Wikipedia talk:Requests for comment/BASC reform 2014

Recusals
How would the new committee handle appeals where one of its members was the administrator who imposed the block? Presumably they would recuse from the decision-making, but would they be permitted to make a statement in defence of their action? Would the discussion move to a secondary mailing list, similar to the way ArbCom handles recusals? Would admins serving on the subcommittee be expected to avoid making blocks that had the possibility of ending up in the subcommittee's lap (in the same way that I gather AUSC members don't generally use their CU rights for routine matters)? I'm putting the cart before the horse a little, and I'm entirely in favour of the horse, but I'm curious to know whether these things have been thought about. Are there processes in place for the existing BASC? HJ Mitchell &#124;  Penny for your thoughts?  23:32, 15 October 2014 (UTC)


 * I touched on this in the proposed new structure: "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." BASC doesn't currently use a secondary mailing list in such cases but it may something to consider since this would nearly double the membership. Beeblebrox (talk) 01:50, 16 October 2014 (UTC)


 * Quite importantly would this information be shared with the appellant? All the best: Rich Farmbrough, 13:23, 11 November 2014 (UTC).

Statistics

 * How about before we discuss the expansion of what is already an overly bureaucratic process, we get some statistics on the number of requests and the number of unblocks that have been carried out by BASC since its formalization into an "official" subcommittee in 2009? Risker (talk) 01:41, 16 October 2014 (UTC)
 * Looking at the archives of WT:BASC, it appears that 9 people have been unblocked this year and 17 people were unblocked in 2013. I have no idea what that is as a proportion of all appellants. NW ( Talk ) 01:46, 16 October 2014 (UTC)
 * It might take quite a bit of digging to compile a total number for the year, but just to give a general idea, we have already rejected 14 appeals this month, and it's barely halfway over. Beeblebrox (talk) 01:54, 16 October 2014 (UTC)
 * Geez, Beeblebrox, this is a really important factor that the community needs to know before making a useful decision. I know those 14 appeals mean at least 30, and probably closer to 60, volunteer hours spent on reviewing unblock requests, on top of all the volunteer hours that had already been spent on earlier levels in the unblock hierarchy. This isn't good use of volunteer energy - not on the part of Arbcom (you've got enough "real arb" work to do), and it wouldn't be good use of time on the part of functionaries (many of whom are already providing support to the other levels of the unblock hierarchy) or the rest of the community.  Risker (talk) 02:21, 16 October 2014 (UTC)
 * I'd say that's inaccurate Risker. I've been doing the heavy lifting on BASC for the past 6-9 months, it's not hard for one individual to manage the queue. There's no way that 60 hours have been put into BASC this month, and depending on the effort put in by other Arbs, I'd say it's closer to 15. Stats below. Worm TT( talk ) 07:37, 16 October 2014 (UTC)

They're approximate, based on my notes. This is number of cases closed, rather than cases that have come in. Worm TT( talk ) 07:37, 16 October 2014 (UTC)
 * 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.


 * Hmm - so you think one person can easily manage the queue (and that largely, that person is you), but Beeblebrox feels its such an onerous and time-consuming task that it needs to be separated from the Arbitration Committee's functions. (On this point, I more agree with Beeblebrox, to be honest.  You're only counting your own time, not that invested by the rest of your colleagues. And if they aren't investing time, then essentially you've already got a system where one admin is unblocking.)  I took a very brief look at some of the unblocks listed at WT:BASC and noted one had been reblocked within a few days, and that at least two more were (even on a quick look) violating the conditions listed in the unblock notice; as I came across this during the course of a debate and not because of their behaviour, I won't reblock but will allow someone else to review and consider.  Risker (talk) 11:17, 16 October 2014 (UTC)
 * I'm not managing it. The quiet July and September was me taking time away from it. I'd certainly say that with the rest of the committee work, it's exhausting. Worm TT( talk ) 11:22, 16 October 2014 (UTC)

Please let me ask a question, from a position of complete ignorance. I am guessing that what happens is that, for example for the 14 cases so far in October, there are 14 e-mails to BASC, initiating each of the 14 appeal requests. Then for each one, I figure there are some messages back and forth between BASC and the appealing editor, some messages of discussion within BASC, and some amount of time looking into material by way of investigation. I'm also guessing, based upon the large percentages of declines, that a large percentage of the appeals are fairly obviously groundless. So I'd like to ask where amid all that does it become so time-consuming? That's a sincere question, because I have the feeling that a lot of the time, it's a matter of telling people politely "no", and because it seems to be in everyone's interest to figure out ways to decrease what is clearly, nonetheless, an excessive workload. --Tryptofish (talk) 20:11, 16 October 2014 (UTC)
 * I'm going to take a guess and say it's what I like to call the "herding cats" effect. I don't know how arbcom does things, but I know the oversight team recently tried to handle the decision-making process for a thorny oversight issue, and while people offered opinions on the mailing list, there was no way to make everyone weigh in, there was no voting process that could determine when enough people had weighed in which way, and there was no one whose job it was to determine the consensus of the discussion. As a result, the discussion sat for days, maybe even a week, with various people going "so...did we make a decision?" "I dunno, I gave my opinion" "Yeah, I gave mine too but I mean is that The Decision? Should we vote? Did we already vote?" "I dunno, let's wait to see if other people speak up" ad infinitum. Without a coach/secretary/overseer role and/or a defined discuss-vote-decide process, things get stuck in discussion mode much longer than they need to (and I don't think this proposed reform is necessarily going to fix that pitfall, if it's something BASC suffers from). A fluffernutter is a sandwich! (talk) 20:32, 16 October 2014 (UTC)
 * Thanks, that's a very helpful insight. Pending other people replying, of course, it seems to me to point to a need for some kind of process improvement within BASC or whatever might replace it. If the problem isn't coming primarily from the appellants or from the need to investigate, then it's actually a self-imposed wound. One solution might be to have a process with just two or three people (but not one) deciding on the appeal, even if the committee as a whole is larger. Another might be to define the decision process differently, such as, perhaps, setting a time frame for discussion, after which the quorum effectively drops to whoever has already taken part – or to assign, on a rotating basis, one person the responsibility for closing each appeal. But there is no good reason for the workload to be increased by the need to ask other members of the committee to express their opinions. --Tryptofish (talk) 20:48, 16 October 2014 (UTC)
 * One solution is to have 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 12:15, 17 October 2014 (UTC)
 * I like where this is going. In my opinion this is one of the issues that does hamper the smooth functioning of BASC currently. Technically, there are only four members. Yet for some reason the entire ArbCom is subscribed to the mailing list. This results in a situation where what constitutes a quorum fluctuates on a case-by-case basis and basically everyone and no one on ArbCom is a member. Beeblebrox (talk) 04:30, 18 October 2014 (UTC)

Contacting BASC.
I believe that a banned editor's talkpage (if blocked) should be automatically unblocked after 1-year of his/her ban, so she/he can contact BASC for reinstatement. Afterall, not all editors have Wiki-email. GoodDay (talk) 11:53, 17 October 2014 (UTC)
 * BASC isn't generally contacted by WP's internal email system anyway. There are dozens of ways to get as many email accounts as you want for free, I don't think it is an impediment to appealing that you use email, and I do think it is an important component of the BASC process that this final avenue of appeal is entirely different from all other forms of block or ban appeal. Decisions are not made by a single admin or UTRS agent, and there is no "peanut gallery" of users who offer unhelpful comments during the proceeding. It generally creates a calmer environment for all parties involved and gives the subcommittee a chance to work things out with those users whose appeals do show merit, as well as a way to quietly but firmly dismiss the cranks and trolls.
 * But even if you don't agree with any of that, this would open the door to literally tens of thousands of troll and vandal accounts to again waste the community's time with their obnoxious behavior. BASC is there to review the blocks or bans of people who have some chance of becoming productive members of the community again, this idea would open the floodgates and hopelessly backlog the subcommittee no matter how many people were on it. Beeblebrox (talk) 04:23, 18 October 2014 (UTC)
 * My point though, is an editor shouldn't have to contact an administrator to unblock his/her talkpage, while logged out. In my situaton, once I got my talkpage unblocked, I was able to place my reinstatement request on my talkpage, so that another editor could send my request to BASC. In order for me to get my talkpage unblocked first, I had to contact an administrator while logged out. GoodDay (talk) 17:02, 19 October 2014 (UTC)
 * And my point is that I have no idea why you believe you had to do it that way when you could have just used gmail or some other free email service that is no harder to sign up for than WP is. Beeblebrox (talk) 17:22, 19 October 2014 (UTC)
 * I didn't know how & couldn't learn how, to set up gmail or any other email service. IMHO, an banned editor's talkpage access should be restored after one year of her/his ban, to avoid these difficulties. Anyways, it's food for thought. GoodDay (talk) 17:26, 19 October 2014 (UTC)
 * I don't mean any offense, but I don't think this is a real problem for anyone but you. I have difficulty understanding how a user who has been on WP for nine years and made 140,000 edits would not be able to "learn how" to sign up for a free email account. Beeblebrox (talk) 18:56, 19 October 2014 (UTC)
 * The important thing is to accept that it is possible. GoodDay (talk) 16:18, 21 October 2014 (UTC)
 * The major providers now want invasive levels of detail. I signed up to Yahoo in order to help with Flikr/British Library, I was not happy with the personal details they demanded.  All the best: Rich Farmbrough, 01:58, 6 November 2014 (UTC).

Support make life better for arbcom
I've noted multiple editors going into arbcom one way and come out ... can I say "cranky"? And some folks just don't make it through their terms. It seems like over time the committee became the dumping ground for all privacy related tasks -- ban appeals, appoint CU/OS, misbehavior that can't be reported on-wiki due to privacy issues. So I support / encourage the committee and the community to split these tasks as much as possible, so the committee members can focus on the primary job of arbitration cases. NE Ent 22:41, 19 October 2014 (UTC)
 * I agree strongly. In an ideal situation, ArbCom would be only for arbitration. (See also User:Tryptofish/Draft for ArbCom, the top part of which has already been implemented, thanks to Roger Davies, and the bottom part of which got blown off.) --Tryptofish (talk) 22:47, 19 October 2014 (UTC)
 * (NE Ent asked me at my user talk to modify my choice of words where I said "blown off". I'm commenting here to acknowledge NE Ent's concern, and to acknowledge that my choice of words is debatable. After all, the very fact that I commented here reflects my appreciation for ArbCom and my wishes that their tasks could be made less onerous, so it wouldn't make sense for me to accompany those sincere wishes with a flame. I do think my words are debatable: one editor's "I got blown off" is another editor's "we listened to you and just were not convinced". But, having acknowledged that, I'm not actually going to change or retract what I said. Feel free to ask me at my user talk about it. In any case, let's not get sidetracked by a minor point. The important point here is that ArbCom needs to reassign all the privacy-related tasks elsewhere so that they can focus on arbitration. Peace. --Tryptofish (talk) 17:51, 20 October 2014 (UTC))

Adapt/adopt
In the question at the top, should "adapt" not read "adopt" ? W. P. Uzer (talk) 09:29, 20 October 2014 (UTC)
 * Yes, and now . PhilKnight (talk) 18:35, 21 October 2014 (UTC)

More Arbitrators= less work per Arbitrator
There is no good reason why ArbCom cannot consist of, say, 30 Arbitrators such that only a few who have enough time will be involved in any particular task. Count Iblis (talk) 03:13, 23 October 2014 (UTC)
 * There are two main issues that will need to be addressed for this to work. Firstly, there are not enough willing or/and suitable candidates to go around in recent years. Secondly, the current arbitration policy is designed in a way that the more arbitrators in a committee, the slower is the decision-making process. The arbitration policy will need to be substantially redesigned. - Mailer Diablo 09:27, 23 October 2014 (UTC)


 * Yes it should probably be redesigned first. If that's done, far more editors will run for ArbCom, because they then know that they can deal with the workload. Count Iblis (talk) 17:03, 23 October 2014 (UTC)


 * Much as has been discussed above about BASC, it may be less a question of having more members total, than of having fewer needed to take part in any given decision. --Tryptofish (talk) 18:47, 23 October 2014 (UTC)


 * Yes, having more people in total and splitting the functions of cases, appeals, and audits would reduce the workload and speed up the decision making process, but having a single large committee would actually slow things down. I wrote an essay - WP:ArbCom reform - about two different ways this could be achieved. PhilKnight (talk) 19:42, 23 October 2014 (UTC)


 * They already had to drop the requirement down from 70% to 50% approval to fill the seats we have. I can't possibly see that many more due to the lack of people who want the job.  Dennis 2&cent; 20:12, 23 October 2014 (UTC)


 * Although I don't think this is part of what is impairting BASC, I would also note that we've had dropouts as well, one before we even ofcicially started our term this year. What started out as a 15 member body has actually been more like 11-12 throughout most of the year.
 * Although their reasons for departing were varied, it is worht mentioning that being on the full committee is more of a real commitment than most WP users are acustomed to in that there are always things that have to be done and there are only a few people to do them, so you get less of a "bysatnder effect" than you do elsewhere. Every day there about aleast a dozen emails circulating about committee business, plus all the on-wiki activity (we've has an awfu lot of case requests lately) plus communications with the clerks, and then appeals on top of it all. And if you do dare to do any regular editing or admin work there are almost always people who will try and read some arbcom-related motivation into them, no matter how clear you are that you are acting only as yourself. Beeblebrox (talk) 22:34, 23 October 2014 (UTC)


 * Even if this were true, this would result in a drastically less efficient committee, and it is most assuredly not something we want. To make a well informed decision, arbs need to be well aware of the facts, so they would need to invest just as much time, maybe they wouldn't thinking that they are numerous enough, but then either decisions would be made by a random and unrepresentative subset of the members, or by uninformed people. The solution is not to increase the size of a committee with too many responsibilities, it is to divide it in several committees with more focused duties. Cenarium (talk) 18:52, 29 October 2014 (UTC)


 * The solution to this is to have a panel sitting on each case, rather than en banc. Five and a spare would be ample. All the best: Rich Farmbrough, 02:02, 6 November 2014 (UTC).

If an appeal is denied
I think BASC should also report all denied unblock case to Wikipedia talk:Arbitration Committee/Ban Appeals Subcommittee, which includes reason of declining the appeal. Otherwise, admins or other users which received emails of banned users will not know the user has made an unsuccessful appeal to BASC.--GZWDer (talk) 04:40, 2 November 2014 (UTC)

Progress
Well, this seems to have stalled out. What I believe I am seeing is that there is general support for spinning BASC off from arbcom, but a lot of users have concerns about the details. My plan as of now is to open a second phase of this RFC in order to refine the proposed changes and find consensus on what the new structure should be. However, this will not be happening real soon as I have volunteered to be one of the drafters on the GamerGate case as my last big project as a sitting  arb, and that is likely to take up most of my available wiki-time until sometime in January. After that I won't be dealing with cases, or BASC for that matter. Of course if anyone else would like to push this ahead in the meantime they are more than welcome to do so. Beeblebrox (talk) 19:55, 4 December 2014 (UTC)