Task Force, Gamergate, Lightbreather, and Tarc that you now know precisely how to harass a Wikipedian in order to extort their silence, confident that Wikipedia will not stop you. You still, to be sure, need to worry about the FEC, about law enforcement and about civil lawsuits, but these are familiar perils in your line of work. You can assign part of your staff to harass opponents on Wikipedia while using the rest of the team to tag-team the target page and to wikignome for camouflage, while also building some sock or zombie accounts just in case they're needed. (Gold farmers have been selling account credentials for years, and leveling players for hire; what do you suppose the going rate is for a sparingly-used 5-year-old Wikipedia account?)

Politics is not beanbag. What is to prevent this team from riding roughshod over its opposition to create the appearance of a false consensus, to the great discredit of the project? And what is to prevent them from doing real harm to the real people who volunteered in good faith to edit the encyclopedia anyone can edit, not having been warned that attempting (for example) to follow reliable sources that this faction finds inconvenient could subject them to dangers unforeseen? Surely you have thought this through, but we’re not privy to your thinking. How is Wikipedia to avoid becoming a place that neither condones nor encourages extortionate behavior? MarkBernstein (talk) 17:08, 26 September 2015 (UTC)

You are perfectly right that any organization that permits anonymous editing will offer possibilities for the sorts of abuse you mention, and many other forms also. There are possible arguments for abandoning the principle of anonymity, but this view does not have the least possible chance of becoming adopted by the community, and the Foundation would undoubtedly consider it a violation of the basic principles. It's really that simple. DGG ( talk ) 17:22, 26 September 2015 (UTC)

Not sure how you mean in light of those cases. With Gamergate, the discretionary sanctions allowed administrators to come up with a solution that has dramatically slowed down the amount of disruption on those pages, and allows it to be quickly reacted to and shut down. It's not a solution I'd hope to see needed often, but in this case, I think it was the least bad alternative. In the Lightbreather case, one editor was banned for off-wiki harassment. I'm aware that another individual was widely accused of such, and all I can say is it wasn't conclusive enough. Wikipediocracy will, undoubtedly, continue to disagree.

I wish I had an easy solution to the issue in general. I wish, for that matter, I had a hard solution. If I did, I certainly wouldn't be keeping it a secret. ArbCom cases won't necessarily work against what you're talking about, especially if they're as good as you say and don't slip up (though leaving no trace of collusion gets harder to do with more people involved and higher activity). We don't have anything like subpoena or investigative power. Allowing robust investigations into editors' real life identity would entail substantial changes in our outing and harassment policies, and those changes could easily raise the specter of other types of harassment that could be as bad or worse. There is, of course, the risk of very negative publicity if such an attack team is caught in operation, and that may actually be more of a worry than anything technically illegal. The other possibility would be for the WMF to be more willing to take robust legal action against people doing things like that for violating the terms of service, but that, too, is not in our hands to decide. And even with that willingness, we'd still have to catch them first. So let me ask, how would you propose to catch a well-coordinated, careful team like that? Seraphimblade Talk to me 17:54, 26 September 2015 (UTC)

To assert that coordination will be sanctioned seems risible: coordination against the Gamergate victims, and against named Wikipedia editors, has been carried on in plain sight for over a year. Far from discouraging the coordination, Arbcom continues to reward the collusion by giving them pretty much everything they request.
There is no risk at all of negative publicity if such an attack team were caught: "a campaign spokesman said some overzealous volunteers got carried away" is the most you can hope for.
The answer, however, is not to sanction the colluders, but to remove or limit the rewards they can expect for harassing Wikipedians. If it were clear, for example, that the admins would allow extraordinary latitude to victims of harassment, that would limit the opportunity to use ARBCOM as a lever to get rid of opponents. The lesson from Lightbreather is (a) harassment will work, and (b) don't confess. If you made it clear that, in content disputes, the appearance of offsite extortion and harassment should weigh heavily against the side apparently benefiting from it, that too would reduce the temptation. Otherwise, extortion will become routine at Wikipedia. If Wikipedia cannot find a way to prevent widespread and routine harassment and extortion while retaining anonymous editing, society can and will take away anonymous editing.
Making an effort to deplore and then to stop extortion and harassment, rather than to encourage it, would be a really great place for ARBCOM to start. MarkBernstein (talk) 23:00, 26 September 2015 (UTC)
If Wikipedia cannot find a way to prevent widespread and routine harassment and extortion while retaining anonymous editing, society can and will take away anonymous editing. - you are pontificating using false hypotheses again. I've seen no evidence of widespread and routine harassment and extortion, and I'm one who has suffered harassment here. You've got to stick to the facts, MarkBernstein, not some dystopian vision that suits your sense of social justice etc. - Sitush (talk) 09:30, 27 September 2015 (UTC)
Good manners start at home. --DHeyward (talk) 23:14, 27 September 2015 (UTC)

Arbitration motion regarding Liancourt Rocks

Original announcement

Arbitration motion regarding Asgardian

Original announcement

Arbitration clerks seeking new volunteers

Original announcement
So...what's the pay? :p—cyberpowerChat:Online 14:59, 28 September 2015 (UTC)
With our new raises, I believe it is now on-par with the arbitrators'. :P L235 (t / c / ping in reply) 20:03, 28 September 2015 (UTC)
It certainly is. I've also heard they're planning to double the pay next year. Seraphimblade Talk to me 21:26, 28 September 2015 (UTC)
I've heard a rumor that Say on pay is about to be implemented. --Tryptofish (talk) 21:34, 28 September 2015 (UTC)
Well, you know it's not the pay, it's the benefits, especially the counseling for stress. Always comes in handy. Liz Read! Talk! 01:08, 29 September 2015 (UTC)

You load sixteen templates, and what do you get? Another day older and deeper in debt. St. Peter don't ya call, cause I can't go. I owe my soul to the Wikipedia store. NE Ent 23:23, 29 September 2015 (UTC)

NE Ent, me thinks that song is an age give-away. I never got the full impact of a company store until I worked for a week in Kohler, WI, and my stay was accommodated in a remarkably beautiful climate-controlled suite. If my memory serves, there was a pretty fancy golf course nearby. Utopia. Atsme📞📧 00:15, 30 September 2015 (UTC)

Promotion of Liz to full clerk

Original announcement
Wow, how did I miss this, it's a big surprise. Thanks Courcelles. Liz Read! Talk! 02:39, 30 September 2015 (UTC)
(edit conflict) Congratulations! Kharkiv07 (T) 02:41, 30 September 2015 (UTC)

Notice: Proposed change to WP:INVOLVED

Discussion is at Wikipedia talk:Administrators#Proposed change to WP:INVOLVED --Guy Macon (talk) 19:43, 2 October 2015 (UTC)

Arbitration temporary injunction for the Genetically modified organisms arbitration case

Original announcement
  • Good news, but can we please be clear on the included articles? Not that anyone would game the system but clarity would be helpful. Will my list of articles at the Workshop do for now? Guy (Help!) 11:02, 6 October 2015 (UTC)
  • @JzG: An uninvolved admin (i.e. not me) will have to do that. I added a talk notice of DS to the one article mentioned by name in the sanction; admins will have to add it to others at their discretion and probably create a standard 1RR template. (I've created the "gmo" DS topic code for that.) Thanks, L235 (t / c / ping in reply) 11:32, 6 October 2015 (UTC)
  • I want to say thank you to the Committee for doing this, because it was very much needed. --Tryptofish (talk) 17:05, 6 October 2015 (UTC)
  • Absolutely. I don't edit those articles so I'm not aware of this particular issue. I did cover a cloning break-thru for CNN's Environmental/Health Unit back in the 90s, and I also have a vague familiarity re: the cloning of horses but I depend on experts for the details (science). The role I've always played involves public dissemination; i.e. making the science understandable so the general public can understand it, and have always employed experts (on both sides of the debate) for input. The stations and networks I worked with (PBS, CNN, Discovery, Nat Geo, etc.) determined neutrality and whether there was any political bias for us to be concerned about. Atsme📞📧 18:15, 6 October 2015 (UTC)

New arbitration trainee clerks

Original announcement
Thanks for volunteering! All the best: Rich Farmbrough, 21:39, 15 October 2015 (UTC).

Gun control amendment

Original announcement

I for one am pleased with a constructive outcome. I hope it works out for all concerned. All the best: Rich Farmbrough, 21:33, 15 October 2015 (UTC).

Richard Arthur Norton (1958- ) October 2015

Original announcement
  • I've tried to make working section anchors for the section-targetted wikilinks for this section and the "original announcement" one but after 10 edits I've failed miserably, chewed off an entire finger, and will run away before anybody hits me with a chair for meddling where I shouldn't have. I'M SORRY. >_<  · Salvidrim! ·  17:34, 16 October 2015 (UTC)
Its fine I've spotted my own screw up, am going and correcting them. Amortias (T)(C) 20:41, 16 October 2015 (UTC)
  • I want to thank all the members of ArbCom for considering my request for clarification, which could easily have been shunted off into a corner and forgotten. I realize your jobs are very difficult, and I appreciate the time and energy you all put into doing them. (Better you than I.} My hope is that RAN will now concentrate on cleaning up his CCI backlog, ask for and get his topic ban removed, and then return to putting his considerable talents into writing new articles. I doubt he'll ever forgive me, but at least Wikipedia will benefit, and so, in the long run, will he. BMK (talk) 22:07, 16 October 2015 (UTC)
  • It's a crying shame that after many years of copyio free contributions we are not able to move forward, instead of backwards.
  • Expecting RAN to go through his past edits and remove perfectly good material to appease the purported desires of the community, is not useful. The vast majority of the edits (and I have looked a several hundred) are the addition of much needed reference with a small quote - small in comparison with the footnoted quotes of many of the more academic sources in the humanities that I come across.
  • Procedurally the finding fails to deal with the eventuality that the text-based CCI is completed without RAN's help.
All the best: Rich Farmbrough, 20:31, 17 October 2015 (UTC).
The point is that the consensus of the community is that RAN cannot be trusted, at the current time, to write draft articles without copyright violations and so all need to be checked in a way that other editors' work does not. There is also not a consensus that he has had "many years of copyvio free contributions" or even one year of them. The point of requiring RAN to put effort into clearing his CCI backlog is the principle of "you made the mess, now help clear it up". If anything on Wikipedia changes such that a person can no longer comply with an arbcom restriction then they may request a clarification or amendment and we'll remove or amend the restriction as appropriate to the circumstances. It is however interesting to compare this scenario with the "it is impossible to clear a CCI" rhetoric that is repeatedly heard whenever RAN's restriction is brought up. Thryduulf (talk) 10:21, 18 October 2015 (UTC)
So "the community" does not trust RAN on copyright, yet believes it's a good idea for him to police his own edits from a decade ago.
Lets look at one random edit:
Here a quote is added to an existing citation

"Ex-Diplomat Says Georgia Started War With Russia". New York Times. 25 November 2008. Retrieved 2009-07-10. A parliamentary hearing on the origins of the war between Georgia and Russia in August ended in a furor on Tuesday after a former Georgian diplomat testified that Georgian authorities were responsible for starting the conflict. {{cite news}}: Cite has empty unknown parameter: |coauthors= (help)

This is fully attributed, is no legal problem is not a policy problem, and is unexceptional in length (in fact it should be longer: the quote shows the part of the text that supports one of the claims in the article - it should be extended to support another one of the claims, at least). If any uninvolved person were checking this it would rank as "no problem detected".
However we must ask WWRAND (what would Richard Arthur Norton (1958- ) do?). He has no choice but to remove the quote, or risk be accused of "leaving a copyvio in" and possibly being banned for life or some such nonsense.
Quite possibly this is why he has found the game is not worth the candle, despite some attempts to work on the investigation.
(In this case the entire citation and quote were removed [18] together with the paragraph they supported, without explanation, by an anonymous IP.)
All the best: Rich Farmbrough, 18:50, 18 October 2015 (UTC).
Unless you can prove that RAN has ever been directly asked to remove a single-instance direct one-sentence quotation from a major independent reliable source from any of his articles, that argument doesn't hold water. Softlavender (talk) 19:26, 18 October 2015 (UTC)
The selection is at random: the same pair of arguments apply (by induction if you will) to longer quotes:
  • If we do not trust RAN and want his every edit checked, why on Earth are we asking him to do the checking?
  • If on the contrary we AGF we are placing him in an insupportable situation by asking him to make a call on his own edits. If he "errs on the side of caution" he will be deliberately making the encyclopaedia worse. If not he risks being challenged and dragged through the mire once more.
If we can come to an agreement that, for example, we are not requiring him to do anything about attributed text, that he can rely on Earwigs copyvio detector if the match is less than or equal to 5%, that unattributed PD text needs to be attributed, then we perhaps have a basis for asking him to work on the CCI. (I have not discussed this with him, so I have no idea of his position.)
All the best: Rich Farmbrough, 18:54, 19 October 2015 (UTC).
  • I agree with Rich here. IMO RAN has hurt his situation over and over again by pushing boundaries. And I can fully understand concerns he'll do so in the future. But A) people pick on some of his good edits and call them copyright violations when they are clearly fair use and B) his actual problems (things which are clearly an issue) are long in the past. WP:ROPE exists for a reason. It's sad that we don't give him some. There is a real chance that he'll screw up badly enough to get banned, but there is, IMO, a much better chance we'll have a lot of much better articles. Hobit (talk) 22:52, 19 October 2015 (UTC)

NOTICE: Volunteers are called upon to review RAN's userspace drafts and move them to the mainspace. They're listed on his user page or can be seen from this PrefixIndex query. Most of them don't seem to have any copyright issues. 103.6.159.89 (talk) 17:03, 21 October 2015 (UTC)

My comments should only be read as someone who has a very superficial knowledge of the issues involved. However, sometimes, outside perspective is helpful. A dissenting judge opined that there is something wrong with not declaring nuclear weapons illegal when the Dum-Dum bullet was declared illegal. Sometimes following the tangled reasoning of the letter of the law, one falls into absurdities which make no logical sense. Ok, back to the topic. I was involved in only one ANI thread about alleged copyvio. The allegations I found to be so far-fetched that I couldn't even believe them. If what RAN does is copyvio, I can barely make an edit on Wikipedia. Reading the clarification case, it seems that nobody else found any examples of copyvio either, which is utterly unsurprising to me. Rather the motion seems to be have been carried out for failure to disprove a negative - that RAN does not engage in copyvio. Kingsindian  17:29, 21 October 2015 (UTC)

  • There have certainly been actual and serious issues in the past. I felt at the time they were grossly overblown, but they were significant and block-worthy. But I'm unaware of any issues that people agree are still a problem--as you say some (actually all AFAIK) of the identified problems wouldn't be a problem for any other editor. 108.67.100.65 (talk) 01:23, 24 October 2015 (UTC)
Original announcement

Philippe appointed as Checkuser and Oversighter

Original announcement
A number of issues with the process of this. Policy states that Checkusers and Oversighters are appointed by Arbcom with community consultation (Quoted from CU and OS policies: "The permission is granted by the Arbitration Committee after community consultation and significant review of the user's contributions.")- which was entirely absent here. It was proposed by motion and closed within 2 days. I suspect with off-wiki discussion on mailing lists. If Arbcom had genuinely wanted community consultation as per the CS&OS policies, there should have been a notification at somewhere like AN, and as yet, the *result* of this motion has not been posted to AN (unlike other motions) indicating Arbcom did not, and does not want any community scrutiny on this. Second problem is that the WMF policy says those with access to private data (CS&OS) "requires that they have passed an RfA or similar process." Philippe passed RFA in 2007 when the process was both very different and much less scrutiny was applied to candidates. Technically he fulfills that criteria but RFA then is not RFA now, and I suspect that even if he passed now, it would be a long drawn out gruelling affair like Liz's RFA. Overall this gives the impression that Arbcom wish to circumvent/ignore the community consultation part of the process for CS/OS and pass out advanced permissions as their personal rewards to give - rather than something that should be earned. Only in death does duty end (talk) 12:44, 2 October 2015 (UTC)
Are you being serious, Only in death? Community comments were requested by ArbCom (on WT:CU and WT:OS), and were received properly... Wikipedia:Arbitration/Requests/Motions#Community comments (Philippe Beaudette to be given CU and OS tools).  · Salvidrim! ·  13:38, 2 October 2015 (UTC)
Well yes, the community would be those editors actively involved in wikipedia - for the most part long-standing editors and admins. If you think putting a notice on the OS and CU (as far as I can see it was on OS only) talkpages is a serious attempt to reach those people, you really dont know much about where the community hangs out (so to speak). Likewise as pointed out above, this was proposed, voted on, and closed in a small period of time - for a non-standard action this should have been given at least 7 days and been advertised in an appropriate place. At this speed, little actual consultation was had and the speed of the voting suggests that even had more comments been placed, it would have little effect as it would have been voted on and passed... It had the required 7 majority the same day it was posted. I will repeat the wording from the OS and CU policies for your benefit "The permission is granted by the Arbitration Committee after community consultation and significant review of the user's contributions." I have seen no sign of this taking place and thats a very loose definition of 'after' you are applying here. Only in death does duty end (talk) 13:53, 2 October 2015 (UTC)
Noting also that it's standard procedure that editors who have not had the tools removed for cause can request them back at any point. Given Philippe's years of experience (albeit on the other side of the fence), Philippe is clearly a good candidate for the tools. WormTT(talk) 13:45, 2 October 2015 (UTC)
With respect Worm, thats a ridiculous argument. "His fitness for the tools has been shown by his use of them while as an employee of the WMF" Really now? And how the hell is anyone supposed to verify that? I think thats something that could certainly be open to scrutiny. (I have not actually been able to find if he had access to the tools prior to his employment for the WMF - records of CU/OS are not as organised as RFA's - so correct me if I am wrong there) But its irrelevant to the above points - no community discussion or scrutiny was held, it is far from being a foregone conclusion he would pass an RFA in today's community process, and the overall impression is that Arbcom do not feel they have to follow their own policies as written. Only in death does duty end (talk) 13:58, 2 October 2015 (UTC)
Simply, I do not see the point in following process for process sake. The only people who can see tool use are CU/OS, that's the whole point of the tool. That we have an individual who is experienced in that tool use and can be vouched for by a number of functionaries is undeniable. This is not the same as scrutiny of an editor, this is more akin to someone who has previously had access to the tools, which is why I made that analogy. As for your argument about RfA, surely that's irrelevant - unless 3/4 of administrators (i.e. the those given the bit prior to 2007) should be desysopped? These rights are within Arbcom gift. They give it temporarily to scrutineers during Arbcom Elections with a simple motion, they return it to previous functionaries with a simple motion. If there is an individual who is clearly qualified for the tools, i.e. trusted by them and well experienced - why should they not? Our encyclopedia is built on IAR - "if a rule prevents you from improving the encyclopedia, ignore it". They didn't ignore it, no, but they also didn't bureaucratically go through the entire process. I'm comfortable with that. WormTT(talk) 14:12, 2 October 2015 (UTC)
Process exists in sensitive areas like tools because of the potential for abuse. 'Not following process for process sake' is an acceptable argument for not placing tags on obvious vandals etc, it is *not* an acceptable reason when you are granting a user the ability to have access to every editors private data. As for RFA - we have recently had admins (even one case of serious tool misuse) who were given the tools in earlier years not having moved with the times and be aware of current process. 'Those rights (cs&os) are an arbcom gift', yes this is exactly the attitude I pointed out above. You see them as a gift for you to hand out. If you cant see why that is problematic... Recently AUSC (made up of people with the tools, 50% appointed by Arbcom) came out with a finding that a staff member of WMUK who used the tools based on information received in his day job, and made pronouncements that were not supported by the facts, was not a 'misuse' of the tools. Forgive me if I reserve judgement on Arbcom's ability to judge who should have them. "You should just trust us we know whats best" has not, and never will be, encouraging to anyone of independant thought. Your argument drills down to 'We didnt need to consult the community on this'. Well thats not what the policy says. Only in death does duty end (talk) 14:25, 2 October 2015 (UTC)
By "These rights are within Arbcom gift" I mean they are for Arbcom's to hand out - not that they are given as presents. Poor language choice perhaps. WormTT(talk) 14:40, 2 October 2015 (UTC)
And yet you are still omitting the additional 'with community consultation' that is required by the policy. NEnt pointed out before it was closed that last time out-of-process additions were proposed there was a significant consultation period, which didnt happen here. it was half a day between filing and reaching majority. This had all the hallmarks of a pre-planned, pre-agreed motion. Even *had* the community responded quickly enough (I will point out of those very few who responded, a number are ex-arbs who no doubt have all the lesser-watched arb pages on their watchlists, and as I recall, as a courtesy keep access to tools and mailing lists - hardly a representative sample of the community!) I cant think of any situation recently where arbs have switched their votes based on post-vote discussion. I'm sorry but if the best you can come up with to explain why you didnt do it according to the policy is 'we didnt feel we had to'... Whats the point of having any processes at all? Only in death does duty end (talk) 14:49, 2 October 2015 (UTC)
Noting, I'm not on Arbcom, I didn't vote on this, I'm just another editor putting forward their opinion. As I see it, Philippe had this access then lost it due to a job change. I'm comfortable that he continue to have those rights and I don't have an issue with how it's happened. Pushing for more appears to be pushing for process for the sake of process. WormTT(talk) 14:53, 2 October 2015 (UTC)
Wikipedia policy exists to prevent abuse. Once Arbcom start ignoring it in favour of 'we dont think we have to' then they lack the moral/ethical justification for passing judgement on editors and admins who also ignore 'policies' they dont agree with. You wouldnt trust a policeman who doesnt think he needs to obey speeding limits when off-duty because he thinks he is a better driver and gets to break them in his day job. Only in death does duty end (talk) 14:59, 2 October 2015 (UTC)
Only in death does duty end, it's clear that you are unsatisfied with the process of this appointment. But the permissions were voted upon by ArbCom and granted, and according to Wikipedia:CheckUser#Assignment and revocation and Wikipedia:Oversight#Assignment and revocation, permissions can be revoked for inactivity and for cause. Unless there is a case that Philippe has misused the tools, there are no grounds at this point for taking away the permissions and there are no processes like a recall for Checkuser and Oversight.
This was an unusual situation--an admin who already had the permissions due to a job with WMF, who lost them due to a job change and then was regranted them by ArbCom--that is unlikely to occur again in the future. I hope you will participate in evaluating future candidates to receive these permissions as that process doesn't always draw a lot of participation. Liz Read! Talk! 19:31, 2 October 2015 (UTC)
Plus, this is Philippe we're talking about. I can count on the fingers of one foot the number of genuinely committed Wikipedians who would oppose his being granted CU and OS permissions. The correct response to those who cite process here begins with "f" and ends with "uck process". Guy (Help!) 22:36, 6 October 2015 (UTC)
I agree: for the sake of institutional integrity, it's best not to duck process. NE Ent 01:27, 7 October 2015 (UTC)
Actually, for the sake of building an encyclopedia, the thing we're all supposedly here for, it's best to edit articles and do other helpful things rather than spending all one's time commenting on that muck, process. BMK (talk) 02:52, 14 October 2015 (UTC)

Clearly the committee has met WMF Policy for the appointment. Equally clearly it has chosen to not follow its own policy in this case, and arguing otherwise is ridiculous. Although the motion announcement stated "Community comments are welcome on the motions page" the pace of voting indicated that a majority of members did not need community input to determine their vote on this particular issue. As the committee invitation was Discuss this at: Wikipedia talk:Arbitration Committee/Noticeboard/Archive 29#Philippe appointed as Checkuser and Oversighter, not click here to suck up to the committee, the criticisms here of Only in death for expressing an opinion are unwarranted and ridiculous. He did not say the committee should reverse his decision, CU / OS are not rights you just ask to get back if they've lapsed, Wikipedia was built on five pillars, including "consensus," not one, and so on. I agree the date of the Rfa should not preclude eligibility; that rule was imposed by WMF, not en-wiki. NE Ent 21:46, 2 October 2015 (UTC)

Although I support Philippe having these bits, User:Only in death is perfectly correct. The discussion was not advertised at all widely, the committee would have done well to have refrained from commenting and simply gauged the community reaction to their proposal. There seems to be a real communications issue here. A significant number, I would even say a significant proportion of established editors, including some who are strong believers in IAR think that in matters relating to functionary powers scrupulous, even punctilious, observation of process (which in some cases means due process) is essential if functionaries as a whole and individually are not to fall into disrepute.
All the best: Rich Farmbrough, 21:35, 13 October 2015 (UTC).
Personally, I would consider Phillipe's track-record as adequate justification for not allowing him access to rollback or reviewer rights.—Kww(talk) 22:41, 15 October 2015 (UTC)
How staff rights work
  • "Philippe has experience using both tools both on the English Wikipedia and others, including supporting the community in the WikiPR case and the recent Orangemoody incident" [19]
His logs do not show that he has ever had CU on en:WP - and that he has had oversight for only one minute. I already know that these rights logs are sometimes broken. Would be nice to have some information about how long he had these rights (he certainly didn't when I checked the rights special pages on 7 September - he had edit filter manager, afttest and afttest-hide).
All the best: Rich Farmbrough, 19:59, 19 October 2015 (UTC).
@Rich Farmbrough: +staff is a global user right and contains both checkuser and oversight on all projects. --Guerillero | Parlez Moi 13:26, 20 October 2015 (UTC)
Thanks, useful to know (though it doesn't explain other oddities I've seen unrelated to staff). I might put a fabricator request that global user rights show on local wikis. All the best: Rich Farmbrough, 16:06, 20 October 2015 (UTC).

Notification

I have no problem with the process or the outcome (granted - I didn't investigate and am just relying on what people said here : )

But after reading the above, I am wondering... For most other permission requests/granting, we have a notification system of varying types. RfA/B has a template, arbcom and the steward elections - a watchlist notice, etc. I don't think it's beyond the pale that CU and OS (permissions not auto granted, or granted merely by admins) should have some notification system too. If there is one, please point me to it. Else, even if it's just a template like Cent, or RfA, I think that that would be an improvement over the current situation. What do you think? - jc37 21:09, 25 October 2015 (UTC)

Here is the motion where granting permissions decision was discussed. This announcement on AC/N and AN was the notification. Liz Read! Talk! 16:31, 28 October 2015 (UTC)