Wikipedia talk:Arbitration Committee/CheckUser and Oversight

The project page associated with this talk page is an official policy on Wikipedia. Policies have wide acceptance among editors and are considered a standard for all users to follow. Please review policy editing recommendations before making any substantive change to this page. Always remember to keep cool when editing, and don't panic.

Step 6 currently reads:

Any unbanned editor who has made at least 150 mainspace edits by the first day of the calendar month before the election may vote.

Is "banned" being used in the context of Wikipedia:Banning policy, or is it referring to any blocked user? (nitpicking, I know, but the way it is currently written, I think that someone who is blocked but not formally banned could request to have a vote cast on his or her behalf) J.delanoy^gabs_adds 03:14, 3 February 2009 (UTC)[reply]

I'd say a blocked user could request to vote as long as they are not formally banned. I say that because it's happened in arb elections IIRC. Speaking for myself here, not for the whole committee necessarily. — Rlevse • Talk • 03:34, 3 February 2009 (UTC)[reply]

See here. >Radiant< 17:34, 3 February 2009 (UTC)[reply]

This is a delegation of extant Arbitration powers, which is part of why its in the spot that it is. There was also an RFC which had strong approval. Its essentially arbitration policy, and thus within the remit of the committee--Tznkai (talk) 00:52, 4 February 2009 (UTC)[reply]

I've removed the word count requirement for two reasons:

1. It's overly-bureaucratic and serves no one
2. It's being ignored, and policy pages should be descriptive, not proscriptive

If you disagree with the edit, please discuss here before reverting. Thanks! --MZMcBride (talk) 22:15, 3 February 2009 (UTC)[reply]

That's extremely low, considering bureaucrats need 90%. Majorly talk 04:35, 5 February 2009 (UTC)[reply]

That's not the best comparison, I think; arbitrators only need 50%, for example. Kirill ^[pf] 04:40, 5 February 2009 (UTC)[reply]

Jimmy Wales only picks the top through, as not everyone who has over 50% can become an Arbitrator. Techman224^Talk 19:50, 14 February 2009 (UTC)[reply]

I'm still of the opinion this policy is the wrong way round - preselection of 'preferred' candidates by Arbcom does little to assauge concerns of cronyism. Surely the community can be trusted to put forward and elect its own candidates, and then only in exceptional circumstances, Arbcom may veto the the result. This has worked fine for Arbcom elections after all - why should this process be any different? --Joopercoopers (talk) 19:46, 5 February 2009 (UTC)[reply]

"Any editor who has made at least 150 mainspace edits by 31 January 2009 may vote."

I assume this only applies to the current election. Shouldn't the wording be changed to clarify for future elections - for example, "Any editor who has made at least 150 mainspace edits by the day elections are announced."? Bsimmons666 (talk) 00:28, 6 February 2009 (UTC)[reply]

Don't worry, we'll have a chance to change it before the next election comes around. ;-) Kirill ^[pf] 02:39, 6 February 2009 (UTC)[reply]

No time like the present, right? Bsimmons666 (talk) 02:46, 7 February 2009 (UTC)[reply]

This policy is invalid. The ArbCom does not "hand down" policy. Policy grows from the ground up from the community. Andre (talk) 05:20, 6 February 2009 (UTC)[reply]

(This is effectively a mini-RFC for sitting Arbs and current sitting CU/OS users to answer)

This mini RFC is specifically for the sitting Arbs, but I would also like to see answers from any current CU/OS operators if they're willing to be bold and be counted. I'd like to see an answer from as many of you as possible, for the benefit of the community. As this will be the process and requirement for all new CU/OS operators, beside Arbs--who are community vetted via their own election for the duration of their term--can you tell us specifically the answers to these five questions? Any non-sitting Arb CU/OS can just add their own named section below.

1. Do you support having all current CU/OS operators who are not sitting AC members undergoing this election/trust vetting process, to see if the community today trusts them with the tools? This would be any CU/OS operator--including former ArbCom members--who received the rights at any point
   prior to this first election in February 2009
   .
2. So that there is no loophole for the current AC09 members, would you be willing to stand for this confirmation after your Arbcom tenure ends, and having this be a requirement for continued access to the tools for any future ex-Arbcom member?
3. Why or why not, in regards to your answer to questions #1 and #2?
4. Specific question I and I'm sure others would like answered in regards to questions #1 and #2, if your answer is not "Yes", and your #3 answer didn't cover it: Why are the old CU/OS different than all the new ones, to be exempted from a frank public examination of whether the community trusts them to do their jobs as CU/OS "today"?
5. When, if your answer to question #1 is "Yes"?

Thank you, and the community thanks you for your frank and forthright answers, to dispel any shadow of inappropriate politics in this process permanently, including retroactively in regards to the old users, so ALL CU/OS operators are held to matching and identical standards.

• Placeholder - I intend to answer these questions at some point. Please feel free to remind me on my talk page if I forget. For the record, I recently asked for the CU and OS bits to be flipped on on my account, but I only intend to use them for review purposes (i.e. to review evidence in arbitration cases and in cases where the use of OS and CU is questioned). I also intend to ask for the CU and OS bits to be removed at the end of my term. Carcharoth (talk) 07:48, 25 February 2009 (UTC)[reply]

Casliber

Cool Hand Luke

1. Yes. I imagine in the future we would want to require re-confirmation after terms. Not so urgent though.
2. Yes, I will stand for such a confirmation election (assuming I'm still here).
3. To #1, I would like such positions to have term limits so that the tools are held by people with fairly recent mandates of confidence from the community. To #2, it's mostly a personal decision. I was very specific about having a 3 year term limit as arbitrator. If the community would like me to serve in some other capacity beyond that date, they should give me a new mandate to do so.
4. N/A.
5. We would need to decide term limits for starters... No rush; perhaps within the next year. (I generally agree with Sam's answers on all of these questions).

Cool Hand Luke 01:45, 16 February 2009 (UTC)[reply]

Coren

1. No, at least not in this format. The review board proposal should take care of problems that may currently exist, but I would otherwise be inclined to consider the previous holders as "granfathered in".
2. Not for 2009, but probably for 2010. The currently sitting arbitrators were elected taking into account that they could elect to get those rights and retain them; that may not be so after next elections (and might, indeed, make the election process go a little easier on the candidates).
3. Because I'm no fan of changing the rules after the fact. Because the previous holders were expected to retain those right indefinitely, much more care has gone into their selection or election that would otherwise have. Retroactive changes in rules are always inherently unfair and should never be done unless there is a pressing need to do so.
4. They are not, but they were picked under different presumptions. What's needed is not reconfirmation so much as a means by which any errors or misuse can be handled, regardless of how the bits were gotten in the first place.
5. I would support reconfirmation of existing right holders on the recommendation of the review board once it has been set up.

Dominic As a current CheckUser and Oversighter, and former arbitrator, I broadly agree with your premise and think there should absolutely be periodic "trust vetting" of all people with privileged access, including former arbitrators with mailing list access as well as CU and OS.

What we at least need is for the internal assessment that went on before the candidates were put forth for the election to happen for all CU, OS, and former arbitrators on the mailing list. What initially happened was that all of the potential candidates were listed on the private arbwiki, and arbitrators left opinions; only candidates that were unanimous (or maybe had one or so weak objections at most) moved on to the election. Current holders of privileged access should be able to pass a similar standard of scrutiny from ArbCom, though not necessarily a private vote unless certain rationales include reference to sensitive material, and I think this would be an important undertaking, since there has really never been any review process for such editors. For one thing, several of the current holders would likely deservedly fail a vetting on the basis of their inactivity alone, and should not retain their access while they do not need it, for the sake of security and privacy. Public input, or evidence if there is a specific evidence, would be accepted, but without a community (re)vote. The same vetting could be undertaken by a "review board," but I am skeptical that it could be set up in a timely manner and without simply creating another layer of bureaucracy that requires its own vetting/elections/review.

However, I do not think an election is workable. In general, as with administrators, once the community has indicated its support, I think we should only put them through it again if they have to reapply after having it removed under controversial circumstances. Elections are a lot of time, effort, and potential stress for candidates and community members alike; and with dozens of people with privileged access not only will we be multiplying that problem, we will probably start to see diminishing interest from the community. I am already concerned with the relative amount of participants in the election as compared to the ArbCom elections, and we don't want it to turn into something like RfA, where only a small, predictable segment of the community participates. Dominic·t 12:07, 17 February 2009 (UTC)[reply]

FayssalF

1. Yes.
2. I don't think I'd be interested in keeping the tools after my ArbCom term comes to its end. A requirement for ex and future arbitrators? Why not?
3. This is an open project. We cannot call it 'open' if only a few people retain some very sensitive tools for life. And in order to maintain it open we can only rely on elections and reconfirmations.
4. N/A.
5. Probably after the constitution of wp:Review Board?

FloNight

Jayvdb

(quickly answering..)

1. No; or at least, not yet, and not in this format. The proposed WP:Review Board is one mechanism which I would prefer operate for a while before current OS/CU are subjected to community reconfirmation, which if done right now would be a popularity contest due to a lack of increased awareness of their invisible contributions to the project.
2. I will be happy to stand for an election like this to retain OS/CU after my term is over, or any other format decided upon by the committee at the time.
3. I believe that the committee should be free to try new measures as they see fit.
4. Standards evolve. Old OS/CU should be given an opportunity to adapt. Many of them are keen to see improved standards; some feel the higher standards mean they are unable to continue in the role because the additional transparency/accountability means the role is more demanding than they signed up for (e.g. time wise, or more philosophical reasons) but many are happy to adapt if more volunteers are on deck. I would not like to assume that the old members are unable or unwilling to rise to the challenge. If they dont wish to adapt, they will be looking down the barrel of either the review board or, failing that, a committee investigation (probably private out of respect to the changes afoot)
5. I would be happy to enforce old OS/CU undergoing the process in the same timeframe as I will likely undergo it, which is in two years from now.

Kirill Lokshin

I support the idea of reconfirmation in principle; as a practical matter, however, I don't believe that a straight reconfirmation vote would be tenable in the foreseeable future.

Past results (e.g. the past Arbitration Committee elections) suggest that a significant, if not overwhelming, proportion of incumbents—and, particularly, incumbents who are politically prominent within the community, whether because they have served on the Committee or for other reasons—will fail to be re-elected if any meaningful support requirement it set. A straight reconfirmation vote (with the number of candidates equal to the number of "open" seats) can only result in our having fewer CU/OS operators, never more; and we are faced with a very real possibility that we would lose enough to cripple normal operations.

A more palatable alternative would be to combine reconfirmations with normal elections, such that new candidates could potentially replace those who were not reconfirmed; but even here, care must be taken to avoid losing too many at once.

(It's worth noting, incidentally, that the current elections are not for fixed terms; once the users are given CU/OS tools, there is nothing in current policy that would mandate reconfirmation at any point in the future. The question may well be asked, in a year's time, whether they should stand for a confirmation vote; would this be any different from the current situation?)

As for myself, I've needed access to the tools primarily for my arbitration work, and I intend to relinquish them when I am no longer on the Committee. Kirill ^[pf] 05:00, 17 February 2009 (UTC)[reply]

Newyorkbrad

Risker

1. I think more value would come from the proposed review board/audit panel.
2. I have no problem with the concept of community reconfirmation, should I choose to actively continue using the permissions at that time.
3. I believe that, on granting access to Checkuser and Oversight permissions, community trust is one of several factors to take into consideration; others include interest, demonstrated ability to refuse inappropriate requests, understanding of policy, etc. When considering if permissions should continue to be extended, one has to consider whether they are being used, whether they are being used in a manner consistent with current policy, and the community's longstanding tradition of granting permissions and not revoking them unless there is uncorrected, actual abuse. At this point, there are no meaningful metrics to determine if these considerations are being met.
4. I think #3 covers most of it.
5. To some extent, I think we need to get the review board/audit panel fully functional before we determine timing of reconfirmation. Without useful assessment of how someone is using the tools, any reconfirmation process would be a popularity contest. Risker (talk) 17:24, 16 February 2009 (UTC)[reply]

Rlevse

Roger Davies

Sam Blacketer

I think the first thing to say is that the current election process now coming to an end was by way of an experiment. From my perspective it is agreed that those users with access to features such as Checkuser and Oversight should be approved by the community. One of the things that the arbitrators will have to decide after the end of it will be whether this process worked well as it was tried, or whether it could be improved.

1. Yes, eventually; but it is not urgent.
2. I have neither Oversight nor Checkuser and do not intend to take them in future.
3. Because I am conscious that monitoring the operation of Oversight and Checkuser are very difficult for ordinary users to do. By helping to reassure users that those with these powers are only those in whom the community has the highest confidence, it helps to stop frivolous complaints which distract from the writing of an encyclopaedia.
4. First, high-profile administrators can often become unpopular for reasons unconnected to their trustworthiness when using Oversight or Checkuser. Second, an election process would take up a great deal of time for the candidates. Some perfectly able operators would withdraw. Third, there already is a system of checking for abuse of Oversight and Checkuser; we know much about the trustworthiness of the current operators by the complaints we receive (or the lack of them).
5. I think it should be a rolling process over the next two years, taking several candidates at a time. Sam Blacketer (talk) 17:30, 15 February 2009 (UTC)[reply]

Stephen Bain

I don't think that this process is necessarily suitable for reconfirmation. There was an internal discussion in December (and again a bit in January) about possibly introducing term limits; it didn't produce anything concrete, however I liked the idea and I think that could be something adopted in the future. There would then be something similar to the elections for the Committee itself. Downsides would include:

• that the performance of CheckUsers and Oversighters is almost entirely insusceptible to community review, unlike the bulk of arbitration work;
• that the timing of new appointments to these roles is better arranged on a supply-and-demand basis rather than there being regular scheduled elections, which there would need to be if there were term limits.

Frankly, I myself was comfortable with the system trialled late last year (although, I'm right here in the cabal for the time being, so of course I'm comfortable with that) and I think it would be good for those advocating a transition to an elected system to think about these problems and how to address them. Can we learn from other projects in this regard?

As for my own access, I have it only for arbitration work, and I don't intend to keep it beyond leaving the Committee. --bainer (talk) 10:11, 17 February 2009 (UTC)[reply]

Vassyana

Wizardman

General comments/questions

• Election #3: "Candidates may post a short statement, and editors may pose questions of the candidates during this period."
• Election #4/5: "Candidates are encouraged to post a concise statement during the preliminary phase, outlining their credentials. Editors are encouraged to put brief questions to the candidates and to make brief comments."

FT2 ^{(Talk | email)} 01:00, 29 July 2009 (UTC)[reply]

This page might get a new policy category; the discussion is at

+closdmouthOUNDSde disabld!!![fair o/wmf????---Please note, I have [[Repetitive Strain Injury]] and find typing very hard. I use a form of shorthand, which may be difficult to understand. I can be contacted through MSN (sven70) or Skype (sven0921) if my meaning is unclear. (talk) 01:18, 2 June 2010 (UTC)[reply]

Recently, while commenting on a CU candidate,

I would also note that there are currently no instructions at the top of Wikipedia:Arbitration Committee/CheckUser and Oversight/2013 CUOS appointments/CU. One would think that anyone who follows a link would instead end up at Wikipedia:Arbitration Committee/CheckUser and Oversight/2013 CUOS appointments (which does have instructions and transcludes the CU subpage), but our new notifications feature sent me directly to the CU subpage when I was mentioned there. Editors who are mentioned by name on that page are especially likely to have had previous encounters with the candidate in question, and thus are, IMO, in special need of clear instructions. Could we have some commentator's instructions at the top of that page and future pages like it? In addition to a paraphrase of Kumioko's comment above, it would be desirable to instruct commenters that the comments should be relevant to the CU right and that this is not RFC/U. --Guy Macon (talk) 23:05, 14 August 2013 (UTC)[reply]

Because CU can track the IP addresses of an editor and vice-versa, the following questions arise each time this tool is utilized:

• How willing are you to block a user unilaterally (CU block)? With the advent of "Wi-Fi" and shared IPs would you consider whether or not the IP address was shared before making a decision?
• How much faith do you have in the "Duck Test" theory? With millions of registered editors, how would you to assume a match if you discover 2 different editors share topic interests?
• What is your stance on the tool "not being pixie dust" or "for fishing"? Many editors often question the tool's accuracy — what red flags do you look for?

I ask these questions because this is not only used to track vandalism but people who abuse the account creation process or the create excessive accounts, leading to accounts being block solely because "userA & userB logged in from the same IP or IP range". In light of the recent nominations, I think this is a valid discussion. Anon (talk) 01:25, 16 March 2015 (UTC)[reply]

Original discussion

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

---

Initiated by L235 at 05:52, 3 October 2015 (UTC)[reply]

Case or decision affected

Wikipedia:Arbitration Committee/CheckUser and Oversight

List of any users involved or directly affected, and confirmation that all are aware of the request:

• L235 (talk · contribs · deleted contribs · logs · filter log · block user · block log) (initiator)

Confirmation that all parties are aware of the request

Statement by L235

The banner at the top of

WP:A/P#Ratification_and_amendment. It appears to be closest to a procedure

(being amendable by the Committee, and being a subpage of "Wikipedia:Arbitration Committee" rather than "Wikipedia:Arbitration"), but it may not even be that because I don't see that any of it has been formally adopted by motion or resolution of the Committee. Could the Committee clarify the status here?

@Thryduulf: What I'm really asking here can be summed into:

• Can the Committee change it, notwithstanding the community's wishes? If so, then it is not a policy per Wikipedia:Arbitration/Policy#Policy_and_precedent; it is a Committee procedure (or maybe an information page), and should definitely not be called an "extension of the arbitration policy";
• If the Committee can't change it without community consensus, then the Any significant edit to this page should have approval from the Arbitration Committee part of the banner should be removed, and either link to Wikipedia:Arbitration/Policy#Ratification_and_amendment or just not be there, implying that the community may change it as they see fit;

• Also, if the Committee can't change it, then is it binding on the Committee to do all appointments this way?

I understand that

ping in reply) 19:10, 3 October 2015 (UTC)[reply

]

• I agree with
  ping in reply) 22:49, 3 October 2015 (UTC)[reply
  ]

• To expand: policies such as
  ping in reply) 22:52, 3 October 2015 (UTC)[reply
  ]

• @NE Ent:, why make this complicated and use a free-form template? There's already the standard:

{{Policy| subcategory = procedural | textoverride = <div style="padding-top: 0.5em; padding-bottom: 0.5em;">This page contains a procedure of the '''[[Wikipedia:Arbitration Committee|Arbitration Committee]]'''.{{resize|85%|2=Please do not change this page without the committee's authorisation.}}</div>|1 = WP:CUOS|2 = WP:AC/CUOS}}

Though it does look like the Committee is done with this request and will not action it, so no real use discussing further.

ping in reply) 02:40, 6 October 2015 (UTC)[reply

]

Answer from NE Ent

"Can the Committee change it," yes per own policies. It's generally understood any page title starting with Wikipedia::Arbitration falls under management of the committee, and editing there requires at least the passive approval of the committee. NE Ent 19:32, 3 October 2015 (UTC)[reply]

Per request, suggestion:

This page documents a procedure used to implement Wikimedia Foundation Policy. Any significant edit to this page should have approval from the Arbitration Committee. If in doubt, consider discussing changes here.

Note the existing page header isn't actually true -- although the en-wiki arb policy page does say the committee appoints CU, it's actually the WMF page which is binding. NE Ent 02:06, 6 October 2015 (UTC)[reply]

Statement by Hawkeye7

WP:Oversight is not an arbitration policy; it is a deletion policy. And Wikipedia:Arbitration Committee/CheckUser and Oversight is a procedural policy that implements it. Per Wikipedia:Arbitration/Policy, the Committee may create or modify its procedures, provided they are consistent with its scope. It is therefore the case that ArbCom can change it without community consultation. Hawkeye7 (talk) 20:50, 3 October 2015 (UTC)[reply]

Statement by Rich Farmbrough (CUOS)

It's agreed that this is not "policy". Therefore L235's suggesting that it not bear a label calling it "policy" - even if it is qualified in such a way as to make it clear (if someone looks it the meaning of "procedural policy" in the history of a discussion linked to from the talk page of a category) is hardly bureaucratic.

I'm sure the combined intellect of those perusing these pages can come up with better terminology. (It reminds me of soap-free soap...)

All the best: Rich Farmbrough, 13:06, 5 October 2015 (UTC).[reply]

Statement by {other-editor}

Other editors are free to make relevant comments on this request as necessary. Comments here should opine whether and how the Committee should clarify or amend the decision or provide additional information.

Status of WP:CUOS: Clerk notes

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

Status of WP:CUOS: Arbitrator views and discussion

• It clearly says it is a "procedural policy", which links to
  WP:NOTBURO firmly applies I think. Thryduulf (talk) 17:42, 3 October 2015 (UTC)[reply
  ]
  • @Rich Farmbrough: you (or anyone else) are free to make a specific request for a change to a different terminology and we'll consider it. Until that point it's not worth any more of the committee's limited time. Thryduulf (talk) 22:22, 5 October 2015 (UTC)[reply]
• Agree with Hawkeye and Thryduulf.
  [•] 23:20, 3 October 2015 (UTC)[reply
  ]
• As do I. Doug Weller (talk) 13:28, 4 October 2015 (UTC)[reply]
• the above three have already said all there can be on this one. Courcelles (talk) 05:47, 5 October 2015 (UTC)[reply]
• Indeed. This request can be archived at any time. Salvio ^{Let's talk about it!} 09:43, 5 October 2015 (UTC)[reply]
• Agreed. DGG ( talk ) 00:45, 6 October 2015 (UTC)[reply]

---

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