Wikipedia:Arbitration Committee/CheckUser and Oversight appointments

For the current procedure for appointing users to CheckUser and Oversight, see Wikipedia:Arbitration Committee/CheckUser and Oversight.

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

---

Please note the consultation period will end on 29 January 2009.

Closed, with thanks to all who participated: --ROGER DAVIES ^talk 10:37, 30 January 2009 (UTC)

Broadly, CheckUser and Oversight permissions are currently given to sitting arbitrators requesting them and also to a small number of editors recommended by ArbCom. Many current ArbCom members consider this to be ripe for reform. This proposal, therefore, aims to increase the number of editors with these permissions, to introduce a process for appointment; and to split the scrutiny process between ArbCom and the community.

This proposal has been developed following a lengthy discussion at ArbCom and therefore reflects current ArbCom consensus. It attracted 14 supporters (with no objections and no abstentions) with the following arbitrators supporting: Carcharoth, Casliber, Cool Hand Luke, Coren, FloNight, FT2, John Vandenberg, Kirill Lokshin, Risker, Rlevse, Roger Davies, Sam Blacketer, Vassyana and Wizardman.

The aim now is to get community feedback on this proposal. If the community prefers this process to the present method, and it works well in a trial, it is likely to be the basis for future appointments. You can provide feedback in two ways: you can participate in the straw poll below and/or you can leave comments for discussion on the talk page.

Proposal

Preamble

This describes the proposed method for determining which suitably qualified and trusted editors are recommended to the Wikimedia Foundation (WMF) for granting of CheckUser and Oversight permissions. This process is not set in stone and will inevitably change based on experience and evolving best practise. Suggestions for improvement are welcome and may be made on the talk page. Please note that CheckUser and Oversight permissions will be subject to periodic review.

Roles

1. ArbCom's role is to pre-vet candidates (this includes assessment of technical competence and familiarity with applicable policy) and to maintain a list of pre-vetted candidates to be put to the community from time to time.
2. The community's role is to vigorously scrutinise the pre-vetted candidates and to determine those most appropriate for CheckUser and/or Oversight.
3. For legal and policy reasons, the Wikimedia Foundation has final authority over access to Checkuser and Oversight permissions.

Candidacy and vetting

1. ArbCom will periodically invite applications from the community for CheckUser and Oversight permissions, although any editor may apply for pre-vetting at any time.
2. ArbCom will carefully vet all applications; very clear consensus among ArbCom members is needed for a candidate to be added to the list.
3. ArbCom will normally call an election (i) when there is consensus for further CheckUser and/or Oversight appointments and (ii) once there is a pool of about three pre-vetted candidates for each position.
4. In the event appointments are very urgently needed and there are insufficient immediately available candidates, the community will be consulted as to its wishes.

Election

1. Only pre-vetted candidates may stand.
2. The election is in two phases; a four-day preliminary phase during which nominations are posted, followed by a ten-day voting phase.
3. Candidates are encouraged to post a statement (up to 400 words) during the preliminary phase, outlining their credentials.
4. Editors are encouraged to put brief questions to the candidates and to make brief comments.
5. ArbCom clerks will monitor the election for decorum and proper process.
6. Any unbanned editor who has made at least 150 mainspace edits by the first day of the calendar month before the election may vote.
7. Voting is by simple approval voting (voters can support but cannot oppose). Editors may vote for as many candidates as they wish, and may modify their vote during the election. Brief explanations may be included.
8. Editors who are arbitrators, or who have been arbitrators in the past 12 months, are disqualified from voting.
9. Following the election, ArbCom will ask WMF to grant CheckUser and/or Oversight permission to the successful candidate(s) in order of votes cast. The final number of candidates recommended may be varied by ArbCom by majority vote, for example, in the event of a tie or near-tie or new vacancies arising during the election.

Straw poll

Please indicate your broad agreement or disagreement with the proposal. Discussion is best on the talk page not here.

Broadly agree

1. Tiptoety ^talk 18:25, 8 January 2009 (UTC)
2. I'm Mailer Diablo and I approve this message! - 18:28, 8 January 2009 (UTC)
3. Broadly agree, but not with Arbcom varying the number of appointees during or after the election. DuncanHill (talk) 18:38, 8 January 2009 (UTC)
4. Aye.--Tznkai (talk) 18:44, 8 January 2009 (UTC)
5. This has been needed for a long, long time. Majorly talk 18:45, 8 January 2009 (UTC)

   :-) FloNight♥♥♥ 18:49, 8 January 2009 (UTC)

6. Yup. I like the simple approval voting. Negative voting is quite poisonous and should be avoided. Jehochman ^Talk 19:00, 8 January 2009 (UTC)
7. Looks pretty good. --MZMcBride (talk) 19:13, 8 January 2009 (UTC)
8. Rjd0060 (talk) 19:14, 8 January 2009 (UTC)
9. Caulde 19:17, 8 January 2009 (UTC)
10. Yup, allows community input whilst still giving the Arbitration Committee oversight of the tools. Ryan Postlethwaite^{See the mess I've created or let's have banter} 19:19, 8 January 2009 (UTC)
11. Nicely thought out. Kingturtle (talk) 19:35, 8 January 2009 (UTC)
12. Logical, clueful: an excellent plan. RichardΩ612 Ɣ ɸ 19:50, 8 January 2009 (UTC)
13. More community input is (almost) always good. -- lucasbfr ^talk 20:01, 8 January 2009 (UTC)
14. Seems fine to me. Tim Vickers (talk) 20:10, 8 January 2009 (UTC)
15. I like it. Elbutler (talk) 20:43, 8 January 2009 (UTC)
16. I am broadly supportive. I do have a certain concern about the term "vacancy", which I shall spell out in more detail on the talk page. [[Sam Korn]] ^(smoddy) 20:54, 8 January 2009 (UTC)
17. Mostly agree, not sure about the approval voting part though, but so much to not support it. Mr.Z-man 21:04, 8 January 2009 (UTC)
18. Seems fine, so long as the pool of selected candidates is small enough relative to the pool of vetted candidates such that a candidate whom the Arbcomm finds appropriate but the community does not can be prevented from standing as a CU/OS without the use of negative votes. If we make it so that only one (or worse, 0) person "loses" an election, this has no real teeth from a community standpoint. Protonk (talk) 21:05, 8 January 2009 (UTC)
19. Yes. Crystal whacker (talk) 22:02, 8 January 2009 (UTC)
20. Was this developed with the review board in mind? I notice that the introduction refers to oversight (not Wikipedia:Oversight), but no mechanism is included. Avruch ^T 22:06, 8 January 2009 (UTC)

    I think the first use of the term in both the introduction and the preamble links to Wikipedia:Oversight. That is the intended reference, and this process can proceed independent of the review board proposal. Risker (talk) 22:29, 8 January 2009 (UTC)

    Knew I'd get into trouble using the same term - what I was referring to, and the term I should've just used, is "scrutiny" from the first para. It's possible, likely even, that the author meant scrutiny of the selection process rather than of the post-selection appointees. Avruch ^T 04:14, 9 January 2009 (UTC)

21. Very well done, especially using approval voting, to jab a stake in the heart of drama. rootology (C)(T) 22:17, 8 January 2009 (UTC)
22. Curious as to how this is going to work. Rgoodermote  22:48, 8 January 2009 (UTC)
23. Sounds sensible, provides more transparency and hopefully little drama.  Sandstein  23:01, 8 January 2009 (UTC)
24. I am not sure about the "arbcom decides when more are needed" part, I'm thinking that may not be ideal. There should have to be an election at least every X months (duno what X is but at least once a year?) The rest seems sound enough to broadly agree with. ++Lar: t/c 23:33, 8 January 2009 (UTC)

    Discussion moved to the talk page. Carcharoth (talk) 01:48, 9 January 2009 (UTC)

25. Sounds good. Cla68 (talk) 00:28, 9 January 2009 (UTC)
26. Excellent idea. More community input and less concentrated power will be great! Cheers, Jake Wartenberg^Talk 01:20, 9 January 2009 (UTC)
27. Agree, but only if we also have a de-CU/oversight policy....--Cometstyles 01:23, 9 January 2009 (UTC)
28. Rschen7754 (T C) 01:38, 9 January 2009 (UTC)
29. This sounds like a solid proposal. seicer | talk | contribs 02:11, 9 January 2009 (UTC)
30. I have some misgivings about simple approval voting in that if we have concerns about potential misuse of such high-power tools, we should always be able to voice them—even if the candidate is "pre-vetted". That being said, the great majority of the process is sound, and for most cases should be sufficient. {{Nihiltres|talk|log}} 03:08, 9 January 2009 (UTC)
31. I generally support this, echo the thought that there should be a balanced de-checkuser/oversight. Lazulilasher (talk) 03:57, 9 January 2009 (UTC)
32. I give my support for this, in assumption it may be altered later if problems arrise. I also give my support for community consultation of this kind, and welcome it as an improvement on the Arbitration Committee's standards of conduct. --Barberio (talk) 04:25, 9 January 2009 (UTC)
33. krimpet✽ 05:00, 9 January 2009 (UTC)
34. Another idea that's well past due - Alison ^❤ 07:04, 9 January 2009 (UTC)
35. I'm persuaded. Davewild (talk) 08:01, 9 January 2009 (UTC)
36. Good job on simplifying the process of a complicated matter. OhanaUnited^{Talk page} 15:21, 9 January 2009 (UTC)
37. Sounds good! — Twinzor ^{Say hi!} 16:53, 9 January 2009 (UTC)
38. Looks pretty good. VegaDark (talk) 17:36, 9 January 2009 (UTC)
39. Very strong support Willking1979 (talk) 02:29, 10 January 2009 (UTC)
40. I don't approve of the strong degree of ArbCom control over the process, but nevertheless I feel I must support this because it's clearly a big step forward. Everyking (talk) 05:41, 11 January 2009 (UTC)
41. Overdue, yes. About a year ago I supported a similar process. I'm happy this has become a larger issue now. Synergy 17:36, 11 January 2009 (UTC)
42. Yes. Seems good. J.delanoy^gabs_adds 20:55, 11 January 2009 (UTC)
43. Yep. Scarian^{Call me Pat!} 21:15, 11 January 2009 (UTC)
44. Broadly agree, with reservations about the specific proposal. Hopefully it will be tweaked. Enigma^msg 04:58, 12 January 2009 (UTC)
45. Per Everyking.--chaser (away) - talk 05:59, 12 January 2009 (UTC)
46. Strong Support - I think we need more trusted people in these positions. Chrislk02 ^{Chris Kreider} 19:40, 12 January 2009 (UTC)
47. Bbbbrrroooaaadddlllyyy (very broadly) agree. I'd be happy to see this implemented. —Cyclonenim (talk · contribs · email) 19:36, 13 January 2009 (UTC)
48. « Gonzo fan2007 (talk) @ 19:41, 13 January 2009 (UTC)
49. Strongly support this much-needed change. — Athaenara ✉ 03:50, 14 January 2009 (UTC)
50. Sounds reasonable. — xaosflux ^Talk 04:01, 14 January 2009 (UTC)
51. Broadly agree, as long as steps are taken to ensure that it does not degenerate into the tar-pit RfA has become. 78.86.97.203 (talk)

    Claiming this edit as mine (didn't notice I was logged out). Gazimoff 10:48, 14 January 2009 (UTC)

52. Support Zginder 2009-01-14T19:58Z (UTC)
53. Definitely a positive step forward. John Reaves 11:03, 16 January 2009 (UTC)
54. --Conti|✉ 15:58, 19 January 2009 (UTC)
55. Support - Yes, I am overall quite happy with this proposal as a fair mixture of ArbCom and community involvement. Camaron | Chris (talk) 20:30, 22 January 2009 (UTC)
56. My name is trusilver, and I approve this message. Trusilver 04:26, 28 January 2009 (UTC)
57. I would prefer to see a slightly more sophisticated system (ie STV than pure approval voting, but I agree with the reasoning behind using an approval system. Happy‑melon 08:43, 28 January 2009 (UTC)

Broadly disagree

1. Because of aims ... to split the scrutiny process between ArbCom; and the community., per previous comments at Wikipedia talk:Review Board. Though if the proposal simply concerns CU and OS access then I'd be fine (small things are like leaks on subs, as small hole being as potentially significant as a broad one). Deacon of Pndapetzim (Talk) 19:22, 8 January 2009 (UTC)

   Could I trouble please to expand this briefly on the talk page. It sounds interesting and not everyone will be familiar with the discussion you refer to. --ROGER DAVIES ^talk 19:37, 8 January 2009 (UTC)

2. This measure effectively silences the voice of those of us who believe that bureaucrats and other highly flagged users should not be checkusers and oversighters. Concern with concentration of power is not a fringe view, and there needs to be a way to express it. It is clear that ArbCom is not representative of the community on this score, as they award themselves many flags (and not just for purposes of overseeing their use), where as concern with concentration of powers has been expressed by others than myself in the community. --JayHenry (t) 05:25, 9 January 2009 (UTC)

   I find that an unusual opinion: one of the reasons that I broadly agree with the proposal is that it will make the process more open and public, which I think will also have the effect of spreading the rights among users who otherwise have little power, i.e. are not directly members of ArbCom. Do you think that this proposal would increase concentration of power rather than reduce it? If so, why? I'd just like to understand your reasoning better, I don't object to your point of view. {{Nihiltres|talk|log}} 15:44, 9 January 2009 (UTC)

   Responded on talk per the request above. Thanks for your questions. I really welcome discussion. --JayHenry (t) 01:24, 10 January 2009 (UTC)

3. If the committee is going to decided whether or not a user is trustworthy and capable of managing the tool then the only decision left is decided with what combination we can get the best bang for the buck, or with what selection of admins the checkuser backlog can be held at bay with the least number of checkusers (as it is generally held that checkuser and oversight should not be widely distributed ). The arbcom seems fully suited to that that task and an open election would seem to be more likely to go the super popular uber-admin wanabe’s than the users who would be most likely to put in the immense amounts of time required.
   A way for the community to object to appointments might be in order but the committee seems to be doing a good job in this respect already and elections are huge drama mills. Icewedge (talk) 07:12, 9 January 2009 (UTC)
4. It seems questionable to remove the right for users to object. Stifle (talk) 11:56, 9 January 2009 (UTC)
5. I have problems with this on a number of levels. One of the main ones is that I don't think we should open the doors to having large numbers of people with checkuser and oversight permissions. In recent memory, I haven't had an oversight or checkuser request take more than a brief amount of time to get a response so to me, that says there isn't a shortage. I think these positions need to have a good amount of oversight (no pun intended) to guard against abuse and the more people with the permission, the easier it is for something to slip through the cracks. I am far more encouraged by the review board than I am by this proposal. --B (talk) 18:43, 9 January 2009 (UTC)

   Discussion moved to talk page

6. For a different reason than some others--the entire procedures seems unnecessarily complicated. I would favor something as simple as give bureaucrats the power at their discretion, with the individual bcrat answerable if xe were to make a bad appointment. DGG (talk) 04:03, 10 January 2009 (UTC)

   I don't believe the global checkuser policy allows promotion of checkusers other than ArbCom appointment or community discussion. Mr.Z-man 04:13, 10 January 2009 (UTC)

7. The ArbCom's remit, by very definition, has been arbitration. Why is it now discussing why it should be able to play a role in what members of the community have 'oversight' roles? There is near nothing I can think of that would justify this. There are voting processes in place for most things, why isn't this suitable for one? Why the idea of "ArbCom maintaining vetted candidate lists", etc, etc.? This seems like an extension of power. Achromatic (talk) 19:56, 10 January 2009 (UTC)

   Actually, by WMF policy, it is Arbcom must appoint users to CU and OS rights. There is considerable leeway in how this can be carried out though. See m:Oversight#Access and m:CheckUser policy#Access. — Rlevse • Talk • 20:15, 10 January 2009 (UTC)

   Do you mean to say that that 'policy' was actually written, or adopted, by someone in some official capacity? The quality of writing is appalling. Was it adopted by a Board resolution or by some other process? 92.39.197.238 (talk) 19:06, 11 January 2009 (UTC)

   To be clear, it was the oversight 'policy' I was referring to. At first glance the Checkuser one isn't so badly writen. 92.39.197.238 (talk) 19:13, 11 January 2009 (UTC)

   Both those policies state that Arbcom need not be involved "where the community prefers independent elections". Yomangani^talk 18:56, 13 January 2009 (UTC)

8. per JayHenry. Giggy (talk) 06:21, 12 January 2009 (UTC)
9. Don't see much of a problem, and the solution is clunky and bureaucratic, while still not fixing the purported problems (arbcom appoints themselves and their buddies, etc). Would much prefer direct appointment as currently occurs, or some more lightweight process. I really don't see the need for all the hassle here, except to keep a clerk or two busy. — Werdna • talk 08:37, 14 January 2009 (UTC)
10. Tying CheckUser and Oversight privileges together is undesirable. Very few people should have Oversight privileges, probably only ArbCom members and one Wikipedia staff member responsible for urgent ORTS tickets. "Grandfathered" Oversight privileges should be revoked. There should be more people who can see the Oversight log than can exercise the Oversight privilege, to keep an eye on the use of that feature. Edits needing Oversight deletion are rare and should remain so. CheckUser is less of an issue; that could be given to many admins. --John Nagle (talk) 18:56, 15 January 2009 (UTC)

    Point of information - edits requiring oversighting are fairly common (most usually "junk" vandalism about non-users that includes phone numbers or personal details). Arbcom had to appoint a new Oversighter specifically last fall to help, and has lost two since. FT2 ^{(Talk | email)} 04:16, 17 January 2009 (UTC)

11. Disagree on principle. Any changes to fundamental issues of user privacy should be discussed by the community first (and then the community's procedure considered by others) - not determined by arbcom and presented as a fait accompli. Cynical (talk) 15:27, 17 January 2009 (UTC)

Disagree with approval voting

1. I agree with all here listed except simple approval voting. I would much rather prefer a simple system of Aye! and Nay! on the main voting page, but with discussion continuing elsewhere on a carefully monitored talk page. NuclearWarfare _(Talk) 20:50, 8 January 2009 (UTC)
2. I share the same concerns as NuclearWarfare. I'd like to keep the support and opposition on one page, as opposed to having just supports on one page and opposition on a discussion page. Nishkid64 _{(Make articles, not wikidrama)} 21:22, 8 January 2009 (UTC)
3. agree needs standard voting, oppose comments often bring light to issues that require more research. Canis Lupus 06:25, 9 January 2009 (UTC)
4. (a) Negative votes should have their weight (b) what is "very clear consensus" ? if not simple majority, then give a plain number. Say, 8 of 14 is clear but not very, and 9 of 14 is very clear. NVO (talk) 13:37, 9 January 2009 (UTC)
5. Doing this by approval vote is a really bad idea. You cannot generate consensus if you disallow one side of the issue from speaking. This is particularly true for community trust. For instance, check out the 2007 ArbCom election: there were fourteen users with >100 support votes, only six of which had consensual support to be elected to the committee. >Radiant< 21:26, 9 January 2009 (UTC)
6. Essentially the same thing Radiant! said. Oppose votes generate discussion - constructive criticism. Bsimmons666 (talk) ^Friend? 23:08, 9 January 2009 (UTC)
7. Approval voting is evil. More evil than straw polls. —Locke Cole • t • c 01:13, 10 January 2009 (UTC)
8. If there are six bad candidates, then with approval voting you still have six bad candidates, only now you know which ones are worse. A user desiring these permissions should be trusted enough to get a convincing supermajority of votes. -- King of ♥ ♦ ♣ ♠ 02:28, 13 January 2009 (UTC)
9. The point of this is presumably to find out when the community does not trust a candidate. If not, why hold votes at all? There's no place in this to object. (The problems with tactical voting is obvious; but there are more choices here than approve/oppose and approve). Septentrionalis PMAnderson 00:42, 14 January 2009 (UTC)
10. Agree with allowing the community to do so; disagree with a vote. Sceptre ^(talk) 16:52, 18 January 2009 (UTC)
11. In these cases the point of the election is to see if the candidates have the trust of the community and that requires the ability to oppose, and in particular to oppose all. If all the candidates pre-aproved by ArbCom generate 50% opposition than the correct result is pretty clearly to appoint none of them, but simple approval voting could obscure that and lead to the selection of the least distrusted candidate even if none have broad community trust. Eluchil404 (talk) 04:44, 19 January 2009 (UTC)
12. The above have said it well; particularly King of Hearts and Eluchil. seresin ( ¡? )  04:45, 21 January 2009 (UTC)

Polls are evil

1. Sorry, but I really object to the fact we are polling on this and not discussing. How many voters are here who have not read the discussion. (I've posted a long comment on the talk page, but I doubt many will read it. People will vote for preconceived reasons without really thinking. RfA all over again.)--Scott Mac (Doc) 20:25, 9 January 2009 (UTC)
2. As usual, this will only polarize the issue on "for or against" one particular version, instead of having people work towards a compromise. >Radiant< 21:26, 9 January 2009 (UTC)
3. Agree. Also, another beauty contest with no practical means of undoing the result seems like A Bad Idea. So it will obviously pass with flying colours. Sigh.Grace Note (talk) 07:05, 12 January 2009 (UTC)

   Of course abuse of oversight or checkuser rights is infinitely more serious than abuse of the admin tools. John Reaves 11:01, 16 January 2009 (UTC)

---

The above discussion is preserved as an archive. Please do not modify it. Subsequent comments should be made on the appropriate discussion page, such as the current discussion page. No further edits should be made to this discussion.