This is an old revision of this page, as edited by Carcharoth (talk | contribs) at 22:39, 31 December 2013 (→Motion regarding activity levels for holders of both CU and OS tools: closing discussion and marking as archived). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

• recent changes
• purge this page
• view or discuss this template

Currently, there are no requests for arbitration.

No cases have recently been closed (view all closed cases).

Currently, no requests for clarification or amendment are open.

Motions

Shortcuts WP:ARM WP:ARBM This page can be used by arbitrators to propose motions not related to any existing case or request. Motions are archived at Misplaced Pages:Arbitration/Index/Motions. Only arbitrators may propose or vote on motions on this page. You may visit WP:ARC or WP:ARCA for potential alternatives. Make a motion (Arbitrators only) You can make comments in the sections called "community discussion" or in some cases only in your own section. Arbitrators or clerks may summarily remove or refactor any comment.

Motion regarding activity levels for holders of both CU and OS tools

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

Motions ended deadlocked with equal support. Archiving for now and the new committee can return to this later if need be. Carcharoth (talk) 22:39, 31 December 2013 (UTC)

For this motion there are 11 active arbitrators, not counting 1 recused. With 0 arbitrators abstaining, 6 support or oppose votes are a majority.

The current section in the 'CheckUser/Oversight permissions and inactivity' section of the Arbitration Committee's Procedures document (adopted 30 March 2011, amended 11 March 2013) is modified as follows in relation to those who hold both CheckUser and Oversight permissions:

Either text A or text B, whichever gains the greater support, to be inserted before the section beginning 'Holders who do not comply with the activity and expectation requirements'.

Proposed motion (A)

• (A) Holders of both CheckUser and Oversight permissions are expected to maintain the required activity levels on both tools. If the activity levels on either tool fall below the required level, the holder may have that permission removed by the Arbitration Committee.

Support

1. AGK 17:04, 3 December 2013 (UTC)
2. There are two opposing points of view here. On the one hand, the functionary group have been significantly vetted and we trust them. If they remain active generally, then keeping the functionary permissions is sensible to allow them to use them in the rare occasions that they might need them. On the other hand, the functionary tools allow access to privileged information and if they're not being used the functionary, whilst trusted, may fall behind the curve in experience.
   When it comes down to it, functionary tools should only be given on an as needed basis, and if they are not needed they should be removed. The committee is fluffing around because we do still trust these users, but that's not the point. The point is that we're giving access to additional information to people who don't need it and that is simply bad practice. Worm(talk) 12:46, 9 December 2013 (UTC)
3. T. Canens (talk) 02:08, 10 December 2013 (UTC)
4. Having read the opinions below, both from functionaries and community members, I find myself on this side of the debate. The tools are granted for the purpose of using them. The existing process includes easy steps for regaining access: former CU/OS don't have to go through the application process, but they do need to assure Arbcom that they'll be active enough with the tools to meet these very minimal activity standards. I do think, as well, that the "may" clause is important, because sometimes functionaries are active in ways that don't include actual tool use (e.g., training of new users, creating manuals, orienting oversighters to OTRS, or maintaining Checkuser-wiki data); these should all count as "activity". Risker (talk) 03:06, 10 December 2013 (UTC) NOTE: First choice Risker (talk) 03:16, 11 December 2013 (UTC)
5. Second choice. I can see the point of view that this doesn't place unreasonable burdens, especially when once vetted the tools don't need to be sought through a formal process. Der Wohltemperierte Fuchs 14:25, 10 December 2013 (UTC)
6. Kirill  13:01, 15 December 2013 (UTC)

Oppose

1. Salvio 13:12, 10 December 2013 (UTC)
2. Opposing in favour of (B). I can see an argument for first and second choices here, but I think it is cleaner to make a direct choice. I wouldn't be adverse to at some point increasing the activity levels, but that would require a new discussion. Carcharoth (talk) 19:07, 10 December 2013 (UTC)
3. Mechanical counting of log actions is a problem to eradicate, not encourage. Courcelles 04:01, 11 December 2013 (UTC)
4. In favor of (B). Newyorkbrad (talk) 23:01, 14 December 2013 (UTC)

Abstain

1. With certain specific exceptions I traditionally abstain from voting on discussions regarding advanced permissions. This is because I am unclear on the finer details of some of the issues involved (legal, ethical, technical, etc). SilkTork 13:34, 30 November 2013 (UTC)

Comments

• I prefer motion (B), but I'm not sure whether I should cast a second-choice support vote here (on the theory of "better at least to have the issue settled than to leave it undecided") or oppose, so I'll cast only my first-choice vote for now. Made a minor copyedit (changed "any" to "either") for clarity. Newyorkbrad (talk) 16:00, 2 December 2013 (UTC)

Proposed motion (B)

• (B) Holders of both CheckUser and Oversight permissions are expected to maintain the required activity levels on at least one of the tools. If the activity levels on both tools fall below the required level, the holder may have those permissions removed by the Arbitration Committee.

Support

1. In the Committee's internal discussions, I have always been of the view that if a functionary with both CU and OS rights is actively using one of them, there is no point to withdrawing the other for inactivity; the person obviously is still fully engaged with the project, and at some point he or she may have a need for the other permission. Having said that, I might prefer an alternate formulation along the lines of: "Holders of both CheckUser and Oversight permissions are encouraged to maintain at least the minimum activity level on both tools. Should the activity level on both tools fall below the required level, after inquiry from the Arbitration Committee and reasonable notice, the holder may have the permissions withdrawn." Newyorkbrad (talk) 16:00, 2 December 2013 (UTC)

   Unless I've misunderstood, I think you are supporting a motion that says that (minus the "reasonable notice" clause)… AGK 17:08, 3 December 2013 (UTC)

   My alternative wording, which I don't feel strongly about, embodies our encouragement that our dual-use functionaries remain active with both tools, even though we are not making it a requirement for retaining both. By way of addendum on the substance, the comment from Agathoclea below embodies my thoughts. Newyorkbrad (talk) 01:32, 4 December 2013 (UTC)

2. Der Wohltemperierte Fuchs 03:27, 3 December 2013 (UTC)
3. Despite the comments below from some functionaries (though less commented than I had hoped for), I am persuaded that there is no pressing need to maintain activity levels separately on both tools. I would support a more nuanced phrasing such as Newyorkbrad suggests, but that may not be strictly necessary. I do think we need to delegate enforcement of this to a more efficient body, and/or be more efficient ourselves on this over the coming year. Carcharoth (talk) 02:30, 4 December 2013 (UTC)
4. Salvio 13:13, 10 December 2013 (UTC)
5. I'm infinitely more concerned about "is X active as an editor and around and pitching in with the work" than in counting log actions. I actually think our whole approach to functionary activity is in need of a rethink de novo, but this is the better option. Courcelles 03:04, 11 December 2013 (UTC)
6. Second choice. I think Courcelles raises an important point about how we measure activity, as there is much that happens that never shows up as a log entry. Risker (talk) 03:15, 11 December 2013 (UTC)

Oppose

1. AGK 17:04, 3 December 2013 (UTC)
2. Per my comments on the other motion. Worm(talk) 12:46, 9 December 2013 (UTC)
3. T. Canens (talk) 02:08, 10 December 2013 (UTC)
4. Kirill  13:01, 15 December 2013 (UTC)

Abstain

1. SilkTork 13:34, 30 November 2013 (UTC)

Discussion by arbitrators

This motion is proposed to clarify a potential ambiguity in the current wording regarding activity levels for holders of the CU and OS tools. For holders of both tools, some arbitrators are reluctant to remove one set of tools if someone is only maintaining activity with the other set of tools. Conversely, some interpret the current wording to mean that tools should be removed for inactivity regardless of whether someone is actively using the other set of tools. Since this does not require private discussion, I have proposed this motion here and will ask for input from the functionary team (i.e. those who use the tools), the rest of arbitration committee (i.e. those tasked with enforcing these activity requirements) and the wider community. I have no strong views either way, and I've not voted yet, as I first want to see what the consensus view is from the functionary team and others and have created discussion sections below. Discussion should be held open for at least a week to ensure everyone who needs to has seen this proposal and has had a chance to comment. Carcharoth (talk) 01:24, 30 November 2013 (UTC)

• Snowolf, one of the impediments in moving forward on the "inactive" issues is that those who were inactive generally held both tools and were only inactive for one. Historically, there has been a split in the committee in deciding whether meeting one activity requirement was sufficient if the user held both tools. Some argue that the key component is user trust, and if they're still trusted to hold one tool, then there should be no issue. Others argue that the activity levels are so minimal that anyone not meeting them has demonstrated that they no longer have an interest in using the tool. Getting this position straightened out and formalized, one way or another, is the purpose of this motion, so that these questions won't keep coming up. Fluffernutter's point about the difference between "may remove" and "will remove" is also noted. There are occasions when even functionaries take a "leave of absence"; provided Arbcom has been informed of the reason and the anticipated time of return, this is an exception to the rote removal of permissions. Risker (talk) 04:20, 30 November 2013 (UTC)

• I see no ambiguity in the procedures, and accordingly find these motions unnecessary. AGK 10:55, 30 November 2013 (UTC)

The current system where ArbCom (or rather arbitrators dealing with this) remind inactive tool holders, and then engage in a dialogue where needed, has not always been that successful this year (as has been pointed out below). Sometimes it has resulted in explanations with resumed activity later, other times the responses have been less satisfactory. The sticking point has been getting ArbCom as a whole to move on this with any degree of decisiveness, speed or efficiency. To streamline this it may be better to delegate the entire CU/OS review process to a smaller body (2-3 people) that will be able to work and act faster than ArbCom as a whole. This could be an ad-hoc group of current functionaries who would formally be asked by ArbCom to review activity levels (as opposed to some current functionaries reviewing such matters on their own initiative), or it could be delegated to an existing body such as the audit subcommittee.

It may be necessary to formally remove the need for ArbCom to vote as a whole on removing CU/OS tools for failing to meet the activity criteria, though that would require setting up a formal body constituted to vote on such matters. Appeal on decisions to remove tools would be to ArbCom as a whole. Trying to move past the frustration expressed below, would either of these options be viable? Carcharoth (talk) 05:29, 1 December 2013 (UTC)

Thanks Fluffernutter for the clarification. The workflows suggested by you and NuclearWarfare are pretty similar to how it has been handled. What I think has been missing is making clear to inactive functionaries that the next step will be a public (i.e. on-wiki) motion to remove their rights. This would remove the awkward private dialogue and would require arbitrators to vote on the motions rather than prevaricate (or more charitably, assign them a low priority and let them languish). Anyway, there are some current arbitration requests to review, so I'll go and do that now and return to this later in the week. Carcharoth (talk) 05:57, 1 December 2013 (UTC)

• As I said above, there's two different philosophies here. One, trust based, the other activity based. It's clear that the committee is divided on this matter - deadlocked in fact - and that's why we've struggled to effectively remove those users who do not meet activity requirements all year. The best thing to do here is probably put this motion on hold, and revote with the 2014 committee, in the hope that fresh views will prevail. Worm(talk) 08:46, 16 December 2013 (UTC)
• Thanks, Thatcher, for the comments. With voting completed some time ago, the result was a deadlock neither agreeing a new approach (Motion B), nor confirming the status quo (Motion A). What I will do now is mark the motions and discussions with archive tags and ask the clerks to archive them (or do so myself). This will allow the 2014 committee to start afresh when they are ready to look at this again, or at other parts of the procedures page. Carcharoth (talk) 22:28, 31 December 2013 (UTC)

Comments by functionaries

• It makes sense to me to handle each tool separately. It's entirely possible to be, say, a very active oversighter who never uses your CU permissions and hasn't kept up with CU/SPI policy, and find yourself in trouble when you suddenly try to use your CU tools. That said, however, in my perfect world Arbcom would sort out the issue of when "may be removed" is "will be removed" before they bother with "tool" vs "tools" - without some sort of idea for when the line will be drawn and when not, any policy about who lines apply to is rather useless as guidance. A fluffernutter is a sandwich! (talk) 02:14, 30 November 2013 (UTC)

  @Carcharoth: I wouldn't characterize my position as wishing the enforcement of activity policies would be more rigorous so much as I would characterize my position as wishing they would be documented and consistent. I don't see it as a problem for the workflow to go, say, "After one quarter of inactivity, communication from arbcom, then removal if a second quarter", or "Removal after one quarter of failure to notify arbcom of temporary inactivity" or "Removal after one quarter if no satisfactory explanation is provided to arbcom when they ask", or pretty much any other decision schema you can come up with, but I'd like to know which of these it is that arbcom will be applying, rather than the current "After one quarter of inactivity...who the hell knows. Maybe they'll remove it, probably they won't, perhaps my explanation will matter, perhaps it won't, I might lose my bits while someone else doesn't, but maybe if I have a good excuse they'll let me keep being inactive but only if the moon is full and I knock twice upside-down on a Tuesday." A fluffernutter is a sandwich! (talk) 05:38, 1 December 2013 (UTC)

• Futile endeavor, the arbitration committee has shown in the past that it has absolutely zero will or interest in forcing functionaries to be inactive. The committee itself, even when spoonfed the data and research, refused to act for months, and when it did so, it failed to act properly. If there was any will to do the job, it would have been done already, if there isn't, it will not be done no matter what you write. Snowolf 04:01, 30 November 2013 (UTC)
• I concur with Snowolf. Since the previous motion on inactivity was enacted, there have been several checkusers who have qualified as inactive. I have even done the 'hard work' (it took five minutes) of calculating who's inactive, and I informed the Arbitration Committee who those checkusers were. All these months later, these inactive checkusers retain their rights. If the Arbitration Committee could have taken the effort they've spent fiddling with the wording, and instead channeled it to enforcing the original policy, that would have been a far more productive use of their time. At this stage, I really don't care what the Arbitration Committee does, as long as they do something. --(ʞɿɐʇ) ɐuɐʞsǝp 13:33, 30 November 2013 (UTC)

• I have been trying for some weeks to get my colleagues to vote on motions about actual inactive functionaries. What I got instead was this pair of pointless meta-motions. Deskana and Snowolf, I share your frustration. AGK 15:05, 30 November 2013 (UTC)

  There is a bit more to it than that. Do we (ArbCom) need to go into more detail here? For the benefit of those not quite sure what is going on here, the inactive holders of CU and OS tools were contacted. Whether I can say more than that, I'm not sure (the reasons for inactivity vary a lot). If individual functionaries (some of whom have contacted us asking why inactive tool holders aren't being removed) are frustrated that others may not be pulling their weight and that the activity criteria are not being enforced, what should be done? Do we (ArbCom) have an obligation to explain to other functionaries what is going on and what the result was of the dialogue with the inactive functionaries? Or should it be a rote, semi-automated removal process like that done by the stewards (as pointed out by Rschen7754 below)? See also the discussions here back in March 2013. I had hoped that resolving the matter of how holders of both tools are assessed would help, and we could then start enforcing the inactivity criteria properly from that point onwards. Would that satisfy those functionaries who have been frustrated at how this has been handled this past year? Carcharoth (talk) 22:09, 30 November 2013 (UTC)

  There basically isn't, Carcharoth, but as you say I don't think we can go into detail here on specific cases. I think it would satisfy the functionaries' concerns (they have said as much), but my point was that we could have satisfied their concerns by just voting on the motions I've been harping on about these past few weeks. In other words, we could have done without these motions. AGK 17:08, 3 December 2013 (UTC)

• I don't think I have much to add beyond agreeing with all of the above comments. Beeblebrox (talk) 17:36, 30 November 2013 (UTC)
• Agreed. I think that a separation between the tools would make good sense, but the primary concern should be resolving whether the previously passed amendment on inactivity will be enforced at all. NativeForeigner 23:09, 30 November 2013 (UTC)
• Agree on separation between the two. Further suggestion: There should be four dates (e.g. Jan 31, April 31, July 31, October 31) that a single Arbitrator (maybe even one chosen at the start of the year) agrees to go and review activity of each functionary. If any functionary is failing to meet the criteria, they are contacted and given a chance to stick around if they can improve. The second such instance in a year automatically goes to a tools removal motion proposed by the reviewing Arbitrator. All we really need for this is regularity and prompt voting, and putting one Arbitrator in charge with clear guidelines on when to accept a functionary's excuses and when not to should be all that is necessary. NW (Talk) 05:36, 1 December 2013 (UTC)

• Carcharoth, unless things have changed in the last two months, what you're suggesting is more of a how things should be done rather than a how things have been done. I was the one handling a lot of it for the bulk of 2013 and as I recall it, the big bottlenecks were (a) a lack of clarity and agreement among Arbitrators about what we should do once functionaries gave us assurance that they would be more active in the future, (b) what we actually ended up communicating to the functionaries team as a whole about why we were and why we weren't taking action and (c) actually getting agreement to vote on the motions in a semi-timely fashion. Especially with (a), some sort of warning system would be useful, but moreover than that, it just really needs to be someone's responsibility to do it (I dunno, make it AUSC's job and give them something to do) rather than hope that one of the Arbitrators remembers (and simultaneously, that WP:AUSC/STATS is up to date). NW (Talk) 06:04, 1 December 2013 (UTC)

• (As a OSer and an Editor)I agree with Snowolf and Deskana, I can't see arbcom following through with whatever they decide to vote on. Too many meta motions that are useless hand waving and no action. (As an AUSC member) Punting this to the AUSC would be a decent idea since arbcom has a captive audience of labor that is not busy with functionary type tasks. --Guerillero | My Talk 18:51, 2 December 2013 (UTC)
• I think there's an implicit, more fundamental question that should be asked first: are we caring about the inactivity of the tool, or that of the tool user? That is, are we concerned that someone is holding a bit that they are not using, or a disused account having high privileges? The latter is an understandable security concern, the former (IMO) is of dubious value. — Coren  18:38, 5 December 2013 (UTC)
• Excuse my tardiness, I will echo the comments of Guerillero and Deskana. These motions are useless unless some group has the teeth and the will to take care of it. I do think that should be the Audit Subcommittee myself, so we can get someone to actually deal with it. -- DQ (ʞlɐʇ) 22:58, 14 December 2013 (UTC)
• The tools are distinct and require a distinct knowledge base in order to use them in an efficient and informed way. So motion B makes no sense. Motion A is sensible assuming you can actually quantify when someone's skills grow stale from inactivity. It is certainly true that CU and OS have evolved considerably over time -- all things change. How requests are made, evaluated for reasonableness and community standards, communicated, discussed, shared, recorded, acted on. If you aren't "plugged in" to the (sub)community of other functionaries, you can screw up without meaning to (and yes, this has happened more than once). But can you really define when a person's skills grow stale? Thatcher 21:02, 31 December 2013 (UTC)

General discussion

Why are functionaries separate from "others"? It looks faintly Orwellian, regardless of whatever the original intent was. Reaper Eternal (talk) 02:45, 30 November 2013 (UTC)

Mainly because I want to be able to review the responses separately when weighing up how to vote, as opposed to trawling through a long set of comments and trying to remember who are functionaries and who are not. What I'm looking for from those who use the OS and CU tools is direct impressions from their experience of using the tools. I've changed this section header from 'Discussion by others' to 'General discussion' so everyone can discuss down here as needed. Carcharoth (talk) 02:57, 30 November 2013 (UTC)

It's reasonable to have them as a separate section, because these motions directly affect them. If Arbcom makes a motion that affects a specific individual or group, it makes sense to have a section that is roughly the equivalent of "discussion by parties" as seen on case workshop pages. Risker (talk) 04:22, 30 November 2013 (UTC)

I don't see a point in this at all. As long as a functionary is otherwise active, why even remove the tools? Why limit the pool of people able to respond when there's a need for a response? It's not like the knowledge is going to go stale; changes to the CU and OS policies are both incredibly rare and rather difficult to miss. A functionary will lose access to CU and OS if they are desysoped for inactivity, which makes sense as a user can't be up to date if they're simply not around at all, buy why the fuss to strip CU and OS from active users? Sven Manguard Wha? 07:01, 30 November 2013 (UTC)

Why would they lose access to CU & OS if they lose adminship for whatever reason? Callanecc (talk • contribs • logs) 07:05, 30 November 2013 (UTC)

• If I'm correctly understanding the question that is being asked here, I think the best policy would be to consider each function (oversight and checkuser) separately for the purpose of determining who should maintain these rights; that is, if an editor has both rights, but frequently uses one while not using the other, then the second right should be removed. So these two rights should not be considered to be "bundled" together for the purpose of removal for disuse.

  The reasoning behind my opinion is based not on changes in policy -- which, as noted above, change quite slowly -- but on the normal practical conventions utilized in utilizing each right. Wikipolicy is, generally speaking, fairly loosely defined, deliberately so, in order that it can both lead and follow practice. Nevertheless, there are conventions of practice, and functionaries who attempt to make decisions in a realm they're not intimately familiar with, because of their limited use of the right, may be prone to make decisions based on their prior understandings of practice, as opposed to what is currently accepted. Beyond My Ken (talk) 08:15, 30 November 2013 (UTC)

• My reading of the issue is that an oversighter might come across edits that require checkuser and the other way around. So while focusing on just one of the tools they are able to fullfill their role better with the occational access to the other. So as long as they are trusted the overall activity should be relevant not that of the individual tool. Changes in policy will be communicated on the respective mailing lists and part of being trusted is the trust in the ability to keep up with things. -- Agathoclea (talk) 08:39, 30 November 2013 (UTC)
• I've fixed a minor grammatical error in both the motions. Hope this is OK. Graham87 09:56, 30 November 2013 (UTC)
• In recent months, stewards have removed the rights of CU/OS holders who have not used the rights for over 1 year: m:Steward_requests/Permissions/2013-09#Jniemenmaa.40fi.wikipedia, m:Steward_requests/Permissions/2013-08#Smooth_O.2C_Demicx.2C_Emir_Kotromani.C4.87.2C_EmirA_.26_Squirrel_.40bs.wikipedia. There has been some disagreement as to whether that is the proper interpretation of the m:CheckUser policy or m:Oversight policy, but the rights were removed by stewards regardless and have not been regranted. --Rschen7754 19:17, 30 November 2013 (UTC)
  • Also noting that stewards only have access to the OS logs, and not CU (though they can see the CU statistics). At the above links, the rights were only removed because people brought the rights to their attention. However, that doesn't mean that someone wouldn't bring it to their attention once the 1 year mark is hit, or that stewards wouldn't remove the rights out of their own initiative. @Carcharoth: I guess what I'm trying to say here is that leaving inactive rights that haven't been used in over 1 year could be problematic at the global level, as some stewards believe that the rights need to be used at least once a year. --Rschen7754 22:17, 30 November 2013 (UTC)
    • Thanks for the clarification on that. I think one set of inactive rights might be about to reach that one year level, so that is useful to know. Thanks also for the ping, but it didn't register in my notifications screen. I believe this is because you placed the ping as an addition to an existing post - if this notifications 'failure' is not documented anywhere, you may want to try and work out exactly how best to ping someone in such cases. Carcharoth (talk) 04:37, 1 December 2013 (UTC)
    • Interesting; Google translate doesn't really help me to understand the language well enough, but the actual global checkuser policy says that a user account that has been inactive for one year will lose CU rights; it doesn't say a user account that hasn't done checks for a year will lose them, and even other stewards questioned one of the removals. The local policy on Bosnian Misplaced Pages also specified an inactivity period for CU and adminship, after which removal could be automatic. I'll be honest though, it doesn't bother me that CU was removed on a project where there had been no actual checks done for a couple of years. Risker (talk) 17:50, 1 December 2013 (UTC)

• I think NW and Carcharoth have a good idea regarding handing tool inactivity to the AUSC remit. On motion personally I think they should be treated separately, given we have people with one but not the other those with both shouldn't be treated any differently. The sub-committee should be responsible for contacting the functionary who has been inactive in the preceding three months (per current definition) and letting them know that they will need to do 5 logged actions (including the one from the community) in the next three months (and their name is put on a list of people to check in two months time), after two months if the functionary in question still hasn't reached the requirements they are contacted again and reminded that they have a month. If after the third month the functionary still hasn't reached the activity requirement an arbitrator on the sub-committee posts a motion to WP:A/R/M for voting on removal of permission. There shouldn't be any reason for inactivity as functionaries are expected to "consider temporarily relinquishing their permission(s) for planned prolonged periods of inactivity" and their access may be restored also per current procedure - the standards there are up to the committee, however when it comes to this the access has already been removed. Just my 2 cents and opinion in any case. Callanecc (talk • contribs • logs) 13:24, 1 December 2013 (UTC)