Wikipedia:Arbitration/Requests/Motions

Page semi-protected
From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Miniapolis (talk | contribs) at 16:25, 15 May 2016 (→‎Extendedconfirmed user group: Enacted motions). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Motions

Extended confirmed protection and arbitration enforcement

Statement

A1)

On April 5, the rollout of the new extendedconfirmed user group began. This group is being automatically applied to accounts meeting both of the following criteria: at least 500 edits, registered at least 30 days ago. A corresponding new protection level, currently called "extended confirmed protection", has been implemented that restricts editing to members of this user group.

Users
  • No action is required on the part of any current user. User accounts that meet the criteria will be automatically updated with the new user group on their next edit. User accounts that do not yet meet the criteria will be automatically updated with the new user group when they do qualify.
  • The extendedconfirmed user group can be added by administrators to accounts that do not yet meet the criteria. A process for requesting this has been set up here, intended primarily to handle the case of publicly identified legitimate alternative accounts of users whose primary accounts do meet the criteria.
Current uses
  • As of this announcement, this protection level is authorized for use in the following areas:
  • Any sanctions or other restrictions imposed prior to this motion passing shall remain in force unaffected.
Expectations
  • See proposals below (to be added as dot points).


Notes

Enacted - Miniapolis 16:01, 15 May 2016 (UTC)[reply]

Support
  1. Just a factual summary to accompany the proposals below. Callanecc (talkcontribslogs) 23:23, 11 April 2016 (UTC)[reply]
  2. Oops, I missed this subsubsubsection completely. Opabinia regalis (talk) 00:34, 12 April 2016 (UTC)[reply]
  3. As a statement of fact, I suppose. --kelapstick(bainuu) 20:58, 12 April 2016 (UTC)[reply]
  4. --In actu (Guerillero) | My Talk 14:33, 13 April 2016 (UTC)[reply]
  5. Tarapia tapioco (aka Salvio) (talk) 12:52, 15 April 2016 (UTC)[reply]
  6. Not entirely sure it's necessary, but no harm to it. GorillaWarfare (talk) 02:56, 18 April 2016 (UTC)[reply]
  7. per GW. Doug Weller talk 14:00, 22 April 2016 (UTC)[reply]
  8. procedural confirmation Cas Liber (talk · contribs) 10:47, 24 April 2016 (UTC)[reply]
  9. DGG ( talk ) 22:09, 1 May 2016 (UTC)[reply]
  10. Kirill Lokshin (talk) 23:50, 7 May 2016 (UTC)[reply]
Oppose
Abstain

Discussion by arbitrators

Community comments

  • " can be added by administrations to accounts that do not yet meet the criteria" - While "administrations" isn't particularly incorrect, I think the expected word is "administrators". --Izno (talk) 14:19, 11 April 2016 (UTC)[reply]
  • Please find my comments in the Comments by Ryk72 section below. - Ryk72 'c.s.n.s.' 14:54, 11 April 2016 (UTC)[reply]
  • Anita Sarkeesian was 30/500 protected by Ymblanter as a discretionary sanction (log) following the precendent of Brianna Wu. You may wish to amend the motion to account for this. BethNaught (talk) 11:03, 24 April 2016 (UTC)[reply]

Extended confirmed protection

Imposition

This provision cannot pass. Callanecc (talkcontribslogs) 09:43, 2 May 2016 (UTC)[reply]

B1) Extended confirmed protection may only be applied as a discretionary sanction where there is a consensus of uninvolved administrators supporting its use. This provision does not apply to a page or topic area which has been placed under 30/500 protection by the Arbitration Committee.

Support
  1. I would rather it be us only for such a nuclear option --In actu (Guerillero) | My Talk 14:41, 13 April 2016 (UTC)[reply]
  2. Until we have further experience with it, I would prefer that it should remain under arb com control. But as there does not seem to be sufficient support for that, at the very least it should require prior consensus of uninvolved admins. There are already too many problems with individual admins enacting sanctions, and this should be used only rarely. DGG ( talk ) 22:12, 1 May 2016 (UTC)[reply]
Oppose
  1. We don't need a consensus of admins to determine if there is disruption requiring 30/500 protection. --kelapstick(bainuu) 22:05, 11 April 2016 (UTC)[reply]
  2. Weakly - consensus of uninvolved admins can be difficult to come by in some particularly problematic topic areas. Opabinia regalis (talk) 23:19, 11 April 2016 (UTC)[reply]
  3. Weekly, primarily because full protection is already permitted as a discretionary sanction and that admin participation at AE is generally small. Callanecc (talkcontribslogs) 23:34, 11 April 2016 (UTC)[reply]
  4. A "consensus of uninvolved administrators" creates a special class of AE action. Courcelles (talk) 16:59, 13 April 2016 (UTC)[reply]
  5. Tarapia tapioco (aka Salvio) (talk) 12:54, 15 April 2016 (UTC)[reply]
  6. See my comments below. GorillaWarfare (talk) 03:08, 18 April 2016 (UTC)[reply]
  7. Should not be required. As has been pointed out, it's not required for other forms of protection. Doug Weller talk 14:04, 22 April 2016 (UTC)[reply]
Abstain
  1. Drmies (talk) 17:49, 22 April 2016 (UTC)[reply]
  2. Tricky. While I don't readily see an occasion where a group of editors might see some use of this category as the best solution to a problem, I am not sure I am happy to see it barred either. Cas Liber (talk · contribs) 10:54, 24 April 2016 (UTC)[reply]

B2) Extended confirmed protection may be applied as a discretionary sanction by any uninvolved administrator at their discretion. This provision does not apply to a page or topic area which has been placed under 30/500 protection by the Arbitration Committee.

Support
  1. kelapstick(bainuu) 22:05, 11 April 2016 (UTC)[reply]
  2. Weakly - I would be much more comfortable supporting this conditional on D1 (extendedconfirmed cannot be revoked under DS). Opabinia regalis (talk) 23:19, 11 April 2016 (UTC)[reply]
  3. Callanecc (talkcontribslogs) 23:34, 11 April 2016 (UTC)[reply]
  4. Doug Weller talk 14:04, 22 April 2016 (UTC)[reply]
Oppose
  1. per above --In actu (Guerillero) | My Talk 14:41, 13 April 2016 (UTC)[reply]
  2. See my comments below. GorillaWarfare (talk) 03:08, 18 April 2016 (UTC)[reply]
  3. At the very least it should require prior consensus among uninvolved admins. it is unlikely to ever need emergency action. DGG ( talk ) 22:12, 1 May 2016 (UTC)[reply]
  4. Salvio Let's talk about it! 15:11, 4 May 2016 (UTC)[reply]
  5. Kirill Lokshin (talk) 23:51, 7 May 2016 (UTC)[reply]
  6. Keilana (talk) 17:53, 9 May 2016 (UTC)[reply]
Abstain
  1. Drmies (talk) 17:49, 22 April 2016 (UTC)[reply]
  2. As above. I think too novel to be applied by a single admin. However, discussion might make it useful Cas Liber (talk · contribs) 10:54, 24 April 2016 (UTC)[reply]

Requirements for use

C1) Extended confirmed protection may only be applied in response to persistent sockpuppetry or continued use of new, disruptive accounts where other methods (such as semi protection) have not controlled the disruption. This provision does not apply to a page or topic area which has been placed under 30/500 protection by the Arbitration Committee.

Enacted - Miniapolis 16:05, 15 May 2016 (UTC)[reply]

Support
  1. I think it's important to hold the line on this. I would like this protection level to be used - at least at first - specifically where there is persistent socking, SPAs, meatpuppets, etc., and not merely for "disruptive editing" that happens to be coming from relatively inexperienced users. Opabinia regalis (talk) 23:19, 11 April 2016 (UTC)[reply]
  2. Weekly my first preference, I'd like to see 30/500 used on pages where there is consistent socking and use of throwaway accounts rather than just new accounts occasionally breaking through the autoconfirmed barrier. Callanecc (talkcontribslogs) 23:34, 11 April 2016 (UTC)[reply]
  3. very weakly, but if we are going to let admins do this in our name then we need real criteria --In actu (Guerillero) | My Talk 14:41, 13 April 2016 (UTC)[reply]
  4. First choice. Tarapia tapioco (aka Salvio) (talk) 12:57, 15 April 2016 (UTC)[reply]
  5. Second choice. Doug Weller talk 14:11, 22 April 2016 (UTC)[reply]
  6. Yes, needs to be spelled out. This is main reason it would be used. Cas Liber (talk · contribs) 10:57, 24 April 2016 (UTC)[reply]
  7. If individual admins are going to be doing it, as seems likely to pass, it needs definite limits. The ones set forth here are appropriate. Other cases if necessary can be dealt with by arb com. DGG ( talk ) 22:20, 1 May 2016 (UTC)[reply]
  8. Kirill Lokshin (talk) 23:54, 7 May 2016 (UTC)[reply]
  9. Keilana (talk) 17:53, 9 May 2016 (UTC)[reply]
Oppose
  1. Don't think it needs to be spelled out this thoroughly. --kelapstick(bainuu) 22:05, 11 April 2016 (UTC)[reply]
  2. We already have a standing rule that this be applied to all pages in the I-P topic area. Doing this article-by-article with these rules is too burdensome to enforce an existing decision. Courcelles (talk) 17:02, 13 April 2016 (UTC)[reply]
  3. See my comments below. GorillaWarfare (talk) 03:08, 18 April 2016 (UTC)[reply]
Abstain
  1. Drmies (talk) 17:49, 22 April 2016 (UTC)[reply]
The above provision passes in preference to the two I'm hiding. Callanecc (talkcontribslogs) 09:49, 2 May 2016 (UTC)[reply]

C2) Extended confirmed protection may only be applied where other methods (such as semi protection) have not adequately controlled disruptive edits. This provision does not apply to a page or topic area which has been placed under 30/500 protection by the Arbitration Committee.

Support
  1. kelapstick(bainuu) 22:05, 11 April 2016 (UTC)[reply]
  2. Second choice - although IMO the socking angle is more important than the failure of alternatives. Opabinia regalis (talk) 23:19, 11 April 2016 (UTC)[reply]
  3. Close second choice, I'd prefer the use to be a little more limited, however I can live with this, primarily that other methods (semi) need to be tried first. Callanecc (talkcontribslogs) 23:34, 11 April 2016 (UTC)[reply]
  4. Second choice. Tarapia tapioco (aka Salvio) (talk) 12:57, 15 April 2016 (UTC)[reply]
  5. First choice. Doug Weller talk 14:11, 22 April 2016 (UTC)[reply]
  6. Second choice. Bit general but whatever... Cas Liber (talk · contribs) 10:59, 24 April 2016 (UTC)[reply]
Oppose
  1. this is too open. --In actu (Guerillero) | My Talk 14:41, 13 April 2016 (UTC)[reply]
  2. Again, we have a standing decision authorizing its use throughout an entire topic area. Courcelles (talk) 17:02, 13 April 2016 (UTC)[reply]
  3. See my comments below. GorillaWarfare (talk) 03:08, 18 April 2016 (UTC)[reply]
  4. Too broad, if it is going to be applied by individual admins. DGG ( talk ) 22:20, 1 May 2016 (UTC)[reply]
Abstain
  1. Drmies (talk) 17:49, 22 April 2016 (UTC)[reply]

C3) <no restrictions on use>

Support
  1. Within the I-P topic area only. Courcelles (talk) 17:02, 13 April 2016 (UTC)[reply]
Oppose
  1. kelapstick(bainuu) 22:05, 11 April 2016 (UTC)[reply]
  2. Opabinia regalis (talk) 23:19, 11 April 2016 (UTC)[reply]
  3. While we don't specify what a full protection discretionary sanction can be used for, the protection policy is quite clear on what it can be used to control (primarily where there is no other means of control). 30/500 is very new and there isn't the guidance and culture around its use yet. Callanecc (talkcontribslogs) 23:34, 11 April 2016 (UTC)[reply]
  4. --In actu (Guerillero) | My Talk 14:41, 13 April 2016 (UTC)[reply]
  5. Tarapia tapioco (aka Salvio) (talk) 12:57, 15 April 2016 (UTC)[reply]
  6. See my comments below. While I don't think we should create policy on where it can be used, I do think we need to say that it should not be removed where we've imposed it as a part of an arbitration decision. GorillaWarfare (talk) 03:08, 18 April 2016 (UTC)[reply]
  7. Doug Weller talk 14:11, 22 April 2016 (UTC)[reply]
  8. Very much too broad. DGG ( talk ) 22:20, 1 May 2016 (UTC)[reply]
Abstain
  1. Drmies (talk) 17:49, 22 April 2016 (UTC)[reply]
  2. Not sure what this question is aiming at Cas Liber (talk · contribs) 10:59, 24 April 2016 (UTC)[reply]

Discussion by arbitrators

  • Note that this is, in my mind, only applicable to the use of 30/500 when used as a Discretionary Sanction/Arbitration Enforcement. Should the community decide to extend the use of this protections use beyond DS/AE (and amend the protection policy), that is outside our remit. In other words, the committee does not own this protection level on the whole, but we can determine how it is used to for DS/AE. If that makes sense.--kelapstick(bainuu) 22:05, 11 April 2016 (UTC)[reply]
  • @BethNaught: They don't affect the ARBPIA3 remedy. As Kelapstick says, the aim is to develop a structure for how to use this tool going forward in the DS/AE area. @Liz: The intention is to distinguish the PIA3 case - where this is already authorized and no consensus is needed - from other areas. Opabinia regalis (talk) 22:42, 11 April 2016 (UTC)[reply]
  • @Liz: I've clarified that, it was supposed to be as a discretionary sanction rather than as an AE action. Fixed now. Callanecc (talkcontribslogs) 00:00, 12 April 2016 (UTC)[reply]
  • @Courcelles: I think you may have misinterpreted, these provisions don't apply to a page or topic area which is placed under 30/500 by the Committee but rather to admins imposing it on their own cognisance in an area subject to discretionary sanction. I've (hopefully) made that clearer in the provisions. Callanecc (talkcontribslogs) 10:57, 14 April 2016 (UTC)[reply]
  • We should ensure that people are not removing 30/500 protection from articles where we have imposed it; now that the technical restriction is in place, there is no reason to remove it and go back to the earlier state of affairs where edits were being undone if the editor did not meet requirements. That said, I feel strongly that the Arbitration Committee should not be creating new policies as we would be doing if we pass B1, B2, C1, or C2. The community is entirely capable of deciding how they want this new feature to be used, and codifying it in policy. The Arbitration Committee should not do this for them. GorillaWarfare (talk) 03:04, 18 April 2016 (UTC)[reply]
  • I think it advisable to go cautiously on this until we get experience. While GorrilaWarfare is correct that we should not in general be making new policy, this is so directly related to those matters which are under our specific control, that it is the most feasible way to proceed. DGG ( talk ) 22:24, 1 May 2016 (UTC)[reply]

Community comments

  • DS already allows that any "uninvolved administrator may impose on any page or set of pages relating to the area of conflict semi-protection, full protection, move protection, revert restrictions, and prohibitions on the addition or removal of certain content (except when consensus for the edit exists)". 30/500 is actually a step down from full protection. It allow admins to apply a protection to an article or topic at a level that is less restrictive than what is currently allowed. In other words, it adds a new scalpel to the tool tray that cuts slightly finer than one already on it. And, 30/500 isn't particularly exotic. It is the existing semi-protection, which is 4 days/10 edits, with somewhat more restrictive parameters. 30/500 should just be another option available to admins at AE, in addition to all the other ones already available. Zad68 14:25, 11 April 2016 (UTC)[reply]
  • Comment on B1 and B2. The alternatives offered are that "Extended confirmed protection may only be applied as an arbitration enforcement action where there is a consensus of uninvolved administrators supporting its use" versus "Extended confirmed protection may applied as a discretionary sanction by any uninvolved administrator at their discretion." There's an abyss between those alternatives, and I'm not 100% sure about leaving it all up to individual admin discretion, as Zad proposes, but I think it should be tried and then perhaps assessed in six months or so months. Also, I want to point out that caste articles are a special case (as are Gamergate articles, but that's not my baby). Bear with me and I'll explain:
After Zad68 had the first brainwave about Gamergate,[1] in May 2015, some of the caste editors and admins took up the idea with respect to caste articles. Please look at the caste AE decision, and you will see that four caste articles and one talkpage were immediately placed under 500/30 restrictions (Nair, Jat people, Vanniyar, Bhumihar, and Talk:Nair), and it was emphasized that any placing of further caste articles under 500/30 should be within admin discretion. If it was to involve asking for consensus, going to AE, or any other of the usual pushing-a-locomotive-up-an-incline effort, there would be little point to the whole thing. It was so decided. Therefore, I'm a little alarmed to see ArbCom not only taking over the community's idea (you're welcome) but proposing in B1 that extended confirmed protection (from now on?) requires a consensus of uninvolved administrators supporting its use. I hope that's not meant to imply that admins in caste areas should henceforth be constrained by a consensus requirement. Caste articles are something so exotic to most admins that it's cumbersome and timeconsuming to try to get consensus for anything in the area. The original AE discussion, with its carefully detailed decision, should be enough. Note incidentally that there has not been a bonanza of 500/30 restrictions on caste articles since the decision in September 2015 — in fact I'm not aware of a single caste page being placed under the restrictions since then, beyond the original five. Caste admins haven't shown themselves to be trigger-happy. Bishonen | talk 16:00, 11 April 2016 (UTC).[reply]
  • Hi Bishonen, I share your concerns around unilateral admin action; especially as a discretionary sanction would not be reversible by other admins. Would "a consensus of 3 uninvolved administrators" bridge the gap? - Ryk72 'c.s.n.s.' 17:13, 11 April 2016 (UTC)[reply]
I share some of Bishonen's concerns above. The matters of getting enough admins involved, and, maybe, determining who might and might not be "uninvolved" in some way could maybe be clarified. If we were dealing only with 3 such admins, for instance, 2 new admins coming in with opinions based on a poorly phrased story or article could basically overturn the reasonable prior consensus, and, maybe, screw up everything for a while. Also, meaning no disrespect to any admins involved, would possibly any admin who is from India, or of Indian descent, or similar, qualify as "involved" based on their involvement in the Indian cultural milieu in which the caste system was created? John Carter (talk) 19:40, 11 April 2016 (UTC)[reply]
  • John Carter, it would be a very terrible idea to sideline Indian admins, i.e. those few admins in this thinly-adminned area who know the subject best, as involved. I only wish we had more Indian admins active on caste pages, and more Indian admins altogether. Bishonen | talk 10:11, 12 April 2016 (UTC).[reply]
  • Wholehearted agreement with you for the most part. However, like with some of the religion articles and topics, there is a reasonable possibility that in some cases, like breaking or controversial stories or sources regarding such a topic, any individual's pre-existing biases might cause them to react more forcefully to a potentially dubious story than circumstances warrant. The same has been known to happen a lot in religion topics. One thing which I think might help a hell of a lot would be to have someone review the comparatively few obvious reference works out there that deal with the topic and find what they cover and where, similar to some of the pages at Category:WikiProject prospectuses. I looked at the list of such works, and libraries holding them, generated at WorldCat, and, unfortunately, don't find any copies within a few hundred miles of me. If anyone with easier access to them would review them, though, that might be very useful. John Carter (talk) 14:18, 15 April 2016 (UTC)[reply]
  • John, how do you define pre-existing bias? Are we now going to require admins to identify themselves on the basis of nationality, ethnicity, language, race, caste, religion, tax filing status etc? Would this same yardstick be applied to American admins on WP:ARBAP2, all humans in the case of abortion, anyone who has filed an IRS tax return on WP:ARBSCI? —SpacemanSpiff 14:49, 15 April 2016 (UTC)[reply]
  • How do these motions affect the general prohibition in ARBPIA3? Do they forbid the use of 30/500 protection on relevant pages which have not yet been subject to disruption? If so, that contradicts the universality of the prohibition. BethNaught (talk) 22:11, 11 April 2016 (UTC)[reply]

30/500 is a new policy to Wikipedia, I think that it is very good, ArbCom should open a public vote about this group and its protection and policies. 333-blue 11:58, 21 April 2016 (UTC)[reply]

  • I feel as if I missed something here. The last community discussion I remember approved 30/500 as a protection level to be applied by Arbcom decision only. All of the proposals here seem to be assuming that it will be applied without explicit Arbcom decision as part of AE, or even without AE. I have two questions. First - has there been a community discussion approving the extendedconfirmed/30-500 protection for general use, not as an Arbcom-imposed remedy? If not, there needs to be before this discussion takes place. Second - if I missed something and 30/500 has been approved by the community for general use, why are there discussions pertaining to it tucked away here instead on at VP? This is a really important discussion with wide-ranging consequences, and it seems really inappropriate to hash it out on this sub-subpage only frequented by a very select community. Taking myself as an example, I take interest in policy and particular interest in protection levels and watch the VP as a result, but only discovered this page accidentally by clicking through links from the Signpost. Given how few comments there have been on such important questions, I imagine there are many others like me. These discussions desperately need a broader forum and any decisions reached in this rarefied environment should be taken with a big grain of salt. A2soup (talk) 01:36, 25 April 2016 (UTC)[reply]
Comment by Liz
  • B1) Extended confirmed protection may only be applied as an arbitration enforcement action where there is a consensus of uninvolved administrators supporting its use. This provision does not apply to a page or topic area which has been placed under 30/500 protection by the Arbitration Committee.
I don't understand this. I would think that extended confirmed protection provision could only apply to "a page or topic area which has been placed under 30/500 protection by the Arbitration Committee." This provision seems to say that a consensus of uninvolved administrators can apply extended confirmed provision on any article topic they wish, whether narrowly defined (by page, as in caste restrictions) or broadly defined (anything having to do with the Israel-Palestine conflict).
It seems uncharacteristic of ArbCom to encourage the regular application of a new restriction that has only now been implemented across several areas. I would think that ArbCom would prefer to see how the 30/500 restriction works in practice and then, later, consider it as a regular remedy admins can turn to in AE discussions. I would be remiss not to point out that sometimes AE cases have the participation of one or two admins and rarely more than four or five. So, a consensus of uninvolved administrators could, in practice, be one or two admins who are weighing in on a complaint. Personally, considering the wide impact these restrictions could have (especially on uninvolved IP editors since, thus far, these restrictions have never been lifted once they are imposed), I would urge the committee to set a higher bar for approval. Liz Read! Talk! 19:38, 11 April 2016 (UTC)[reply]
Maybe, at least specifying a minimum number of required admins to be involved to determine consensus, for instance? John Carter (talk) 19:43, 11 April 2016 (UTC)[reply]
I fully agree with Liz on this point. My impression was that the community had only okayed 30-500/extendedconfirmed protection as an Arbcom (not AE) remedy. Am I mistaken? If not, can someone point me to the community discussion in which this protection level was approved for general use? A2soup (talk) 01:36, 25 April 2016 (UTC)[reply]

Extendedconfirmed user group

D1) Administrators are not permitted to remove the extendedconfirmed user group as a discretionary sanction.

Enacted - Miniapolis 16:24, 15 May 2016 (UTC)[reply]

Support
  1. I agree with the statement by NE Ent, and my preference would have been for this to be similar to autoconfirmed, where it was automatically added, but could not be taken away. --kelapstick(bainuu) 21:57, 11 April 2016 (UTC)[reply]
  2. My far-and-away first choice on how to manage this is to make the extendedconfirmed usergroup technically impossible for admins to revoke. Failing that, I think it should be absolutely prohibited for any reason other than the explicit request of the affected user. Opabinia regalis (talk) 23:05, 11 April 2016 (UTC)[reply]
  3. Removing it shouldn't be needed, topic bans would be much more preferable. Callanecc (talkcontribslogs) 23:22, 11 April 2016 (UTC)[reply]
  4. --In actu (Guerillero) | My Talk 14:38, 13 April 2016 (UTC)[reply]
  5. Per kelapstick, Opabinia regalis, and Callanecc. 30/500 is a very binary thing: either a user meets the 30 days/500 edits criteria, or they don't. While I see the benefit to adding it (in the case of alternate accounts), I do not think it should be removed. If a user cannot contribute constructively to a subject under 30/500 restriction, they should be topic banned. Otherwise we end up with a confusing scenario where someone who is, say, disruptive in the Arab-Israeli conflict suddenly finds themselves banned from editing Brianna Wu. GorillaWarfare (talk) 03:15, 18 April 2016 (UTC)[reply]
  6. Drmies (talk) 17:47, 22 April 2016 (UTC)[reply]
  7. Doug Weller talk 10:57, 24 April 2016 (UTC)[reply]
  8. Kirill Lokshin (talk) 23:53, 7 May 2016 (UTC)[reply]
Oppose
  1. This should be a community policy decision. Courcelles (talk) 17:06, 13 April 2016 (UTC)[reply]
  2. Tarapia tapioco (aka Salvio) (talk) 13:09, 15 April 2016 (UTC)[reply]
  3. I also think this should be up to the community; there is merit in both positions. DGG ( talk ) 00:30, 2 May 2016 (UTC)[reply]
Abstain
  1. Undecided on this. can see both sides. am of a mind to have community-wide input on this. Cas Liber (talk · contribs) 11:03, 24 April 2016 (UTC)[reply]

D2) Administrators must not remove the extendedconfirmed user group as means of bypassing defined arbitration enforcement procedures (for example, removing the user group as a normal administrative action to avoid banning an editor from the Gamergate controversy article).

Enacted - Miniapolis 16:24, 15 May 2016 (UTC)[reply]

Support
  1. @Kelapstick and Opabinia regalis: This proposal is different to D1. D1 is about removing it as a discretionary sanction, this proposal is about removing it as a normal admin action to avoid having to use arbitration enforcement procedures. Callanecc (talkcontribslogs) 23:22, 11 April 2016 (UTC)[reply]
  2. Callanecc I see what you mean, maybe all these options are just confusing me. I thought the first one was all encompassing "just don't do it". --kelapstick(bainuu) 23:38, 11 April 2016 (UTC)[reply]
    We originally considered proposing this however the concern was that saying 'don't remove it at all ever, no, nope' would be stepping on the community's toes by creating policy for them rather than defining only how admins use it in our name (discretionary sanctions). Alternatively we can make another proposal stating that the user group is not to be removed except if the community authorises removal for an individual or a process for removal. Callanecc (talkcontribslogs) 23:56, 11 April 2016 (UTC)[reply]
    I meant just don't do it as a means of AE/DS. And no more options, my head is already starting to hurt. If the community wants to develop a process for this, that's their prerogative. --kelapstick(bainuu) 00:01, 12 April 2016 (UTC)[reply]
  3. I also misread this (I blame Kelapstick, naturally ;) Opabinia regalis (talk) 23:41, 11 April 2016 (UTC)[reply]
  4. --In actu (Guerillero) | My Talk 14:38, 13 April 2016 (UTC)[reply]
  5. Tarapia tapioco (aka Salvio) (talk) 13:09, 15 April 2016 (UTC)[reply]
  6. Per my comments on D1. GorillaWarfare (talk) 03:15, 18 April 2016 (UTC)[reply]
  7. Doug Weller talk 10:57, 24 April 2016 (UTC)[reply]
  8. . Cas Liber (talk · contribs) 11:03, 24 April 2016 (UTC)[reply]
  9. This at least is an obvious and necessary precaution. DGG ( talk ) 00:30, 2 May 2016 (UTC)[reply]
  10. Kirill Lokshin (talk) 23:53, 7 May 2016 (UTC)[reply]
Oppose
As redundant to supporting D1. --kelapstick(bainuu) 21:57, 11 April 2016 (UTC)[reply]
Hmmm, yeah this seems to make sense, I guess. --kelapstick(bainuu) 23:38, 11 April 2016 (UTC)[reply]
Per Kelapstick. Opabinia regalis (talk) 23:05, 11 April 2016 (UTC)[reply]
  1. This should be a community policy decision. Courcelles (talk) 17:06, 13 April 2016 (UTC)[reply]
Abstain
These proposals are 0-5, suggesting they won't pass at all. Collapsing unless another Arb wants to support one of these. -- Amanda (aka DQ) 19:10, 15 April 2016 (UTC)[reply]
The following discussion has been closed. Please do not modify it.

D3) Administrators must not remove the extendedconfirmed user group as normal administrative action as means of bypassing defined arbitration enforcement procedures or remove the user group as a discretionary sanction.

Support
Oppose
  1. As redundant to supporting D1. --kelapstick(bainuu) 21:57, 11 April 2016 (UTC)[reply]
  2. Per Kelapstick. Opabinia regalis (talk) 23:05, 11 April 2016 (UTC)[reply]
  3. This one is redundant to D1 and to D2. Callanecc (talkcontribslogs) 23:22, 11 April 2016 (UTC)[reply]
  4. --In actu (Guerillero) | My Talk 14:38, 13 April 2016 (UTC)[reply]
  5. This should be a community policy decision. Courcelles (talk) 17:06, 13 April 2016 (UTC)[reply]
Abstain

D4) The extendedconfirmed user group may be removed as a discretionary sanction if the intention is to prevent the user editing every area covered by 30/500 protection both at the time and in the future.

Support
Oppose
  1. Being disruptive in one area covered by this does not default in disruption in another area by this. We have blocks and topic bans to deal with specific disruption. --kelapstick(bainuu) 21:57, 11 April 2016 (UTC)[reply]
  2. Per Kelapstick. Opabinia regalis (talk) 23:05, 11 April 2016 (UTC)[reply]
  3. Per Kelapstick. Callanecc (talkcontribslogs) 23:22, 11 April 2016 (UTC)[reply]
  4. --In actu (Guerillero) | My Talk 14:38, 13 April 2016 (UTC)[reply]
  5. This should be a community policy decision. Courcelles (talk) 17:06, 13 April 2016 (UTC)[reply]
Abstain
  1. Undecided on this. can see both sides. am of a mind to have community-wide input on this. Cas Liber (talk · contribs) 11:03, 24 April 2016 (UTC)[reply]

Discussion by arbitrators

  • @Ryk72: The scope of this discussion is the usage of this tool in DS/AE areas - i.e., places that are already under arbcom's remit - and not to produce binding constraints on what the community might choose to do with it. My personal (strong) preference is to treat this as effectively a pilot project. Wait to see how this works in a few known trouble spots over the course of a few months - long enough for a few cycles of new users to pass the thresholds - before starting a broad community RfC informed by the experience gathered from the DS/AE applications. I don't think what we know so far from the previous implementations of the idea (manual reverting and edit filters) is enough information. Opabinia regalis (talk) 22:45, 11 April 2016 (UTC)[reply]
    • To expand on my votes above: the implementation of this usergroup for the purpose of administering the PIA3 remedy comes with significant potential unintended consequences, one of which is the possibility to backdoor-enforce topic bans by technical means. This means that the entire editor base over the 30/500 thresholds has been newly exposed to an administrative "power" that didn't exist before, in a context where there is already common feeling that admins have too much power (regardless of whether that's true). Technically enforced topic bans have not explicitly been considered by the committee in this context, and IMO it would require an explicit affirmative community consensus before this user group could be used this way.

      More generally, I strongly oppose any mechanism that gives admins the ability to eject users from the (ahem) "established editor" group, which in the context of a heated dispute will inevitably be read as deeply disrespectful of the affected user's contributions. As it is not possible to revoke user flags for predefined limited time, this action would effectively put that user in the position of going cap-in-hand back to admins in the hopes of being readmitted to the club. As tempting as it doubtless sounds - "surely it's better to be half-blocked than blocked altogether?" - I think permitting the revocation of this right would be socially toxic far in excess of the possible advantages in borderline cases. Opabinia regalis (talk) 23:06, 11 April 2016 (UTC)[reply]

      • To the surprise of probably nobody who's read the above, I was the one who made the "nope not ever" proposal while we were first discussing this. I'm still unconvinced by the "creating policy" analysis; this user group exists entirely because an arbcom remedy effectively required it in order to be administered properly, and as a side effect, a new administrative power was created. Preventing the removal of the extendedconfirmed flag for any reason, pending development of community policy, would just be restoring the status quo ante. But "treat DS applications as a pilot project" has essentially the same effect. Opabinia regalis (talk) 18:30, 13 April 2016 (UTC)[reply]
  • Re to NE Ent and Ryk72 admins are not permitted to remove the user group period was one of the things proposed in the original discussion however it wasn't included here given the issue with ArbCom creating policy for the community rather than defining how its/our discretionary sanctions are enforced. Callanecc (talkcontribslogs) 23:22, 11 April 2016 (UTC)[reply]
  • @Courcelles: The community can't really define what can and can't be done as a discretionary sanction or what can and can't be done as arbitration enforcement. So whether (even if) the community decided one way or the other we'd need to pass a motion or statement either confirming that the result of the community RfC applies to arbitration enforcement and discretionary sanctions or that it doesn't. These provisions only apply to 30/500 as it relates to arbitration enforcement. Callanecc (talkcontribslogs) 11:03, 14 April 2016 (UTC)[reply]

Community comments

Comment by Ryk72

Respected Arbitrators, I thank you for the opportunity to comment on these motions; but also must state that it would have been preferable for these decisions to be handed to the community as a whole.

Given that we have limited data of the effect of this type of sanction (and some of that data would suggest that it has not been a positive effect; serving only to entrench article WP:OWNership), I encourage taking the most cautious approach at this stage, by selecting the least permissive options (B1, C1, D1/D0*). The usage can be opened up once we have additional data & feedback (in 6-12 months).

I also encourage the removal of Talk pages, and of Brianna Wu, from A1. This level of protection for Talk pages is not required to protect the encyclopedia mainspace. Talk pages should remain open to allow good faith suggestions; bad faith comments by 30/500 banned editors can be simply ignored. The Brianna Wu article had no recent history of disruptive editing by 30/500 banned accounts, and was protected by a single admin acting unilaterally, explicitly on the basis of comments in unspecified Reddit threads. This would seem a poor, even quixotic, reason for such protection.

This level of protection will be used in topic areas which are subject to discretionary sanctions, where tempers are already heightened, and suspicion of admin actions can be high. To protect our admins from accusations of impropriety, I believe that it is important that we make the workings of this protection level as automatic as possible. I propose that we do not allow admins to either add or remove the user rights. Therefore, please add "D0) Administrators are not permitted to remove the extendedconfirmed user group."

This would mean that users either qualify for extended confirmed or they don't, only on the basis of their history. Users gaming the system or proving disruptive can be dealt with as usual (blocks &/or topic bans).

Apologies for the length; please let me know if anything is unclear or if you have any questions. - Ryk72 'c.s.n.s.' 14:51, 11 April 2016 (UTC) - amended Ryk72 'c.s.n.s.' 15:37, 11 April 2016 (UTC)[reply]

  • Ryk72 at Brianna Wu I don't agree with your assessment of the history. Look here. On 11 April 2015 the article was at FULL protection. On 26 September HJ Mitchell tried stepping it down to SEMI, the only protection level available down from FULL. Two days later there was an article edit by an autoconfirmed but not extended-auto user that was so bad it had to be oversighted. HJM had to put it back up to FULL. On 10 March, 30/500 was applied via edit filter. Since then, there haven't been isues. The evidence shows that at that article, 30/500 is working great and allowing experienced editors to make direct edits to the article, whereas without it they had to wait for an admin to apply the edits for them. Zad68 15:23, 11 April 2016 (UTC)[reply]
  • Hi Zad68, Minor observations - No changes were made while the article was under full protection, so no waiting for admin was required; 30/500 was applied 19 days before full protection was due to expire; we have no way of knowing if it was required - it's a long time from September to March. The unilateral decision is troubling, and "unspecified Reddit threads" remains a poor rationale for any protection. Caution is still preferred. - Ryk72 'c.s.n.s.' 15:33, 11 April 2016 (UTC) amended Ryk72 'c.s.n.s.' 15:46, 11 April 2016 (UTC)[reply]
  • Commenting here as an administrator, not as an arbitrator, as it was in that capacity I applied the 30/500. Framing the rationale as merely "unspecified Reddit threads" is being willfully obtuse. As an experienced editor in this topic area, you know the many, many problems caused by new editors attempting to slur living individuals in this topic area and you know that it would be inappropriate to link to those specific threads due to BLP. When there is off-site coordination targeting the article of a BLP in a topic area where there has been frequent off-site coordination from the very same forums, it would be irresponsible to "wait and see" if further action was needed to protect a BLP who has already been the target of onsite disruption. As far as the usefulness of the sanction, I can think of at least two disruptive editors from the talk page off the top of my head who were prevented from editing the article. If this sanction was so obviously inappropriate, then you should have appealed this at WP:AE when it was imposed. Or now. Gamaliel (talk) 17:35, 11 April 2016 (UTC)[reply]
  • While my familiarity with Reddit is obviously not as great as that of those who actually use the site, as a casual observer, the whole "co-ordination of harassment" narrative seems terribly overblown - I have never seen a Reddit thread which suggests "let's go to Wikipedia and fuck shit up", far less one which suggests "let's go to Wikipedia and fuck this person's article up". I have seen a lot of disquiet with our articles; with admin actions or the lack thereof; and a good deal of satire (well deserved for the most part; and certainly no worse than WO or SLOWP).
    The biggest concern in this instance, however, would be that, at the time the sanction was imposed, only two active threads [redacted] were discussing that page in any manner; neither of these makes any suggestion of an attack on that page; both question the removal of a Talk page discussion about a potential COI, by the editor being discussed; which, in cynical minds, would raise the question of who exactly we are protecting. This comment would also, for those minds, raise the same concern that the person being protected is that editor, and not the article subject. I would prefer administrators, by virtue of making a decision collaboratively, be freed from the possibility of such cynical accusations. It may, of course, be that other Reddit threads exist which actually do contain evidence of an "off-site co-ordination of attacks"; if so, please feel free to contact me privately with details - I will be happy to sit in the corner wearing a funny hat - until then, I will maintain that we should be cautious in allowing unilateral administrative application of this protection level. Where we do have a well-founded apprehension of attacks against living persons that cannot be evidenced on Wiki, I'd prefer to see a decision involving more than one admin. - Ryk72 'c.s.n.s.' 12:30, 13 April 2016 (UTC)[reply]
  • No, we're not liking to offsite attacks and harassment. You know better than that. Gamaliel (talk) 12:39, 13 April 2016 (UTC)[reply]
I agree caution is preferred. But when it comes to a BLP, that may mean being more protective then we would be in a case where such concerns don't arise. In particular, if we get even a very small number of edits which require suppression, it may very well mean we need to look at greater protection. Particularly in a page where the requirement for 500/30 is not excessively problematic (unlike if we were to implement it at, say, ANI). Nil Einne (talk) 18:29, 11 April 2016 (UTC)[reply]
Echong what was said above, Brianna Wu is a LP and so it's likely many of the comments on the talk page of her article will relate to her or other LPs. Even though comments on talk pages have less visibility then stuff that happens in article, we cannot simply ignore BLP problems but have to deal with them including by deletion when they are severe enough. If this happens often enough, this suggests we need to stop it happening. Nil Einne (talk) 18:22, 11 April 2016 (UTC)[reply]
Comment by Jo-Jo Eumerus

Just a brief comment before I return to images and articles; I'd say that this protection/permission system should be used in the same way as any other arbitration sanction. That also means that removing the permission should happen as a sanction like any other AE topic ban or block as an enforcement measure. For example, if an account is banned from editing a certain area where 500/30 protections apply, withdraw the permission. When the ban expires or is successfully appealed, the permission can be regranted. If the community decides to extend extendedconfirmed and associated protection to non-arbitration uses they should have their own policy created. Jo-Jo Eumerus (talk, contributions) 15:20, 11 April 2016 (UTC)[reply]

Comment by NE Ent

Please no, don't let admins take away the extendedconfirmed permission, because one of the unwritten rules of Wikipedia dispute resolution is anything that can be removed will be argued about incessantly to get restored. If an editor reaches 500/30 and hasn't figured out how not to act like a dweeb, the appropropriate sanctions should be applied everywhere, not just the 500/30 articles. NE Ent 20:40, 11 April 2016 (UTC)[reply]

Comment by T. Canens

There is at least one known incident in which two editors attempted to circumvent a similar restriction by creating articles with hundreds of edits, each adding a single word. The editors involved was indefinitely blocked as the actions were obviously in bad faith under the circumstances, but it is not inconceivable that a genuinely clueless new editor may attempt something similar, in which case removing the permission obtained from the artificially inflated edit count (until the 500 edits have been made in the normal course of editing) seems like an appropriate remedy. T. Canens (talk) 05:14, 13 April 2016 (UTC)[reply]

Comment by Salvidrim

Has the committee considered the possibility that it shouldn't be the one to decide when and how 50/300-protection and the extconf user-right should be used, but that this decision should be left to the community (just like it was last time we discussed protection levels with PC1/PC2 and user-right levels with Template-protection and template-editors)? Of course, ArbCom is always free to place articles under protection as remedies (50/300 or otherwise), and such Committee actions would be unrevertable (as all Committee actions are), but I guess I'm just a bit uneasy at this decision being taken mostly out of the hands of the community.  · Salvidrim! ·  19:19, 13 April 2016 (UTC)[reply]

@Salvidrim!: We're not defining how the community can use it, in fact the provisions have been written in a way which allows the community to use the protection level however they wish. This discussion and voting is about defining how the extendedconfirmed user group and 30/500 should be used in arbitration enforcement where, once an admin action is taken, it is difficult to have it changed. And also to define what administrators are doing in the Committee's name. The community is absolutely able to define how and when 30/500 protection is used in its name and how and when the user group can be removed (and I sincerely hope that it does very soon). If the community decides that the user group can be removed then admins will be able to do this, just not (assuming that provision passes) as a discretionary sanction. When/If there is a community RfC, and the results of that are different to what ends up being decided here then I'd be very open to changing what we have done here, but until then some guidance from the Committee on what is done in its name and with its protection is needed IMHO. Callanecc (talkcontribslogs) 11:11, 14 April 2016 (UTC)[reply]
Comment by Davidwr

Administrators should be allowed to remove this user-right as a less-severe substitute for a more-severe sanction that they are already authorized to implement, provided that the removal of this user-right, alone or in combination with other authorized actions, would be no more severe than the "more-severe" sanction or collection of sanctions that would otherwise be imposed.

For example, if an otherwise-well-behaved editor was repeatedly violating discretionary sanctions on an article related to the Arab-Israeli conflict and as a result is already topic-banned, violating that topic-ban would normally warrant a limited-time block. However, removing this user-right for the shorter of the duration of the topic band or the duration of the block he would get if removing this user-right were not allowed is better than imposing a block: It allows the editor to continue contributing in other areas of the encyclopedia that are not affected by this user-right.

In other words, think of removing this user-right as a "mini-block" - a way to enforce an existing topic ban where the editor has violated the ban and where a "regular block" would be worse for the encyclopedia as a whole than removing this user-right. davidwr/(talk)/(contribs) 02:33, 27 April 2016 (UTC)[reply]

Comment by Anarchyte

I believe admins shouldn't be able to remove this right because, hopefully, by the time an editor has reached the 30/500 mark, they know the basic rules and policies of the site. If they don't know how to edit a page under these special sanctions (where all the rules for that article are in a nicely placed editnotice), they won't know how to edit other articles. Anarchyte (work | talk) 00:50, 15 May 2016 (UTC)[reply]

Implementation notes

For these motions there are 14 active arbitrators. 8 support or oppose votes are a majority.

Majority reference
Abstentions Support votes needed for majority
0 8
1–2 7
3–4 6

Statement

Support: 9
Oppose: 0
Abstain: 0
Status: Passed

Extended confirmed protection

Imposition
B1
Support: 2
Oppose: 7
Abstain: 2
Status: Six support votes needed to pass
B2
Support: 4
Oppose: 4
Abstain: 2
Status: Four support votes needed to pass
Requirements for use
C1
Support: 9
Oppose: 3
Abstain: 1
Status: Passed
C2
Support: 6
Oppose: 4
Abstain: 1
Status: Two support votes needed to pass
C3
Support: 1
Oppose: 8
Abstain: 2
Status: Can't pass

Extendedconfirmed user group

D1
Support: 8
Oppose: 3
Abstain: 1
Status: Passed
D2
Support: 9
Oppose: 1
Abstain: 0
Status: Passed
D3
Support: 0
Oppose: 5
Abstain: 0
Status: Can't pass
D4
Support: 0
Oppose: 5
Abstain: 1
Status: Can't pass