Wikipedia talk:Arbitration Committee

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
(Redirected from Wikipedia talk:Arbitrators)
Wikipedia Arbitration
Open proceedings
Active sanctions
Arbitration Committee
Audit
Track related changes

Scope of ECR restrictions in I/P area[edit]

How do we determine if an article falls under ECR restrictions in the I/P area? I have Jonathan Freedland specifically in mind, given recent IP edits. Coretheapple (talk) 16:23, 6 April 2024 (UTC)[reply]

I'm interested in this question too (in addition to the question, what exactly is the "topic area", but that is another story). To me, those particular edits are unambiguously within scope of the ARBPIA restrictions. I routinely revert edits like this with an edit summary along the lines of "This is not an edit request. Editors must be extended-confirmed to edit or discuss this topic except for making edit requests. Edit requests most likely to succeed are those that are 'Specific, Uncontroversial, Necessary, Sensible' per WP:EDITXY." And I do this regardless of whether someone has placed a {{ArbCom Arab-Israeli enforcement|relatedcontent=yes}} template on the talk page. Then I add a template if it is absent. It would not make sense that the first non-extendedconfirmed person to add content within scope of ARBPIA is immune from the restrictions if the template is not there. Perhaps this is not the correct approach.
Another interesting question is "how often is the template not there" for things within the topic area, and the answer seems to be quite often. I have been looking at this out of curiosity and adding templates. Right now, I'm looking at protections and templates for everything in Category:Israel–Hamas war and all of its subcategories as a test set. Since it is an ongoing event, it is in relatively good shape, although the protections and templating are patchy. This graph gives you a general idea for everything in Category:Israel–Hamas war. Squares=categories, diamonds=articles, circles=talk pages, red=not EC protected, blue=EC protected, yellow=no ARBPIA template. Sean.hoyland (talk) 17:05, 6 April 2024 (UTC)[reply]
One area that I would like to see clarified is whether ordinary editors like moi can simply add the talk page template noted above. I imagine such templates can be subject to abuse if added inappropriately. Coretheapple (talk) 16:04, 7 April 2024 (UTC)[reply]
You can add, only uninvolved admins can remove apparently.
The current rules
I haven't looked at this but I'm guessing that there may be a significant mismatch between {{ArbCom Arab-Israeli enforcement}} and {{ArbCom Arab-Israeli editnotice}} deployment because there are differences in the user permissions requirements. To add an editnotice requires page mover permissions or something like that, I forget.
Sean.hoyland (talk) 02:46, 8 April 2024 (UTC)[reply]
Helpful. Thank you. Coretheapple (talk) 14:41, 8 April 2024 (UTC)[reply]
  • Another question (which I'm sure has been answered I guess) is contributions to RSN and user talk pages. I noticed a non-EC user weighing in at RSN but wanted to check first. A centralized FAQ might not be a bad idea. Coretheapple (talk) 16:17, 12 April 2024 (UTC)[reply]
    The only engagement with the topic that is allowed anywhere on en-wiki is edit requests on article talk pages. ScottishFinnishRadish (talk) 16:21, 12 April 2024 (UTC)[reply]
    OK, thanks for that clarity on this. Coretheapple (talk) 16:42, 12 April 2024 (UTC)[reply]

ARBCOM[edit]

Is ARBCOM the appropriate place to report mass canvassing that had been occurring on and off WP, and which the community hasn't been able to resolve? The idea is to describe a problem and request/propose amendments to current WP:ARBPIA guidelines. Makeandtoss (talk) 16:08, 11 April 2024 (UTC)[reply]

@Sean.hoyland: Makeandtoss (talk) 18:07, 11 April 2024 (UTC)[reply]
@Makeandtoss If this involves significant offwiki/private evidence, it might be a good idea to email us. Moneytrees🏝️(Talk) 18:09, 11 April 2024 (UTC)[reply]
@Moneytrees: @Firefly: Significant would be an understatement and it is not so private: [1], within hours of the posting of this Reddit thread, the concerned talk page discussion was suddenly flooded with opposing votes [2], including from two zombie accounts.
This is part of an ongoing trend since October 2023, in which at least half a dozen editors with pro-Israel viewpoints were topic banned or banned for mass canvassing through user emails, sockpupptery and/or meatpuppetry (and also part of a wider trend since 2011 that involved hundreds of users who were also banned for similar disruptive behavior). Notably one of the editors mentioned in the motion has contributed in the the talk page discussion without any meaningful argument or policy citing. There is a consistent, systematic and severe disruption of Wikipedia and its processes that is seriously damaging its reliability, and this is a pressing issue that needs to be addressed ASAP; because clearly the traditional solutions are being circumvented. Makeandtoss (talk) 18:46, 11 April 2024 (UTC)[reply]
Yes it's a lot of the same suspects and MO from the PIA Canvassing case.
I also have concerns regarding active sockpuppetry which I've been meaning to go to AE about.
I'm unfamiliar with how ArbCom and ArbCom cases work but perhaps the committee should consider a case reviewing the entire topic area regarding pro-Israel POV pushing, involving canvassing, sockpuppetry and excessive battleground/tendentious editing.
- IOHANNVSVERVS (talk) 18:55, 11 April 2024 (UTC)[reply]
I'll add that I've had some concerns myself regarding possible canvassing; at this requested move most notably. Not sure what Makeandtoss has in mind but my concerns are regarding canvassing along the lines of (or possibly a continuation of) the recent Arbcom case PIA Canvassing. IOHANNVSVERVS (talk) 18:28, 11 April 2024 (UTC)[reply]
Assuming there is private evidence involved, which I believe there would be given the description of "canvassing [...] occurring on and off WP", please send said evidence to arbcom-en@wikipedia.org (or use Special:EmailUser/Arbitration Committee if you prefer). firefly ( t · c ) 18:31, 11 April 2024 (UTC)[reply]
The only evidence I have for my concerns is public/on-wiki. I can briefly elaborate here on the concerning elements I've indentified if desired/appropriate. IOHANNVSVERVS (talk) 18:40, 11 April 2024 (UTC)[reply]
@IOHANNVSVERVS if the evidence you have is entirely on-wiki then ANI is the best place for it. firefly ( t · c ) 18:46, 11 April 2024 (UTC)[reply]

Yikes. Apparently, I'm not as observant as Makeandtoss. I hadn't noticed the apparent canvassing. I thought maybe Mila (Mbz1) had come back to visit judging from an uptick in aggressive editing by editors who have recently acquired EC privileges. Sean.hoyland (talk) 01:33, 12 April 2024 (UTC)[reply]

@Sean.hoyland: @Moneytrees: @Firefly: Update: the requested move has been stopped with the closing editor not taking into consideration the massive canvassing that took place just after the Reddit thread was opened [3]. The ease in which WP can be disrupted from accurately reflecting RS, and in the process wasting the time and efforts of good faith editors and rewarding editors who flout its internal processes and guidelines, is quite shocking and disappointing to me and I am sure to others as well. What is the way forward here in your opinion, should I request an amendment to the existing ARBPIA motion? Makeandtoss (talk) 21:46, 17 April 2024 (UTC)[reply]
If there is no off-wiki material, I would have thought AE was the way forward for the sort of "group" behavior being talked about here and it would need careful prep, a single RM would not be enough evidence imo, a pattern over time would have to be established. Selfstudier (talk) 22:06, 17 April 2024 (UTC)[reply]
As you can see here, there were 15 recorded instances of mass canvassing in just a few weeks [4]. And there are at least a dozen other instances were there is suspected mass canvassing taking place. Not to mention the two hundred accounts banned in the past decade for the same sockmaster [5]. The pattern is pretty much established and has been so for a very long time. Makeandtoss (talk) 08:49, 18 April 2024 (UTC)[reply]
Hmm, aren't we looking backwards though? I would have thought the idea would be to bring things up to date. Is off wiki canvassing the only issue? Selfstudier (talk) 10:15, 18 April 2024 (UTC)[reply]
Looking backwards to establish the pattern and to highlight how WP has been unable to deal with this issue and that new and extremely strict measures are required to prevent disruption. Both off wiki and on wiki canvassing has been taking place, and the current measures haven't been able to deal with either in a timely way. Makeandtoss (talk) 14:46, 18 April 2024 (UTC)[reply]
What sort of strict measures are you proposing? Outside of trying to identify and topic-ban or block specific editors who are clearly acting as WP:MEATPUPPETs via reddit posts, the 30/500 restriction is already as strict as any existing tool for this AFAIK. One thing I did notice is that there are more editors aggressively seeking to meet the 500-edit-count requirement (someone mentioned elsewhere that people who hit 500 edits are often posting in Wikipedia:Requests for permissions, not knowing that their 501st edit will grant it automatically), but if they're not overtly gaming it I'm unsure what can be done about that. --Aquillion (talk) 18:27, 18 April 2024 (UTC)[reply]
Two prominent editors do appear to have engaged in EC gaming. Not sure it I should mention them here but it's something I've been meaning to go to AE about. IOHANNVSVERVS (talk) 20:42, 18 April 2024 (UTC)[reply]
Perhaps ECR should not allow minor edits and/or include a requirement of a minimum number of different pages edited. That would prevent a lot of gaming I believe. Or perhaps ECR should not be granted automatically but one would have to apply for it after meeting the 30/500 requirements, which would allow the applying user's edits to be briefly reviewed for gaming. That would be a little bit inconvenient but would save a lot of time/effort overall if it prevents the disruptions caused by those who game ECR. IOHANNVSVERVS (talk) 21:02, 18 April 2024 (UTC)[reply]
What sort of pattern are we looking for? What would we use to identify editors who were likely canvassed? My thinking is that it'd probably be editors who contributed to multiple discussions that were posted in the same place, and whose contributions to them are surprising (generally because they either rushed to get 500 edits to participate in those precise discussions, or because they're "zombie" editors who returned for those precise discussions and haven't edited much recently otherwise.) Those are the clear-cut cases. Proving canvassing for established editors who are heavily active in the topic area is difficult because they could have just come across all those discussions naturally - if an editor contributes to every major I/P discussion, we can't conclude much. --Aquillion (talk) 18:27, 18 April 2024 (UTC)[reply]

<- What stands out to me in that RfC is not evidence of the effects of canvassing, it's how poorly the RfC process samples the editor population, making the result highly sensitive to random factors, bias etc. There is presumably very little that can be done about canvassing, but I assume a lot more can be done to encourage editors to participate and make RfCs in the topic area less susceptible to the side-effects of small population sizes. Sean.hoyland (talk) 13:26, 23 April 2024 (UTC)[reply]

Expanding the RFC discussion to dilute the external factors is an interesting idea but it risks overwhelming already complex discussions. I think other things can be explored to prevent such mass canvassing:
  • Setting a rule that sanctions editors who are later found out to have been canvassed through exposure of any on/off WP evidence (this would act as a strong deterrent regardless if such evidence is found).
  • Setting instructions to closing editors that compel them to take into greater consideration the canvassing effect by examining the circumstances surrounding each case.
  • Restrictions on WP email sending for users engaged in PIA articles or at least showing if editors participating in RFCs if they received emails in the past 1-2 weeks (or as long as RfC lasts) and from which users.
  • Preventing zombie accounts from participating, i.e. a user cannot participate in a RfC if they haven not been active for at least a week and then suddenly appear somewhere.
I realize these measures may sound strong-handed but this is the least that could be done to prevent the disruption of what is arguably one of the most important sources of information online. I am sure others have more ideas, that's why I opened this discussion, and to see how they can be implemented as soon as possible. Makeandtoss (talk) 14:12, 23 April 2024 (UTC)[reply]
I try not to get too involved in these discussions because I don't believe I should have much of a hand in crafting the policy that I enforce with almost no oversight, but while the zombie account thing looks good at first sniff, it could lead to more of the same involved people having the same discussions. I think it's pretty clear that non-arbpia editors just don't want to get involved so adding another fence for them to hop to get involved may backfire. ScottishFinnishRadish (talk) 17:01, 23 April 2024 (UTC)[reply]
Penalising *receiving* emails is not a good idea as it can be easily gamed by sending emails to users one doesn't agree with.
Penalising *sending* emails is pointless as usually throwaway accounts are used for this purpose. Alaexis¿question? 05:16, 24 April 2024 (UTC)[reply]
  • Makeandtoss makes some highly original suggestions. I suspect the first two might be hard to either establish or enforce—surely only devs can check the emails that have been sent, and both points seem to rely on a degree of OUTING, or private off-wiki evidence, so arbcom's new waterslide could find itself overwhelmed in the tide of contentious RfCs, no? But the third point... I like it. I like it a lot. It would, at the stroke of a hat and with almost minimal too or technical implementation required, not just make canvassed votes easier to see and dismiss but make canvassing itself pointless from the start. It wouldn't affect any established ethnocentric-politico POV pushers who are already within the article; but then, they are already there, and the current system is incapable of stopping them either. But this new approach (#3 above) would strip all the blatant, last-minute canvassed votes. And make it easier to address, ——Serial Number 54129 15:03, 23 April 2024 (UTC)[reply]
Regarding "Restrictions on WP email sending for users engaged in PIA articles", as far as I'm aware, the accounts used to distribute canvassing emails via WP mail in the ArbCom case were not active in PIA articles because they were not EC confirmed. Emails were distributed using disposable sockpuppet accounts by a ban evading user, perhaps not in all cases, but that seemed to be the MO. The recipients were presumably selected on the basis that they are or had been active in topic area at some point. Email restriction based remedies are not technically possible once a communication network has been established with no dependency of WP mail. Sean.hoyland (talk) 16:53, 23 April 2024 (UTC)[reply]
I'll copy this from Wikipedia talk:WikiProject Israel
Start copy
== Wikipedia Command Center ==
On October 2023, דוד שי (David Shay) announced a WhatsApp group to "fight" against anti-Israel bias in English Wikipedia. According to the announcement on Hebrew Wikipedia [6], the group seems to be coordinated by שלומית ליר (Shlomit Lir). It is also advertised on the Hebrew Wikipedia Facebook group, which is often moderated by Shani (WMF). The group publicly self-styles itself as "Wikipedia Command Center", and seems to serve as a private coordination hub for pro-Israel editing, focusing on a list of English Wikipedia articles. Since the group references WikiProject Israel, I bring it up here. How does this group comply with WP:CANVASS, WP:MEATPUPPET, WP:NPOV, etc? MarioGom (talk) 15:38, 23 April 2024 (UTC)[reply]
End copy. Selfstudier (talk) 18:39, 23 April 2024 (UTC)[reply]
I am not sure Wikipedia email is that relevant when there is a coordination group in WhatsApp with +100 members. MarioGom (talk) 22:09, 23 April 2024 (UTC)[reply]
  • This discussion is getting a bit far off from ArbCom (since we're no longer really talking about solutions ArbCom can / would apply) and might be better held either on Wikipedia talk:Canvassing or on WP:VPP. But one thing I will point out is that currently, WP:CANVASS does not even require that canvassed editors declare that they were canvassed. I feel that adding that minimal requirement (as a "hard" red-line requirement, albeit with a bit of fuzziness because editors arriving as a result of external discussions may not always understand that they were canvassed) would help a great deal in terms of both making canvassing more obvious and serving as an easier way to catch people who are engaged in illegitimate meatpuppetry. Obviously this would still require that we figure out that they're hiding their meatpuppetry somehow, but that brings me to... --Aquillion (talk) 04:03, 24 April 2024 (UTC)[reply]
  • One other thing that is worth at least brooching, although I don't know if it's technically feasible or feasible under our privacy policies: It would be immensely useful if CheckUsers could view the referer data for someone's visits to Wikipedia. Obviously that is highly private information and would have to be handled with extreme care, limited to CheckUsers alone; all they would do with it is give the same sort of "confirmed / no evidence / possible" response if someone is accused of being canvassed by examining to see if they have repeatedly visited Wikipedia via links from a site or page indicated as a source of canvassing during the window of a known canvassing incident. A secondary, less invasive tool for CheckUsers that might help would be one that just lists which inbound links have recently directed the most distinct users who have commented on a talkpage to that talkpage or its associated article page, which could be used to confirm canvassing incidents. Although it's possible to hide referers, an extensive effort to flood a page with meatpuppets is likely to result in a bunch of people clicking through to the page and commenting without realizing that they'd need to do that to avoid getting caught, which would produce a clear indicator that a checkuser could be asked to check and confirm (again, with just a simple yes or no) whenever external canvassing is suspected. --Aquillion (talk) 22:31, 23 April 2024 (UTC)[reply]
    As far as I am aware, checkusers do not have access to refer information (and WP:Checkuser makes no mention of it), so this is almost certainly something that cannot be implemented without action from developers (so don't hold your breath). I'm not even sure whether that data is stored in a way that retains its association with a specific user (Foundation:Wikimedia Privacy Policy doesn't make it clear whether they can say "Users X, Y and Z arrived from the Wikipedia Editors in Fooland Facebook group", "three unique users arrived via that link (but we can't say which three)", or just "three visits to this page came via that link, we can't say whether it was three different people or the same person three times."). If the user association isn't stored already then starting to do so would require more work by developers and (probably) sign-off from WMF legal that doing so is compatible with the privacy policy, etc (definitely do not hold your breath). Thryduulf (talk) 00:13, 24 April 2024 (UTC)[reply]

WP:NOT ENOUGH[edit]

If a certain case had gone ahead, I had proposed "being polite is not enough" as a principle for said case. I still think it should be a principle that can be applied in cases where it is relevant, and have expanded on the subject at the above link. It's in project space, expansion/tweaking is welcome. Just Step Sideways from this world ..... today 22:55, 11 April 2024 (UTC)[reply]

About contentious topics, especially ARBPIA[edit]

Just asking how to appeal a contentious topic, because the restrictions are going too far from "anyone can edit" principle of the pedia. There was once the thread I initiated on why AfDs are ECPed; it got an answer and was archived. I want to request amendment on reducing this restriction, because literally every article gets extended confirmed on sight, even templates, even categories and project pages (!!). This is particular to the Arab Israeli conflict, and even an April Fools Day deletion of Israel was deleted stating that it was part of the restriction. This should be encouraged to be stopped, I do not want to form a democracy, but as I said, restrictions are going beyond the limits. Toadette (Let's talk together!) 13:36, 14 April 2024 (UTC)[reply]

Are you active in the ARBPIA topic area? Are you familiar with the history and dynamics of the topic area, the level of deception/sockpuppetry, the culture of biased editing/POV pushing, the canvassing, the infiltration by staff from partisan organizations etc.?
The statement "literally every article gets extended confirmed on sight, even templates, even categories and project pages" is inaccurate.
The topic area is largely unprotected, or rather it is protected by people rather than the database server.
  • If you pick a category, any category, in the topic area and traverse the network of connections examining every subcategory, article, template etc. the graphs look something like this, with the majority of articles etc. unprotected (not blue). Squares=categories, diamonds=articles, circles=talk pages, red=not EC protected, blue=EC protected, yellow=no ARBPIA template. That's a graph starting at the category Israeli Settlement limited to just a few steps from the root category.
I would argue that protection in the topic area needs to be enhanced.
Opening the topic area up is the wrong way to go in my view. Sean.hoyland (talk) 15:29, 14 April 2024 (UTC)[reply]
And, non-extendedconfirmed users can make edit requests as the {{ArbCom Arab-Israeli enforcement}} talk page template explains. Straightforward, sensible requests are handled. Sean.hoyland (talk) 15:52, 14 April 2024 (UTC)[reply]
The place to request amendments to an Arbitration decision (including the rescinding of a CT designation) is Requests for (Arbitration) clarification and amendment. firefly ( t · c ) 15:49, 14 April 2024 (UTC)[reply]
The April fools Israel AfD was unfunny and in poor taste and was created by a globally banned sock, so I think the G5 deletion was proper and is a great example of why these restrictions exist. Moneytrees🏝️(Talk) 16:38, 14 April 2024 (UTC)[reply]
Without commenting on this specific example, an April fools (or any) joke being funny or not is irrelevant because humour is subjective. Whether something is or is not "in poor taste" is also something that is often debatable and not generalisable. A page created by a sockpuppet of a banned user being speedily deleted under G5 is not evidence that restrictions on editors who are not sockpuppets are required. Such restrictions may or may not be justified, but you have not presented any evidence that they are. Thryduulf (talk) 22:45, 14 April 2024 (UTC)[reply]
@Thryduulf, respectfully, I can't agree. A joke AfD on Israel for April Fools is unfortunate from at very least an optics perspective. One of the reasons CT exists is to keep socks out of controversial areas-- so a G5 for violating ECR seems like good application to me. Moneytrees🏝️(Talk) 01:46, 15 April 2024 (UTC)[reply]
If a user is a sock of a blocked user then G5 applies everywhere, regardless of whether the topic is covered by ECR restrictions (i.e. the page would have been speedily deleteable if it had nominated Luxembourg rather than Israel). That means that ECR is irrelevant to this situation.
If the user is not a sock of blocked user then G5 does not apply and is completely irrelevant to the situation. A proposal to expand to incorporate ECR violations into G5 did not result in consensus, a related discussion is ongoing but the option of extending G5 does not currently have a clear consensus (and has received explicit opposition).
Separately, April fools jokes (good and bad) are very poorly generalisable to anything that is not an April fools joke.
There may be good, relevant evidence for the restrictions but this specific argument is neither. Thryduulf (talk) 02:07, 15 April 2024 (UTC)[reply]
  • You only have to scroll up two sections to see an extensive discussion (with plenty of evidence) about ongoing canvassing, meatpuppetry, and sockpuppetry in the I/P topic area. The 500/30 restriction hasn't been perfect at keeping things functional (as the issues discussed above show), but it's the only really effective tool we have; the suggestion that we could weaken it now, at a moment when the relevant articles and discussions are faced with unprecedented levels of outside canvassing, isn't something we can seriously consider. If anything, ARBPIA needs stronger restrictions for the duration of the current war; the only issue is that we don't really have any other tools in our toolkit, but I certainly wouldn't be surprised if another ArbCom case were to come up with something. --Aquillion (talk) 19:54, 18 April 2024 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Arbitration_Committee&oldid=1220503601"