Wikipedia:Requests for comment/BASC reform 2014

• Should the community adopt the changes to the makeup and procedures of the Ban Appeals Subcommittee proposed below? Beeblebrox (talk) 18:25, 15 October 2014 (UTC)

Background

Up until now the Ban Appeals Subcommittee has consisted entirely of members of the full Arbitration Committee, and all Arbitrators are subscribed to the subcommittee's mailing list. Ban appeals is possibly the most active of all lists associated with functionaries and arbitrators. And yet, only a small fraction of appeals are actually accepted. This has resulted in arbitrators spending a great deal of their time discussing ban appeals, with little tangible benefit to the community. However, the subcommittee is a vital component of the blocking/banning appeals process, serving as a final, off wiki venue and an important safety valve in the event of an improperly applied block or ban. Therefore, doing away with it entirely is not a viable option but it is desirable to shift this responsibility away from the committee and into the hands of the community. It is hoped that unbundling this process from the committee will result in both committee business and ban appeals being handled in a more timely and efficient fashion. (The current turnaround time for ban appeals is 3–5 weeks.)

Experience has also shown that there are many serial appellants, who repeatedly contact the subcommittee out of a desire to overturn their sanctions, sometimes attempting to bargain by promising to cease evading their ban and so forth. This proposal will endeavor to define the subcommittee as an absolute final appeal of any block or ban.

In short, if these changes are implemented, the basic purpose of the subcommittee is unchanged. Its composition will be altered and there will be a greater emphasis on the idea that this is final appeal of any ban or block, and long-term banned users who make repeated appeals will be given less license to take up the subcommittee's time.

Who may appeal to BASC?

(this section is copied from the current BASC procedures page and is presented simply for informational purposes, no change is being proposed to these parameters)

The Ban Appeals Subcommittee (BASC) was formed by the Arbitration Committee to hear appeals, by email, from users who are banned or blocked by community processes. BASC is a last resort, available only if other means of appeal are exhausted.

If you:

1. are community banned from Wikipedia entirely, or you are blocked indefinitely; and
2. have appealed your block using {{unblock}} on your talk page and using the UTRS interface,

then you may submit an appeal to BASC. Please read Wikipedia:Appealing a block and the next section of this page before doing so.

BASC does not hear appeals of topic bans, namespace or single-account restrictions, or appeals of block of short duration. Appeals of remedies imposed by the Arbitration committee must be appealed directly to the committee itself, not to BASC.

Proposed new structure

The Ban Appeals Subcommittee will be composed of one current arbitrator, three functionaries appointed by the Arbitration Committee and three community members appointed by the Committee following an advisory processes similar to the one already used to choose functionaries. The term of the arbitrator member will last six months; of functionaries and community members, one year. Consecutive and multiple terms are allowed. As deleted or suppressed material is often pertinent to ban appeals, due to the Wikimedia Foundation's restriction on non-administrators being given access to deleted revisions without passing an RFA or RFA-identical process, editors who are not administrators are ineligible for community membership of the Subcommittee

Members of the Ban Appeals Subcommittee will be identified to the Wikimedia Foundation, given the CheckUser permission, and have access to the arbcom-appeals-en, Functionaries-en, Checkuser-l, mailing lists and archives, and the CheckUser IRC channel if requested. Alternative (or reserve) members of the subcommittee do not have such rights unless they are appointed to the subcommittee during their reserve term. All current arbitrators will be permitted but not required to subscribe to the BASC mailing list and participate in discussions there as desired, but for purposes of determining the status of appeals only a consensus of subcommittee members is required. All members of the subcommittee will be considered equal when voting on an appeal.

If a member of the subcommittee was involved in the issuance or previous review of a block or ban being appealed they should recuse themselves from the case, but may present Checkuser or Oversight/RevDel evidence to the other members if pertinent.

The role of the arbitrator member of the committee will be to serve as a liaison with the full committee, as well as being a voting member of the subcommittee. Additionally, they will serve as the archivist of BASC proceedings, which will continue to be recorded at the arbwiki. Records of previous appeals from specific users may be copied from the current archives and shared with the subcommittee.

Discussion of proposed new structure

• Support seems like a logical way to spin ban appeals off. By giving three seats to functionaries, the three at large seats should always be filled with non-functionary admins. This should give the widest array of opinions a seat at the table --Guerillero | My Talk 18:49, 15 October 2014 (UTC)

There is no 3:3 split. The split as laid out in this proposal is 4:3. Samsara (FA • FP) 15:13, 20 October 2014 (UTC)

• Support Optimal idea. → Call me Hahc21 20:29, 15 October 2014 (UTC)
• Oppose The existing sub-committee is the final appeal and there's no purpose in creating another. Ban all your serial appellants; problem solved. Chris Troutman (talk) 21:02, 15 October 2014 (UTC)
  • Why should arbs be the only people deciding on this? --Guerillero | My Talk 21:31, 15 October 2014 (UTC)
  • The purpose of this RfC is not to create a new BASC, but to modify the existing one to remove abig chunk of workload from the arbitrators' hands and into the community. → Call me Hahc21 23:58, 15 October 2014 (UTC)
    • ARBCOM is made up of very experienced Wikipedians, typically all experienced admins, and have been elected by the largest community consensus available on Wikipedia. I neither seek to have these decisions made by a lesser group nor do I think another layer of bureaucracy would help. If the arbs are overwhelmed with workload, I simply point out that much effort is probably wasted on repeat offenders. If ARBCOM simply doesn't want this task, I'd like to see a statement in writing from the group explaining such and we can discuss alternatives. Chris Troutman (talk) 22:56, 20 October 2014 (UTC)
      • Considering the number of former and current arbitrators saying similar things in many places over the last several months, including on this page, I don't think such a statement is necessary. --Rschen7754 01:25, 21 October 2014 (UTC)
• Absolute support. We do need an ArbCom (and yes, that's coming from me!), but we need it for arbitration. Over the years, it has acquired (or in some cases grabbed) all sorts of responsibilities that have little or nothing to do with arbitration. I'm very pleased that we've got the sanest and ArbCom for a while and that they're finally trying to get rid of some of these things. A community process ought to perfectly capable of handling block appeals, and this should have a knock-on effect in reducing arbs' workloads and thus making ArbCom more effective and perhaps encouraging more sane people to stand. I can't see a downside here. HJ Mitchell | Penny for your thoughts? 21:45, 15 October 2014 (UTC)
• Support, with very high enthusiasm for the Committee cutting back on its workload in this area. --Tryptofish (talk) 23:07, 15 October 2014 (UTC)

  After reading subsequent comments, I want to add some points. I have no problem with the way that Beeblebrox put this discussion together, and I thank him for it. I also would be just fine with saying that ArbCom will only hear appeals of bans imposed by ArbCom, full stop. Let the community (try to) handle the rest, and the community can approach ArbCom when that doesn't work out. --Tryptofish (talk) 19:15, 16 October 2014 (UTC)

• Oppose. First off, if you're planning to include functionaries in a committee, it would be a good idea to poll the functionaries and see if they're interested in taking on this work. You've not done that, and in fact haven't informed the functionaries about this proposal at all. Secondly, BASC is unnecessary. The only reason Arbcom is at all involved in block appeals is that it took it upon itself back in 2007 to review certain blocks, without any mandate from the community, and without any really good reason except that it seems they wanted to unblock someone with the intention to trap them and reblock them. BASC has resulted in almost no unblocks and a massive amount of work for arbitrators. The community is perfectly capable of handling unblock requests through existing processes. The Arbitration Committee should narrow the scope of BASC to review of blocks directly linked to Arbcom cases only, and devolve all other unblock requests. Just remove them from your portfolio. Risker (talk) 00:46, 16 October 2014 (UTC)

I admit I did not poll the functionaries before posting this. I didn't poll the rest of ArbCom or the current BASC either. I haven't polled admins before proposing changes that would effect them either. Policies belong to the community, not any subset of it, even if it does impact them directly.

I have discussed this in very general terms with a few of the other arbs and some of them would agree with you that BASC should just go away. The reason I do not feel that way is that I think we do need an off-wiki, final, binding appeal. The offwiki aspect can help to promote a calmer discussion without the peanut gallery chiming in with unhelpful or inflammatory comments. Having it be a group, instead of one single admin or UTRS agent helps insure against snap judgements and simple mistakes. We are talking about telling people that they have to go away for a long time, we should be sure that is the right decision. That is why I believe BASC remains relevant, it just needs a major overhaul in order to function more efficiently. Beeblebrox (talk) 01:42, 16 October 2014 (UTC)

Well, you're incorrect in this. You, an arbitrator, are proposing to amend an Arbitration Commmittee policy. It's disturbing that you're openly stating that you haven't even sought the support of your colleagues, or the people whom you've designated to carry out these responsibilities. If you were paying attention to unblocking generally, you'd know that there are already enough challenges to keeping UTRS properly staffed and supported by checkusers, that functionaries are already working very hard (especially as so many are both checkusers and oversighters), and now you're proposing to take them out of the existing system and stick them into a complex bureaucracy and have an arbitrator act as clerk. As well, the plan listed below is a significant rewrite of WP:BLOCK, and I don't see any effort to have changed the unblocking aspects of the block policy before initiating this proposal for a pretty massive bureaucracy. Risker (talk) 02:01, 16 October 2014 (UTC)

Risker, you know me. We have our differences but I think you are aware that I have dealt with literally thousands of unblock appeals as an admin and several hundred oversight requests as a functionary. I know as well as you that these are endless tasks that often don't get dealt with as quickly as we would like. I am absolutely not proposing that any functionary be removed from their current duties in order to serve on BASC. I don't know where you are getting that idea, It would be an additional task done only by those willing to do it. If we can't find three functionaries and three rank-and-file admins then this idea will not work. I think we will be able to find them. I also don't see the word "clerk" anywhere in any of the proposed language. BASC doesn't need a clerk, but it's decisions need to be recorded somewhere, and currently that is done at the arbwiki. I considered proposing a whole new BASC wiki but thought it better to stick with the most urgent issues for now. Those issues are the very slow response time of the current subcommittee and the amount of time sunk into dealing with a very, very small group of people who have little to no chance of ever rejoining the community yet appeal repeatedly. I'm used to you never liking any of my big, crazy ideas by now, but I would appreciate it if you could limit yourself to arguing against things I actually have proposed instead of mis characterizing them. Beeblebrox (talk) 02:23, 16 October 2014 (UTC)

Umm, keeping track of decisions is indeed a clerical function; it's practically the definition of one, and it's the reason for Arbcom clerks. The current subcommittee is of course swamped; it already has a more than full plate, and this is a thankless job, which is inherently demotivating. It will not become more motivating if it is done by other people, nor is there any reason to believe that the request outcomes will be different. Simply put, given the finite number of volunteer hours functionaries are in a position to provide in those roles today, the addition of another responsibility that will consume in the range of 200-300 volunteer hours per month from the functionary team (the equivalent of more than 1 FTE in the working world) means it has to come out of the hours the functionary team is currently working on their primary roles, which already have significant backlogs. Without some statistical evidence that BASC unblocks have a positive effect on the community that is roughly equivalent to the amount of time (and burnout) invested in sorting the wheat from the chaff, it has not demonstrated its value. Several of the "rules" you've suggested below could be applied to current BASC processes if you feel the current processes are being abused. Or you could simply say "the Arbitration Committee will limit itself to reviewing only Arbcom case and enforcement blocks of greater than 6 months' duration" and leave the rest to the community. People who have genuinely followed the unblock expectations of the community are generally allowed back by the community, and Arbcom doesn't need to get into things at all. Risker (talk) 03:10, 16 October 2014 (UTC)

"The Arbitration Committee will limit itself to reviewing only Arbcom case and enforcement blocks of greater than 6 months' duration": that's a good idea too. Another idea would be to just have a new BASC comprised of 1 current arbitrator + 5 community elected users (admins, identified) who would then be handed CU and OS permissions. That way, the current functionaries team won't be affected by it. → Call me Hahc21 04:38, 16 October 2014 (UTC)

• A few questions: would subcommittee members have the ability to run unrelated checks with the CU permissions? Also, would they be provided with oversighted material if necessary? --Rschen7754 01:22, 16 October 2014 (UTC)

It was decided earlier this year that preventing AUSC members from using their tools was hampering both recruitment and sock investigations. However, this isn't exactly the same thing. I think it might be reasonable to say that anyone who already had CU should be perfectly free to continue using it as needed, but perhaps we should consider restricting it for the non-functionary members as this is not intended to be a back-door to becoming a permanent member of the functionaries team. I do think that permission to view both CU information and oversighted material is essential as both can be involved in ban appeals, but speaking for myself in 10 months on the subcommittee I have not used CU a single time, instead relying on others more familiar with the tool to gather the relevant evidence. So the permission to view such material, which will require identifying oneself to the Foundation, is mor important to the functioning of ban appeals than the actual use of the tools. Beeblebrox (talk) 01:35, 16 October 2014 (UTC)

Also, what happens if someone changes categories halfway through the year? i.e. becomes a functionary, becomes an arb, etc. --Rschen7754 04:50, 16 October 2014 (UTC)

• Oppose I agree with Risker almost completely. Additionally, if nothing else, tracking the ban appeals was a logistical nightmare. Is there not any way that WP:UTRS could be used for opening appeals for the folks who have been indefinitely blocked for a long time by the community? Or if some sort of WP:AN-ban is being appealed (as one should have some sort of ability to do in a more formal manner than to the ad hoc group that appears on that page), there is no reason that ArbCom's normal processes cannot be used for that purpose. Handling appeals of sockpuppeteers and people who were blocked two years ago and had maybe 500 edits to their name is not a good use of a Committee's time, yet that's what happened a fair amount. NW (Talk) 01:42, 16 October 2014 (UTC)

• @NuclearWarfare: Anything is possible really. Just let us know if you want something changed. Probs won't happen overnight, but we'll definitely get to it. -- DQ (ʞlɐʇ) 16:12, 16 October 2014 (UTC)

• @DeltaQuad: Just thinking aloud mostly. If there is a concern is that this leaves no avenue for people who have been indef blocked+talk page revoked to appeal, is there a way that UTRS could do a first pass appeal on the talk page lock without considering the merits of the full unblock (without it being too much work for the admins who monitor UTRS, of course). NW (Talk) 20:44, 16 October 2014 (UTC)

• Support in principle. I served on BASC during my term on ArbCom and I think the benefit of having a fresh set of eyes to review the sanctions should not be underestimated. Contrary to popular belief, I also believe that BASC has helped to overturn genuine cases where users may have fallen through the cracks (e.g. appeals of first instance where the sanction was manifestly excessive). I also believe that the role of BASC through its incentive element (of giving genuinely reformed cases a second chance) has actually helped to reduce workload at functionaries, which otherwise disruption, socking and other acts would deteriorate. As to where is the best place to keep the paperwork, I think arbwiki might pose some issues due to its tight security (and that will involve the coordinating arb to go back and forth to retrieve past files), and perhaps a separate workspace needs to be established to maintain a easy-to-access institutional memory (either a BASC wiki, OTRS or some other method). I think the objections raised with this idea has more to do with implementation, and I am open to seeing alliterative proposals to the composition of BASC (and the role of functionaries in this). I would also like BASC to have an option to solicit comments from the community with regards to open ban appeal cases. - Mailer Diablo 03:28, 16 October 2014 (UTC)
• Strong support in principle, oppose as currently proposed I think it's a very good idea to hand over some of these duties to the community, thanks to Beeblebrox for doing the work and proposing it. As currently written I have the following concerns/issues (note that I haven't looked at the thread on func-en yet):

1. It's not made clear whether the community appointees and functionaries will be able to use their CU rights for general use (as we are on AUSC now) as opposed to only to investigate appeals.
2. The appointees are not given OS so are expected to follow blindly what those with OS can view in suppression logs. Counter productive when one of the issues which lead to the block/ban (as far as the blocked user thinks) is that the people deciding either didn't have access to didn't look at all the evidence
3. Given the backlogs already facing the current CU/OS team (SPI, OTRS, UTRS, ACC) I think they/we are already worked pretty hard, losing them to this might cause problems (plus issues with my #1 above).
4. Rather than having strict 3 functionaries to 3 community members I think it'd be better to have a minimum of each, substituting (for example) a functionary for a community wouldn't be a problem.
5. Given oversight blocks are usually discussed on OS-l the appointees will likely need access to that even if they don't have the OS right.
6. I'm opposed to recused members still being able to present evidence on the list unless all parties to a block/ban are also kept completely informed as the recused member is (since they're on the mailing list) and are able to present evidence (I'm not on BASC and haven't had a block looked at by BASC so I'm not sure).

Callanecc (talk • contribs • logs) 04:18, 16 October 2014 (UTC)

(On point 4) There is no 3:3 split proposed. The split as laid out in the proposal is 4:3. And I agree with you on that being a potential source of problems - the ArbCom posse should have to convince at least one community member to enforce its prior consensus, otherwise it's not worth tying up community members' hours in the process. Samsara (FA • FP) 15:13, 20 October 2014 (UTC)

OS-l is a ghost town and there has only been two, maybe three, OS blocks. If my memory suits me correctly they have all been discussed on functionalist-l anyways. --Guerillero | My Talk 04:28, 16 October 2014 (UTC)

Thanks, I've struck that one. Callanecc (talk • contribs • logs) 04:39, 16 October 2014 (UTC)

• Strong support in principle though I think this needs a bit of refinement. Thanks to Beeblebrox for taking initiative and starting this. --Rschen7754 04:50, 16 October 2014 (UTC)

It's not fair to Beebs to say he's not discussed it with anyone, there's been vague rumblings on his ideas which he's made available to the Arbs for months. At the same time, most of us have been so busy that it's not been the focus of many discussions. Personally, I absolutely Support the removal of BASC from Arbcom. I would like to see Arbcom wind down as many responsibilities as possible. The important factors, in my opinion:

• There is an off-wiki committee which can be a final port of call for block/ban appeals. This takes it away from any "pile-on" effect and genuinely gives people a solid chance to put across their view.

  There is a need for a place to review checkuser/oversight blocks.

  There is a need for a place to review blocks which include private information

As long as there's one place to look at those 3, I'd be happy. That might be a beefing up of UTRS, or spinning out of BASC from Arbcom. The way Arbcom works, committee members pay attention to what interests them. In that case, if no one is interested in an areas, say BASC, it languishes. Worm^TT(talk) 07:52, 16 October 2014 (UTC)

• strong support per WTT and HJ Mitchell. There is a need for unbundling of tasks from the Arbitration Committee so they can give their time to arbitrating disputes in a timely manner, and this seems like a sane way of doing it. We should not let perfect be the enemy of the good, so while there might be detail to be worked out we can do that after agreeing the principle. Some detail I think should be considered:
  • I think the AUSC should be formed of between 6 and 12 members, between 1 and 3 of whom are arbitrators, between 1 and 3 are functionaries and at least 4 are identified admins. This would allow space for recusals, changes of status and temporary inactivity during a term without damaging or paralysing the committee (too many or too few members).
  • Each individual appeal must be discussed by at least 4 (5?) members of the AUSC, including at least 1 arbitrator or functionary.
  • There must be a defined process for what happens in the event the AUSC is deadlocked.
  • Having terms run from circa February would probably work best with arb elections in December.
  • Members should have access to OS and CU for the duration of their term, but would not be allowed to use them outside AUSC without already having or subsequently gaining access in another way.
  • The appointment procedure should be lightweight, probably self-nomination with community comments on those nominations guiding appointment by arbcom (initially) and appointment by arbcom with advice (public or private) from the existing AUSC members (2nd term onwards). Jimbo should have a right of veto for exceptional circumstances if he wants it.
  • There should be a public statistical log of the committee's activity (number of cases heard, number of appeals accepted, number declined completely, number declined partially (e.g. block length reduced), average time taken, etc.) updated at least three times per year.

  This is just detail though and should not be taken as opposition to the proposal as a whole. Thryduulf (talk) 12:52, 16 October 2014 (UTC)

  Risker asked me on my talk page why I think "we need a 7-person committee to do what Worm That Turned said on the talk page is essentially being done by one person now."[1]. However, as WTT replied on that page and noted in the introduction, it isn't just him doing the work and it isn't being managed adequately or in a timely manner ("current turnaround time for ban appeals is 3-5 weeks"). You will also note that I explained that 7 people are not needed for each appeal (just 4-5), but that having more people "would allow space for recusals, changes of status and temporary inactivity" while still enabling the committee to function better than it is doing at present. Your (Risker's) opposition seems to stem largely from not wanting you (or other functionaries) to take on the workload - but as has been explained you don't have to all it needs is 1-3 volunteers from the functionaries. Reducing the arbitrator's workload is needed, and this is something they can divest safely. Thryduulf (talk) 12:52, 16 October 2014 (UTC)

• Non-vote comments on everything, for the moment...
  1. The idea of spinning BASC off from Arbcom is a good one, and certainly that team will need to be staffed by people who have access to all the relevant information, which would mean a minimum of one arb, one checkuser, one oversighter, and the rest admins. However, this proposal as written gives what seems to be a disproportionate amount of weight to the "functionary" hat, which seems likely to collapse under that weight. Yes, if BASC is going to be reviewing CU and OS blocks, it needs to have access to that information, but other than in those cases, being a functionary gives one no particular standing in the realm of ban appeals. Unless the intention is to create "handles ban appeals" as another functionary-subsumed role, you could get the same effect by staffing BASC with an arb (who will already have CU/OS/arbwiki access) and then just assigning those rights to whoever else is appointed to BASC if they don't have them, the same way we do with AUSC.

     Reserving half the committee's slots for "pre-made" functionaries requires you to draw from an extremely limited pool of people, the bulk of whom all already wear other committee hats that eat up their time (AUSC, Arbcom, Ombuds...) or whose time is so limited that they aren't particularly active in the roles they already occupy, and none of whom are ban-review experts simply by virtue of having functionary bits. The functionary population is graying pretty rapidly; AUSC and Arbcom appointments don't bring in a lot of people who aren't already functionaries, Arbcom rarely does CU/OS appointments anymore, and as a result people who were a bare minimum of 18 years old when they became functionaries are now increasingly juggling 9-to-5 work, kids, aging parents, etc. It's just not a pool that's very rich to draw from anymore.

  2. Callanecc was actually right when he noted that BASC members would need access to Oversight-l; if we're giving them OS access (especially if we plan to also allow them to use their OS bits generally), they need to have access to any discussions that may have taken place with regard to OS blocks or OS-related blocks. That it's not a particularly busy list isn't a reason to roll OS business into functionaries-l, which contains a number of people who don't have OS access and shouldn't be hearing OSed information.
  3. Three years for serial appellants strikes me as too long. As others have pointed out, you really, really don't want appellants to feel like they have nothing to lose, and for people who are perhaps young and/or impulsive, saying "three years" is basically like saying "forever". I would suggest something closer to 18 months - the same way we tend to block vandals for 31 hours so they don't try the same thing at the same time tomorrow, giving appellants something like 18 months off would avoid the yearly rush but would prevent rapid-fire appeals as well.
  4. "All users subject to a long-term community-imposed block or ban are entitled to appeal to the subcommittee." <-- If your goal is to have a manageable workload, I think this is a non-starter. There is no reason BASC should be the first port of call for anyone who's not under an appeal-only-to-Arbcom/BASC sanction or a CU/OS block. Administrators, whether onwiki or on UTRS, are perfectly capable of first-line reviews of pretty much everything else, and letting people jump the "queue" to go right to Arbcom without using the community process would be forcing a small group to do the work of a large one for no particular reason.
  5. Logging of results onwiki: I think this is something Arbcom should have been doing the whole time; not only would it make it clear to non-arbs what their workload is, but unless there is a reason for secrecy of appeals (and surely in some cases, there is) there is no reason for Arbcom to be keeping pass/fail type information away from the community. Risker makes valid points about attention-seekers and abusive usernames, but these could pretty easily be replaced in logs with "[redacted username] - appeal denied" or whatever, at BASC's discretion.
  6. On the proposing of this proposal: I disagree with Risker's comments that putting this forth to the community first was a bad thing; the community has been begging Arbcom for years to increase their transparency, and I can't see it as anything but a good thing that rather than passing something like this in camera and then informing the rest of us of the done deal, Beeblebrox's first action here was to check with the community. Yes, he could have polled the functionaries first to find out if we'd all just run screaming at the thought of more work, but we can run screaming just as well here as we could on functionaries-l, and this way others can have their say as well.
  7. Though I understand the logistical reasons that Arbcom would need to appoint members of this committee (handing out CU/OS requires this, I think?), I can't shake the idea that BASC (and AUSC, for that matter) shouldn't be serving at Arbcom's pleasure. Have arbcom liaison(s) on the team, absolutely, but if we're going to have a committee reviewing ban appeals on behalf of the community that placed the bans, the community should, at least in a perfect world, be the ones picking the people on that committee. "People who arbcom thinks are suited to ban appeals" and "people who the community thinks are suited to ban appeals" may be intersecting sets, but I doubt they're identical ones. And just because Arbcom has controlled ban appeals doesn't mean there's any particular reason they need to continue controlling ban appeals.

     Phew, ok, I think that's everything floating around my brain for the moment. Sorry for the length. A fluffernutter is a sandwich! (talk) 13:31, 16 October 2014 (UTC)

• Support - as a step in the right direction, however, I'd prefer eventually removing ban appeals from ArbCom altogether. I've served on BASC for a total of 12 months, some of which was before UTRS was up and running, and the workload wasn't excessive. To be honest, most of the appeals from those who just had their talk page access revoked were unpersuasive, and it was those who hadn't appealed for a while who were more reflective and convincing. At the moment {{uw-blocknotalk}} points to UTRS, and so long as that continues, my guess is that UTRS will pick up the majority of the 'just had their talk page access revoked' cases. Otherwise, yes, tracking the ban appeals was a time consuming chore. At the time, I favored migrating the whole thing to a software package that would track the cases in a similar manner to OTRS, and hopefully simplify producing reports. Anyway, while the method of producing the reports could certainly be streamlined, and I'd prefer going further than this proposal, and removing ban appeals from ArbCom altogether, I nonetheless support. PhilKnight (talk) 17:49, 16 October 2014 (UTC)

Earlier this year when we were discussing these issues amongst ourselves I somewhat sheepishly admitted to my fellow arbs that I had been referring all sorts of users to use BASC for years. If only I had known... Beeblebrox (talk) 03:39, 17 October 2014 (UTC)

Strongly support PhilKnight's suggestion -- why not just elect a separate BASC group distinct from arbcom to handle all non-arbcom appeals? NE Ent 22:36, 19 October 2014 (UTC)

Thirded. The less overlap, the more likely people will be to get an "independent trial". I can see that these considerations are woven into the current proposal, and it's definitely a step in the right direction, but since we're at this junction, shouldn't we go further? Does ArbCom feel threatened by not having a perceived majority (i.e. combination of membership and appointment) in this subcommittee? If so, why? Starting on a clean slate sounds like an appealing idea to me. Samsara (FA • FP) 14:55, 20 October 2014 (UTC)

• Strong support. I didn't know that Beeb had this up his sleeve but I've seen something like this coming for a long time. It's time to devolve some of the areas of Arbcom to the community but it needs to be done very carefully. I don't have time to comment t in greater detail but I think HJ Mitchell, Worm, and Fluffernutter have best spelled out most of what I would have wanted to say. Kudpung กุดผึ้ง (talk) 13:20, 17 October 2014 (UTC)
• Oppose - I think there are better alternatives. Why not make it easier on everyone, and let people back automatically after their time in purgatory. If they re-offend, simply double the length of the ban, and repeat. This makes the process self-regulating, and saves everyone time. --Iantresman (talk) 20:33, 19 October 2014 (UTC)

• Comment leaning oppose. Iantresman's suggestion seems sensible. Whether successive bans should double or triple or quadruple (etc) the length, I don't know, but the basic idea of exponentially increasing bans has merit in my view. zazpot (talk) 02:07, 26 October 2014 (UTC)

• Comment. My initial reaction is the same as Risker: do away with it except for ArbCom-related appeals. However, I'll wait for the "final" proposal before actually voting. The most apt analogy for BASC I can think of is non-capital habeas. A criminal is convicted in state court. He appeals once as a matter of right and another (in some states) as a matter of discretionary review. After the direct appeals are complete, he may file habeas petitions in state court. Ultimately, if after all this he loses, he can still file a habeas petition in federal court. I believe the odds of prevailing on federal habeas are tiny (I forget the precise percentage but I think it's less than 1%), and, yet, a ton of work is spent on the petitions before they're (almost always) denied by the district court. After that his rights for appeal to the circuit are limited but do exist. How many chances does someone get before there's finality?--Bbb23 (talk) 22:40, 19 October 2014 (UTC)
• Comment, leaning oppose - I think Risker has some good points. The community is capable of reviewing most any block that isn't Arb enforced. The lack of CUs is part of the problem, as they can be helpful for block reviews, but that is a matter of how few Arb let through the system, as they decide who gets to a community vote when it comes to CU. Even creating a hybrid system where non-admin can view and comment, even if they can't vote (as they can't see deleted contribs or be privy to any CU results, at WP:AN (or a new WP:AN Ban Appeal), would be better for these last ditch ban appeals. If there are bigger issues, then THOSE can be kicked up to Arb by the admin reviewing. I can't say I have the perfect system, but doing most of these at a lower level makes sense. The idea outlined in this proposal has problems. It might be better than the existing system, but by how much? Dennis 2¢ 11:40, 20 October 2014 (UTC)
• Oppose in its current form as per my various comments above. Samsara (FA • FP) 15:13, 20 October 2014 (UTC)
• Support the idea and basic structure. This is a relatively high traffic discussion (of interest to the community) where a pretty constrained set of decisions matter. that's a perfect place to have trusted community members do something for the arbs that isn't arbitration. So the arbs can focus a bit more on that without feeling stretched thin. And increasingly (even for non ARB bans), the community has to have some input into the appeal process in a venue that is free from outside drama and at least somewhat efficient. that environment is precious and conducive to making good decisions about bans, but non-arbitrators should be able to act in that mode with the same effect as it is appropriately devolved. Protonk (talk) 16:37, 20 October 2014 (UTC)
• Oppose. This would make a bad system that does need to be reformed even worse. ArbCom exists because the community cannot deal with certain problems and it needs to use its unique capabilities to solve such problems. In case of people who are not trusted to not cause problems who appeal over and over again, the solution is to let these people edit Wikipedia using a new account monitored by ArbCom or ArbCom appointed undercover Admins. You can then have an evaluation of these editors based on their actual recent editing here, instead of never ending debates with those editors based on old assessments. Count Iblis (talk) 18:19, 20 October 2014 (UTC)

There was historically involuntary mentorship as a condition to being allowed back in with supervision, but there is a lack of willing volunteers and this has been largely superseded by conditional unblocks. - Mailer Diablo 15:19, 21 October 2014 (UTC)

• Oppose We already see problem editors return and cause problems when bans are overturned by the current system. I think this will make it easier for it to happen more often. I think there should be no appeal at all. Its hard enough to get someone banned, if they are banned there is good reason. AlbinoFerret (talk) 00:25, 21 October 2014 (UTC)

At one end we have editors who want to grant an automatic right to return after a certain period. On the other end, we have editors who want to let no one back. The question then is, where should the line be drawn and is it possible to reconcile this difference? What would be the best approach to do so? - Mailer Diablo 15:19, 21 October 2014 (UTC)

@Mailer Diablo. There is only a question concerning where the line is drawn, if the matter is subjective. By having automatic returns, and escalating ban times, you remove the question, and have a self-regulating system that increases ban times for the most persistent offenders. --Iantresman (talk) 17:17, 21 October 2014 (UTC)

@AlbinoFerret. I must disagree, having been on the receiving end of a community ban in July 2007. In my case, not one diff was presented to support the allegation. The discussion took just 5 hours and 11 minutes with no due process. My case was given as one of the reasons for deleting the Community sanction noticeboard[2]. Editors who bothered to examine the case, found not "a single shred of evidence against" me, and that the banning Admin "refused to justify the block". I was accused of harassing one editor that eventually left Wikipedia (he temporarily left Wikipedia for different stated reasons), and I was accused of harassing another editor who turned out be the same editor who was using at least 4 sockpuppets abusively.[3] Although I was entitled to a review after 6 months, it was refused for 4 years. On returning, I was able to bring several articles up to Good article status, and have achieved at least another 10,000+ edits. In my opinion, the system is flawed. I don't claim to be blameless, only that I was blame for things I didn't do. --Iantresman (talk) 17:08, 21 October 2014 (UTC)

On the other hand we have persistant edit wars becuase some users have been banned and let back in and the cycle repetes itself again and again. Ban, let back in, start wars, ban, let back in, edit war..... One ban was ended because the person who banned them left Wikipedia. AlbinoFerret (talk) 17:40, 21 October 2014 (UTC)

Escalating bans would ensure that the cycle would not repeat indefinitely, and automatic returns would ensure that Arbom's time is not wasted. --Iantresman (talk) 23:40, 21 October 2014 (UTC)

An automatic returned editor may quickly engage in the offending behavior that results in a reblock, and the editor would be quick to go through the appeals process up to the level of BASC as the new sanction has to be endorsed as a new instance. This would result in the same level or even an increased level of workload as there would be a sense of entitlement. - Mailer Diablo 03:16, 22 October 2014 (UTC)

My suggestion of escalating bans would not have the same appeal process it has now. For example, if a couple of admins can block an editor, then an appeal through a couple of non-involved Admins should be able to determine whether an editor is allowed back. They simply determine whether (a) procedure has been followed correctly (b) the banned editor was culpable (c) whether there were any mitigating circumstances. This has zero workload on Arbcom/BASC. --Iantresman (talk) 08:35, 22 October 2014 (UTC)

• Oppose. Has the BASC ever lifted a block on a subsequent appeal? There are things to address in the block appeal process, but this certainly doesn't seem to be it.TeeTylerToe (talk) 04:09, 21 October 2014 (UTC)
• General support for the idea of delegating this function to a smaller Committee that is not made up of all arbitrators, but I'm not convinced that there needs to be a community membership with the associated administrative overhead. Can't you just delegate this to, say, three arbitrators, at least for the probably 80% or so of relatively easy cases?  Sandstein  08:57, 21 October 2014 (UTC)
• Support Per above. DogbertYDS (talk) 19:00, 21 October 2014 (UTC)
• Support in principle. Not married to the current details. I hope any final proposal will have a new discussion/straw poll. Skyerise (talk) 19:32, 21 October 2014 (UTC)
• Support - Arbitration of cases takes too long. Off-loading of work from the arbitrators is a good idea. Robert McClenon (talk) 22:08, 21 October 2014 (UTC)
• Support. Arbitrators will be able to focus solely on Arbitration (which is what they should be spending their time doing) and BASC members will be exclusively focused on ban and block appeals. Since members of said committees will be able to focus on one job, there might be time for more detailed consideration to be given to each case, rather than looking at an ArbCom case with one eye and a ban appeal with another eye, so to speak. (It's only natural that this proposal should be good, because Beeblebrox, the creator of it, has two heads. Naturally, people with two heads have two brains instead of only one, which means that Beeblebrox can probably think of ideas two times faster than the average person. ) --Biblioworm 23:37, 21 October 2014 (UTC)
• Question Can you clarify the format of the "advisory processes" in concrete terms? Samsara (FA • FP) 02:31, 22 October 2014 (UTC)
• Support Firstly, since the appeal process appears to primarily exist to limit the 'nothing to lose' mentality, the project should allocate resources to it efficiently. Secondly, since the probability of any positive outcome for the project from an appeal is negligible, the probability of loss due to delegating the decision process is negligible. Lastly, members of a specialist appeals committee would, in all likelihood, be better at reviewing appeals than the Arbcom generalist. Doug (talk) 19:46, 22 October 2014 (UTC)
• Support - Devolution of ArbCom authority is a good thing. Creation of a specialized body that can reexamine cases with fresh eyes is a good thing. Carrite (talk) 06:59, 23 October 2014 (UTC)
• Support - I'm actually liking this idea! - Arbcom can focus on Arbitration, BASC can focus on bans & whatnot. –Davey2010 • (talk) 14:07, 23 October 2014 (UTC)
• Support per HJ Mitchell. This seems to be a very good idea. This gives the community more influence in these situations and helps Arb Com by removing something that takes up way too much of their time.--Mark Miller (talk) 02:04, 25 October 2014 (UTC)
• Oppose per Risker. Actually, given that this RfC is entitled "BASC Reform", it would be good if Risker's proposal could be prominently floated in a separate section so it can be commented on (sufficiently in advance before the Reform RfC comes to an end). Of course, it's very well and good that a a proposal for a change was made by Beeblebrox, but the proposal to change the way people are appointed is putting the cart before the horse. Users on this sub-committee spend/kill a lot of time reviewing appeals, but how much of a positive benefit has it brought to the community? How much of it is simply negative to the community? I think those questions should be asked and answered in order for the community to assess how best to restructure/reform. Ncmvocalist (talk) 10:52, 25 October 2014 (UTC)

I am personally of the opinion that BASC provides substantial benefit to the communuty. However, it is largely similar to the benefit the oversight team provides in that it is largely invisible to most users. Because of the way the subcommittee operated, all the broader community is aware of currently are how many users actually get their sanctions lifted. I think the subcommittee probably should at least publish a record of how many other appeals it hears and declines as that would give everyone a better idea of the total workload involved.

In short, BASC keeps the worst of the worst from wasting more of the community's time with endless appeals that will not succeed, and they are able to have in-depth private conversations in the hopes of finding a way to allow back in users who are not the worst of the worst. It is an important task, and that is the problem. Having it done by a small group of arbs, who have other business that only they can take care of, seems to have a negative impact on the way the process operates, causing it to move slowly and sometimes becoem a bit confusing. The hope is that having volunteers from the broader community, for whom this will be the only thing that they must handle among their own very small group, the process will function more efficiently. Beeblebrox (talk) 15:43, 25 October 2014 (UTC)

Could you please arrange for (or do the needful) to have the stats to-date published for the purposes of this discussion then? In particular, how many users appealed to BASC, how many were declined, and how many were granted altogether each year (since BASC was 'created')? It would also be helpful if perhaps a clerk or other interested person who can spare some on-wiki time to review the relevant on-wiki archives to see how many were referred for the community to reconsider and what the outcome of those were (eg; whether the ban was kept in place or revoked). Additionally, of the number granted (be it directly by BASC or by the community), how many were reblocked or rebanned subsequently. I appreciate what you've said, and it better informs people about the process and the intention. But as always, there are unintended consequences which sometimes flow from any reform also. Often, a reform involves taking a risk and/or hypothesising on whether the current state of affairs or a reform will be an improvement - and seeing how it goes, as we can't see into the future. Here though, I think the stats I've suggested would help clarify the work involved and the type of numbers that need to be appointed to this body assuming it remain in existence (and/or whether a type of reform proposal is possibly needed in relation to the banning policy if it just comes down to an issue of endless appeals). Ncmvocalist (talk) 17:01, 25 October 2014 (UTC)

There was some discussion of stats on the talk page here, it turns out nobody has ever kept a record of the number of denied appeals. WormThatTurned did come up with the following from this year:

• April - 19 cases, 3 unblock, 16 declined.
• May - 11 cases, 2 unblock, 9 declined
• June - 12 cases, 12 declined
• July - 4 cases, 4 declined
• August - 12 cases, 1 unblocked, 11 declined
• September - 5 cases, 3 declined, 2 offers of unblock with no response
• October - 14 cases, 1 unblocked, 13 declined.

I'm pretty sure numbers were higher during the first few months as we were bombarded by serial appellants during that period. Beeblebrox (talk) 18:54, 25 October 2014 (UTC)

• Support Good idea. Arbs can use their time more usefully now. ^{My name is}not^dave (talk/contribs) 10:34, 26 October 2014 (UTC)
• Support. I think this is a great idea; it frees up arbs to concentrate on arbing. On a secondary point, it deflects any possible criticism that arbcom has become a closed loop (ie. that those who feel wronged by arbcom can only have their appeal heard by arbitrators). bobrayner (talk) 12:06, 26 October 2014 (UTC)
• Oppose. As longtime arbitrator Newyorkbrad has pointed out multiple times, ArbCom currently handles far fewer arbitration cases than it used to. If a lot of ban appeals are removed from ArbCom, as per this suggestion, their workload will be further reduced. This may not be a bad idea per se, but it should be combined with an appropriate reduction of number of arbitrators, for instance to 10. (My own suggestion would be a committee of 10, with each new arbitration case set from the outset with 7 arbitrators and a minimum of 5 needed to end the case). The new sub committee should not get more new members than the number the main committee is reduced with. At a time when overall activity level at Wikipedia is rather down than up, we should not pursue a path with a steadily increase of off-site buraucracy. Iselilja (talk) 13:31, 26 October 2014 (UTC)
• Support I see this proposal as an attempt to remove functionary duties from Arbcom and delegate those to a subgroup which need not deal with any private Arbcom information, but who can manage the outcomes of decisions which already are very discussed by the community. I would prefer that Arbcom address disputes which have broader significance to the community than to individuals, and this community seems like a way to retain the Arbcom branch of Wikipedia as an appeals option while freeing up Arbcom time to focus on community issues which have the greatest impact. I support this proposal more as a way to free Arbcom time, but I also think this will bring greater justice to the appeals process because I expect that people who would take appointment on this committee would like to deal with appeals more than typical Arbcom members. Blue Rasberry (talk) 13:42, 29 October 2014 (UTC)
• Support I had suggested something similar a while ago, but never got around to make a full fledged proposal. I am glad to see that this is now moving on. In response to Risker's argument that BASC should rather be disbanded altogether, I am afraid that a committee is the best way to put an end to intractable appeals, it would not be a net positive to have such endless debates reappear on our noticeboards. In fact, I would go further in the mandate afforded to this new committee, extending it to cover all community sanctions (such as topic bans). Now, the matter of its composition ? The structure proposed by Beeblebrox seems adequate for the time being, although in the long run I would prefer that all members be community elected. If the concern is that the pool of candidates is too small, this is already mitigated by the fact that users may stand in both ArbCom and 'AppCom', and if necessary we can have arbs filling missing seats. This would not be an issue in the long term though, see infra.

Something similar should be done with AUSC, there must be a dedicated committee separate from ArbCom with final say on CU/OS matters. I had suggested this in the past but arbs were reluctant to the idea, arguing that this was the 'fiduciary duty' of ArbCom as mandated by the wmf privacy policy. In fact this is not at all the case, the policy makes possible for local arbitration committees matching certain conditions to take upon this duty, but this is in no way a requirement, and indeed several wikis have arbitration committees matching these conditions but without any of those responsibilities. Here again, I would extend the mandate of this 'AudCom', specifically to misconduct by administrators outside formal arbitration cases. Any case overlapping between the three different committees would be resolved by a joint subcommittee.

All in all, I think it is a step in the good direction to 'decentralize' ArbCom, and in fact return it to its more basic role. It is true that there has been lots of progress in the recent years and the threat of an 'all powerful' arbcom has diminished, but it did not disappear altogether and introducing concurrent committees affected to different duties would make this threat much more remote. In addition, these committees would have a more focused role, so would be more efficient, the members wouldn't have to be proficient in so many areas and would have less work, meaning a less demanding job, so more attractive, and the pool of candidates would increase as a result. Cenarium (talk) 18:44, 29 October 2014 (UTC)

I'm going to resume my thoughts in a condensed manner. A large committee with a vast number of responsibilities is bound to be inefficient, the solution is to divide it in several committees with more focused responsibilities. Of course, I'm not advocating the extreme opposite which is just as inefficient, but a middle ground such as described above seems like the most efficient solution, not only for Wikipedia but for the committee members. Cenarium (talk) 18:59, 29 October 2014 (UTC)

• Strong support – Per HJ Mitchell. Very well thought-out proposal, and an excellent way to solve this problem. RGloucester — ☎ 02:19, 31 October 2014 (UTC)
• Support Long overdue.
• Query 1 What happens if a member of the BASC is one of the editors who was in dispute with the appealing editor?
• Query 2 If, as Risker says above, this is a power the committee took unto itself in 2007, would it be better legitimised by a broad-based community consensus?

  All the best: Rich Farmbrough, 01:44, 6 November 2014 (UTC).
  

  Sugggestion replace issuance with issuing. All the best: Rich Farmbrough, 02:05, 6 November 2014 (UTC).
  

• Cautious support - I would hate to have this backfire and see the return of various troublemakers whose bans are richly deserved. I agree with Tryptofish that community bans should remain the responsibility of the community. If the purpose is to allow ArbCom members more time to judge policy matters and contemplate editor actions, and this proposal allows that, then this is a good idea. I strongly disagree that bans should be of fixed duration. A ban should always be indefinite, as I see it. Jusdafax 05:04, 19 November 2014 (UTC)
• Support in principle: However it happens, BASC should be separated from ArbCom's list of duties. This single change will lead to the biggest improvement in how ArbCom functions. As for the details, keep it simple and go with what works. Processes can be adjusted and mistakes can be fixed. Ignocrates (talk) 20:57, 19 November 2014 (UTC)

• This page is probably "stale" by now, but I'll ask anyway...If the community chooses to actually ban an editor. this committee can over-rule the community? I thought only arbcom acting as a committee of the whole is empowered to do such things. And with that in mind, only one arbcom member will be on this committee? There should be at least 3 to give at least the semblance of acting as a committee? maybe there should just be elections for this committee as a doppelganger to arbcom? Otherwise, you'll please pardon me, because this really does "feel" well-meant, but this sounds rather close to a star-chamber of sorts. Which seems anathema/unwiki to me. - jc37 07:38, 13 January 2015 (UTC)

Proposed new policy regarding repeat/serial appellants

Initial appeal

All users subject to a long-term community-imposed block or ban are entitled to appeal to the subcommittee. Users who still have access to their personal talk page may still appeal to the subcommittee if they wish, providing they have attempted appeal at least once via the normal process and at least once by the unblock ticket response system.

However, all appellants should carefully consider how they wish to frame their appeal as the subcommittee is intended as an absolutely final, binding review of the original sanction. Only one appeal based on the merits of the initial sanction will be heard by the subcommittee. Should the subcommittee find that the sanction was valid, that aspect of the case will be considered permanently resolved.

If an appeal is denied

Once one appeal has been rejected by the subcommittee, the block or ban will be considered valid and upheld and may not be overturned by other means. There is no higher avenue of appeal than the subcommittee and a user should only be appealing to it if the community has already rejected previous appeals and/or the user has had their talk page access revoked. Attempting to request unblock by emailing individual admins or asking a proxy to open a community discussion is not acceptable once an appeal to BASC has been denied. This is to prevent blocked or banned users from taking up more of the community's time and resources than they already have. While everyone is entitled to a fair hearing, there is no right to infinite public appeals.

Subsequent appeals

Further appeals will therefore be required to address a way forward as opposed to seeking to simply overturn the block or ban as by this point the sanction has already been reviewed and upheld as valid. Users will be given one warning if they fail to do this, and after that any such communication from that user may be ignored. If the user makes multiple appeals after being advised that their appeal has been rejected, the mailing list moderator may configure the list to reject their communication for a period of up to one year.

Use of other tactics

Users who attempt to make back-room bargains or to blackmail the subcommittee in any way (threatening to regularly evade the ban, to "out" arbitrators or other users, campaigns of email or other off-wiki harassment, etc) will be given one warning that such tactics are not acceptable and will not lead to a removal of sanctions. If they attempt this strategy again, the mailing list moderator may configure the list to reject their communications for a period of up to one year.

Serial appellants

It is a sort of tradition that every year when new subcommittee members begin their terms, many sanctioned users will "try their luck" with the new members, swamping the new committee with requests that have very little chance of being approved, often ignoring previously established restrictions on how soon they may appeal again. There are also users who do respect the time frame restrictions but as soon as the prescribed period has elapsed they will simply send a request to have their restrictions lifted without indicating that they acknowledge their past errors or would do anything different if their sanctions were lifted. Neither of these tactics is acceptable.

Therefore, any users who have appealed more than five times without presenting a compelling case for lifting their restrictions may be limited to one appeal every three years, and the mailing list moderator may configure the list to reject their communications for the entire three-year-period.

Even if a user's communications are rejected there is a log of what was sent and that log may be consulted as evidence during the user's next permitted appeal.

Discussion of Proposed new policy regarding repeat/serial appellants

• I think that establishing a maximum of 2 appeals per year/1 appeal every six months is a more feasible solution. → Call me Hahc21 20:37, 15 October 2014 (UTC)

That's essentially the system we have now. For most cases it is sufficient, but there is a small group of long-term banned users that endlessly appeal without ever making a compelling case. These changes would only effect that small minority, preventing them from wasting the subcommittee's time with repeated, pointless appeals. Beeblebrox (talk) 01:09, 16 October 2014 (UTC)

• I think that this is a pretty serious rewrite of the unblocking provisions of the blocking policy, without having been discussed at that policy. Risker (talk) 02:03, 16 October 2014 (UTC)
• I wouldn't want this to be set in stone - the worst thing that can happen is when banned users start to think that they have "nothing to lose". I think it would be okay if this discretion is left to BASC to decide what would be the best approach. - Mailer Diablo 03:41, 16 October 2014 (UTC)
• Possible alternative option. I think BASC should be given an option to seek the community's input on a particular ban appeal. This can be done in two ways - either the appellant indicates that he is willing to give BASC this option (in which his appeal will be viewed more favorably) or that BASC has an automatic right, if exercised, to make calls for input from the community with regards to a ban appeal. The latter has a powerful deterrent effect of repeat/serial appellants because over time it will become very apparent who are the repeat/serial appellants and would think very carefully before submitting an appeal. Also when input from the community is sought, there is the question of whether the comments should be on-wiki or emailed to the mailing list to be viewed by BASC only. - Mailer Diablo 03:55, 16 October 2014 (UTC)
• Some comments:

1. I absolutely agree with Mailer Diablo regarding recording a log of appeals publicly. As well as the transparency issue which has been with ArbCom for ages it also makes good sense that the username and outcome (could be as simple as Callanecc - sanction valid, still needed - appeal declined) will be publicly recorded considering they can only appeal to BASC (or to Jimbo) in the future
2. The If an appeal is denied section will need more discussion as it changes the unblock policy.
3. I would have no objection to the Use of other tactics section being a longer than one year.
4. Parties involved in the original block (including original blocking admin and admins who have declined appeals) should be notified that an appeal has been received and be give
5. In the Serial appellants section I think five is too many, that's 4 years of them trying their luck. Just tell them that they can't appeal for two-three months either side of a year so that they can't appeal as soon as the new BASC is sitting. If they ignore that rule give them a warning then either enforce the time limit (by having the list reject their email) or apply the three year limit.

I don't have any other issues with it at this stage, though I think Mailer Diablo's option should be incorportated, with the community being consulted about appeals whenever it is possible. Callanecc (talk • contribs • logs) 04:33, 16 October 2014 (UTC)

• One of the reasons that turned-down block appeals were not published was that it was likely to increase the activity from the more attention-seeking types, and would also have a negative impact on those few accounts where there was a belief that another six months or a year would see sufficient change in attitude or increased maturity to consider unblock at that time. Don't forget we have banned users with literally thousands of socks, many of them never tagged. Don't make this a hard and fast rule, or you'll get the abusively named accounts asking for BASC unblock, and they'll be forced to publish the names. Risker (talk) 04:45, 16 October 2014 (UTC)

It does need to be fine tuned a bit, perhaps only publishing the name of the primary account but not if the appeal was out of process and nothing was done (ie if they waited 6 months not a year but weren't restricted further) as well as abusive usernames. I don't see the problem of recording 'declined, but six months will likely do it' since the user is being told that anyway, and might make them consider that it's more likely they will be unblocked in the six months given it's public not a statement only to them. Callanecc (talk • contribs • logs) 05:41, 16 October 2014 (UTC)

• I am thinking only three outcomes and one status should be indicated: "accepted -(state conditions)", "declined", "ineligible", and "processing". Let's keep it simple. Depending on what BASC thinks, it may be okay to omit ineligible appeals altogether from such a list. - Mailer Diablo 06:49, 16 October 2014 (UTC)

I'm not sure I like the idea of publishing the results of each individual appeal. In addition to the concern that it will encourage some banned users (and it will, the exact users these changes are intended to discourage, the worst of the worst) it will also "name and shame" a user every time they have an appeal rejected. I realize we do this on their talk page already with regular block appeals, but we don't do it on a centralized noticeboard that preserves it forever. Given the fact that the list of banned users was just deleted I have the impression the community wishes to move away from that sort of thing.

On the other hand there is nothing stopping BASC from publishing monthly statistics stating how many appeals they declined, how many were rejected as invalid due to being premature, and of course which appeals were accepted. (the only thing BASC currently publishes) Beeblebrox (talk) 04:14, 18 October 2014 (UTC)

As to the proposal for publishing the requesrts, I would say the following is reasonable:

1. The first appeal by any user, unless under a problematic user name, must be published. In case of sockpuppetry, only the first one by any of the accounts operated by the user in question qualifies for this.
2. Any appeal which is accepted should be published, along with any editing restrictions the user is under.
3. BASC will decide, on a case-by-case basis, whether or not to publish any other appeals.

What bothers me more is the issue of having thelisty reject emails - specificly, we need some decent safeguard against both of the following abuse methods:

• Some user, who wants to be sure a specific banned user can never appeal, makes several appeals from many different free email addresses; then, when the real user decides to be allowed to edit again, the legitimate request will probbaly be rejected.
• Some user decides to try their luck by making the request from other, free email addresses, and (once they've been rejected) subsequently pretend (s)he knows nothing about them.

Note that we can't require these users to use their original email address - they may no longer have access to it. עוד מישהו Od Mishehu 10:53, 19 October 2014 (UTC)

• In principle I support limiting appeals but I think this proposal is still too lenient. For any Wikipedian to be banned, there had already been numerous arguments across the wiki resulting in repeated visits to the noticeboards. If someone is that much of a troublemaker, I'd give them one appeal only and then a lifetime ban after that. We have good-faith contributors writing an encyclopedia and yet we're wasting time on ne'er-do-wells that insist on making this a battlefield. How many contributors have quit in the time ARBCOM extends to the usual suspects? Chris Troutman (talk) 23:06, 20 October 2014 (UTC)
• This more or less describes how I imagine it works, or should work, now, so no objections on the merits. But in view of WP:CREEP, is it necessary to codify all of this as policy?  Sandstein  09:00, 21 October 2014 (UTC)
• For those making repeated frivolous appeals, there should be exponential back off. Six months, one year, two years, four years, and so on. This will discourage time wasting. There should be a means for appellants to informally speak with somebody before an appeal is filed so they can get feedback and avoid filing frivolous appeals. Jehochman ^Talk 10:43, 26 October 2014 (UTC)
  • I strongly support the second part of your suggestion. I have had no dealings with BASC, but Arbcom is generally a black hole as far as email communication is concerned. Dialogue is far better than some kind of rationed call-and-response system. All the best: Rich Farmbrough, 01:49, 6 November 2014 (UTC).
    
• I support Mailer Diablo. I think we should report all cases, included declined, to BASC talk page (private information can be removed). For "one appeal every three years" rule, I think it should only be used for users with active evasion or harassment to BASC.--GZWDer (talk) 14:38, 4 November 2014 (UTC)
• Query when you say "many serial appellants" are we talking 50, 1000, more? All the best: Rich Farmbrough, 01:49, 6 November 2014 (UTC).
  

It is actually probably only a few dozen people. That is exactly the problem, this small group manages to take up a disproportionate amount of the subcommitee's time with repeated requests. Because the subcommittee does take each appeal seriously they do review these appeals even though they often conclude that they are totally without merit. This is a change that would only be perceptible to the worst of the worst banned users, the small group who have basically no chance of being allowed to edit again. These are usually also the most stubborn and persistent people, which is part and parcel of why they were banned to begin with. Beeblebrox (talk) 00:20, 7 November 2014 (UTC)

That got me thinking. I realize that this is a sensitive point, and my opinion is hampered by my lack of seeing things from the inside, but perhaps there needs to be some re-titration of "Because the subcommittee does take each appeal seriously they do review these appeals even though they often conclude that they are totally without merit." I'm sure it gets to where the meritless appeals are obviously meritless from the get-go. Perhaps there is some room for, in a thoughtful way, engaging progressively less exhaustively with appeals as they become increasingly obviously meritless. This does not have to be all-or-none. It's possible to move a little closer to a response of "sorry, no, the subcommittee does not feel that this information changes our decision", based on past experience with the appellant and on a simple reading of the appeal, without feeling the need to go into a meticulous examination every time. Given that these appeals present such a strain on the workload, this may be a justifiable change in practice, without really sacrificing the taking of the overall process seriously. I have the impression that, for better or worse, US Supreme Court justices say "no" to death sentence appeals more quickly that ArbCom says anything to trolls who are only asking to edit a website. --Tryptofish (talk) 00:38, 7 November 2014 (UTC)

Discussion of how this affects other policies

This was brought up above and I think it is an important point that should be discussed.

• To clarify, here is what can happen now:

1:A user is blocked

2:They may appeal anywhere they like, their talk page, UTRS, BASC, or even by proxy at a noticeboard

3:If those reviewing their appeals find them without merit and decline to unblock, they may move on to any of the other forums at will

4:Eventually, (usually after about three on-wiki appeals, but there is no actual rule) they begin to get restricted by having their talk page access removed and so forth

5:They often then appeal again to UTRS

6:And then to BASC

7:Usually by this point the user has appealed their block about five times

8:BASC will tell them they can appeal to them again in six months, but if they email the right admin they can get another on-wiki review posted by proxy or even get their talk page unblocked so they can start up again that very same day

• Here is what would change:

Steps 1-7 = no change

8:BASC will tell them they can appeal in six months. They are not to appeal their block in any way to anyone for a period of six months as they have already had multiple appeals denied.

9:When six months have elapsed they may appeal again, but their new appeal must focus on what would be different if they returned. If they are still trying to argue that the original sanction was invalid and should be overturned (as opposed to being lifted to give them another chance) it will be summarily rejected.

• So, yeah, a slight change to the appeals process, not just for the benefit of BASC but for everyone involved in block reviews in any forum or format. People who cannot accept that their behavior is unsuitable for a collaborative project are one of the biggest drains on our volunteer resources. Endless "last chances" are not the right approach to curbing this problem. Beeblebrox (talk) 17:36, 28 October 2014 (UTC)

This is how innocent men get executed in the United States. New evidence comes to light, but they have exhausted their appeals on matters of fact, so regardless of the strength of the case, leave to appeal is denied.

While I understand Jclemens' edict "It is not about justice or fairness." it remains the case that a dysfunctional appeal system leads to a socially dysfunctional community.

So I prefer the current system, for normal community blocks and bans. It is unlikely that significant on-wiki disruption will occur from use of the {{Unblock}} template - admins are (usually) patient with good-faith requests, and will deny talk page access where needed.

All the best: Rich Farmbrough, 01:37, 6 November 2014 (UTC).

Once one appeal has been rejected by the subcommittee, the block or ban will be considered valid and upheld and may not be overturned by other means totally disagree with this. The community is the ultimate authority. The committee was implemented for cases where the community cannot reach a conclusion AND leaving the matter undecided is harmful. Having a method for allowing people to come back in is relatively harmless, sine if they misbehave they will be banned again. But giving functionaries, however capable, well intentioned and diligent, the right to exclude over the wishes of the community is a bad idea. We have already seen this power abused by check-users. All the best: Rich Farmbrough, 02:15, 6 November 2014 (UTC).

As I believe I have mentioned to you before, I find that comparing WP processes to real-world courts of law is seldom useful as they are manifestly not the same thing. I struggle to imagine a scenario where new evidence has ever come to light on the fifth or sixth review of a Wikipedia block. Beeblebrox (talk) 00:24, 7 November 2014 (UTC)

Initial implementation

Arbcom will develop a streamlined selection process (similar to that now used for AUSC) and select the inaugural members, who will then serve until the 2015 AUSC selection process, which will now run concurrently with the BASC selection process. In the event that the arbitrator member of the subcommittee is not re-elected at the end of the year a new volunteer will be sought from the ranks of the new committee even if it has not yet been six months.

Discussion of Initial Implementation

• I'm a little hesitant with the 'we'll select some people' approach but I can see the benefit in having selected canaries for the start. I'd also prefer that a body such as this would have more of an 'election', but as we've seen with AUSC over the past few years that doesn't seems to happen as well as one would hope. Callanecc (talk • contribs • logs) 04:37, 16 October 2014 (UTC)
  • At least as a bootstrap, people selected by ArbCom seems a sensible way to get an initial committee off the ground with minimal fuss and drama. If this is done by next year's committee then approximately half the committee will have been elected with this as part of their remit, meaning the appointees will have been (partly) democratically indirectly elected. For subsequent committees there will, depending on the detail of the proposal be at least one arbitrator who will not be elected to the committee, and at least one person elected or selected from the limited pool of functionaries. A sensible way for the balance of the committee (about 4-10 people I think, depending on proposal) would be elections with people co-opted by the committee if insufficient people are elected. Thryduulf (talk) 00:49, 18 October 2014 (UTC)
    • Yes, that's one way forward. Another might be to hold community elections for BASC on the same ballot as the ArbCom ones. The detail would need working out (like preventing the same candidate being appointed to both, for example).  Roger Davies ^talk 02:06, 18 October 2014 (UTC)
      • That is pretty interesting. I assume this voted for BASC could look at any ban, including Arb bans? I also assume they must have the bit, and identify with the foundation as to get CU access? Would all review all cases, or (for instance) having 9 BASCers, and 3 (or more if they request for a particular case) review each ban? Dennis 2¢ 15:57, 21 October 2014 (UTC)
• This was discussed on the talk page, but I'll put it out here. To alleviate issues relating to slow responses and the lack of consensus in mailing lists, BASC could adopt a voting system (possibly, the BASC wiki) that is not different from the way Arbcom votes for a motion. This makes (1) the process formalized that would make apparent who has not yet voted/given an opinion in a given case and (2) minimizes any possible dispute or ambiguity over a member's decision. For a seven-member BASC, four votes would make a majority clear and 24 hours after the majority is formed the case can be acted upon. Clear-cut cases should not require every member to be present to make the decision (though it is preferable that everyone has given their say). - Mailer Diablo 08:09, 18 October 2014 (UTC)
  • I agree with Mailer Diablo. I hope that editors here will take a look at the talk page section about "Statistics", where this discussion has been taking place. If, in fact, the biggest factor in making the BASC workload too high has been the lack of a way to wrap discussions up expeditiously, then it may be better to try fixing that, before trying to implement the ideas proposed here. (And if there are other factors that impede the BASC workflow, that have not yet been identified on the talk page, it would be very helpful to point those out.) --Tryptofish (talk) 19:08, 18 October 2014 (UTC)

Would Arbcom be interested in UTRS for BASC?

I'm not sure if you are interested, but what if a new userright could be created just for BASC where appeals could be forwarded directly to BASC and hidden from other admins on UTRS? Would it help manage the process? Much of the process of BASC is the same as the old unblock email list that UTRS replaced. It saved a lot of effort. Templates could be created, queues could be separated, and appeals could go direct from the UTRS queue and be promoted up to the BASC queue. Thoughts?--v/r - TP 21:34, 19 October 2014 (UTC)

@TParis:. UTRS works similarly to OTRS, correct? If that is the case I think this would be a big improvement over the way it is done now. Beeblebrox (talk) 17:42, 28 October 2014 (UTC)

It has similarities but it wouldn't be accurate to say they are alike. UTRS is basically a message management, workflow, and template organizing system. Messages are sent and received through the system, able to be transferred to several queues (Individual queues, checkuser queues, open proxies, a hold, and UTRS admin), and it supplies easy to use templates that cover 95% of the tickets. A BASC queue could be added with a specific BASC webpage interface. UTRS patrollers could send tickets to the BASC queue and tickets entered from the BASC interface could go directly to the BASC queue. The BASC queue would be invisible to others UTRS patrollers. It allows patrollers to have a sort of "chat" interfact, so BASC members could talk about the ticket within UTRS. And it has a search feature to bring up old tickets. The only caveat is that tool developers can see all tickets. I'm not sure if that is acceptable or not. An alternative is to give someone on BASC the code and have another instance set up specifically for BASC run by BASC.--v/r - TP 17:49, 28 October 2014 (UTC)

Extending mandate to community sanctions

This is something that I've suggested in my support comment. The main rationale is identical to that for bans and blocks : too much time wasted in community discussions, need for closure, and opportunity for review in a less heated setting. The list of sanctioned users is available here. I propose that we discuss this here, and maybe add this as a supplementary proposal. Cenarium (talk) 19:07, 29 October 2014 (UTC)

While I understand the concern I'm not so sure about this idea. BASC will currently hear appeals of community bans, but generally will not simply overturn them unless there was an obvious problem such as involved admins doing the close, extremely short time frame, and so forth. If they find an appeal of a community ban has merit they will usually only go as far as unblocking the user's talk page and/or opening a new community discussion. BASC has traditionally not heard appeals of topic or interaction bans at all as such users can appeal on-wiki. The purpose of this proposal is to make the subcommittee work more efficiently, expanding their mandate to include whole new areas could have the opposite effect. Beeblebrox (talk) 19:20, 4 November 2014 (UTC)

If there is an efficiency concern, we can wait until the new committee is up and running, and see after a while if it is running smoothly enough to see its mandate extended, which could be done progressively. I wouldn't expect a full committee to be overrun from just appeals of bans and blocks, though. There can be a back and forth with the community, the committee doesn't have to straight out overturn, same as for bans and blocks. Cenarium (talk) 21:05, 8 November 2014 (UTC)

This is attractive, but there are real issues were the committee (or sub-committee) to overrule the community. I have, I believe, suggested in the past that the committee should recommend that a community sanction be overturned (or changed). I see that as something that could be usefully done, given that the discussion would presumably be at a fairly relaxed and considered email pace, rather than on-wiki where people close discussions pre-emptively. All the best: Rich Farmbrough, 01:24, 6 November 2014 (UTC).

A mere recommendation may be insufficient to get closure, but there may be a back and forth up until the decision is made. After the decision is made, there could also be a window of opportunity of, say, one week, for the community to reach a consensus to reject or amend the decision. Cenarium (talk) 21:05, 8 November 2014 (UTC)