Wikipedia:Arbitration Committee Elections January 2006/Quadell's Proposal

Let's think outside the box a minute. Why not have unlimited arbiters, in the same way that Admins and Bureaucrats are chosen? Anyone can run at any time, and the community votes for any candidate, just like for an admin candidate. If someone is sick of the job, or if an Arbcom member has been inactive for a certain period of time, he/she is simply removed from the committee (but can re-run later). Perhaps Jimbo could have official veto power on candidates.

It's very different from the way things are done now, but look at the advantages.

Advantages

 * There will always be enough arbcom members.
 * Because of the above, arbcom members will have to hear fewer cases (each), possibly reducing burn-out.
 * The solution is scalable - as Wikipedia grows (and the number of cases grows), the number of arbcom members will naturally rise as well.
 * All arbcom members will have the consensus support of the community - not just 51%, or an appointment
 * The process is greatly simplified: no annual elections, no tranches, no voting software, no emergency appointments, etc.
 * The system doesn't depend on Jimbo. As great as Jimbo is, it's best that Wikipedia rely on itself as much as possible.
 * Good-hearted editors who aren't sure if the arbcom is right for them could run and get voted in with no pressure. After hearing a few cases, if they don't like it, they can quit. No big deal.
 * The consensus system a system that works well for Wikipedia in many other domains.

Possible objections and responses

 * But then there would be too many Arbcom members! Look at how many admins there are.
 * I don't think so. Look at how many bureaucrats there are.
 * But then a bad apple could slip in - after all, occasionally an admin will pass the test, leading to problems. This would be much worse with a bad arbcom member.
 * Being an admin is, officially, "no big deal". Being an arbcom member is a big deal. It should be much harder to become an arbcom member than an admin. And if Jimbo had veto power, this objection is lessened.
 * How would decisions get assigned to Arbcom members?
 * A taxi system, or "first-come, first-served", or something. I'm open to suggestions.
 * Shouldn't there be requirements, like first being an admin, or having been around for a year, or something?
 * Possibly, but not necessarily. Newbies routinely get voted down as admins. No hard-and-fast rules are necessary.
 * But wouldn't this be a sort of popularity contest?
 * In some ways, yes. But so would any election. (And appointments are popularity contests too, except that you only have to be "popular" to one person.)

Comments
I'd be very interested in hearing comments on this proposal, especially from Arbcom members and seasoned Wikipedians, but also from anyone interested. – Quadell (talk) (bounties) 19:05, 26 October 2005 (UTC) You would have to use first past the post voteing which would make block voteing far more significant.Geni 23:54, 26 October 2005 (UTC)
 * Interesting idea. I have to say this is one of the best ideas I've heard on this topic yet. BLANKFAZE | (&#1095;&#1090;&#1086;??) 20:54, 26 October 2005 (UTC)
 * While I support a much larger slate of arbitrators (perhaps ten or twenty times as many as we now have), and some sort of round-robin, taxi, or random assignment of potential cases, arbitration is still difficult, skilled work that also requires a certain set of personality traits. Having a mix of experienced and inexperienced arbitrators is part of the advantage of staggered terms and a small enough group where you know you'll get a few experienced arbitrators in each case.  Further, based on observing the administrators here, there is a large chance that they will mistake the goal as "punishment" or "justice" instead of "whatever's best for the Wiki".  I'd rather see steady, monthly election of a few new arbitrators to expand the panel until we've reached a good panel size.  64.254.131.65 21:08, 26 October 2005 (UTC)
 * That seems to me like a lot more complication than the system I'm proposing, with no particular gain. (In my opinion.) – Quadell (talk) (bounties) 13:15, 27 October 2005 (UTC)
 * Assigning randomly is a bad idea: people get cases they wouldn't be interested in: recipe for Slowness. Instead: 48h after the accept-a-case threshold is passed, all those who've voted accept are on that case. The delay allows people who didn't want the case, but who want to be on it if it *is* accepted, to change their vote. William M. Connolley 22:08, 23 November 2005 (UTC).
 * I really like the idea of assigning cases randomly. If we want to get as impartial a panel as possible, it would be the way to go. If jurors in criminal trials could chose the case they sit on, they would be more likely to chose a case in which they have a bias or agenda. For example, someone who has been raped may want to sit on a rape trial or someone who is for the death penalty may want to sit on a capital case. I don't see disinterest in the case as a problem. It will help them be dispassionate. Assuming unnecessary cases are/will be weeded out through some process, I think it's a good idea. Sorry about getting off topic. -- Kjkolb 16:19, 1 January 2006 (UTC)
 * Why do you say that having one election a month, with two or three new vacancies to fill each time, requires a particular voting method?    Un  focused  03:06, 27 October 2005 (UTC)
 * that wasn't the proposal I was commenting on.Geni 04:09, 27 October 2005 (UTC)
 * Sorry about the mixup. But I don't see how Quadell's proposal requires any specific voting system, either.  Or is your comment a preference stated as a requirement?   Un  focused  05:04, 27 October 2005 (UTC)
 * if people are voteing yes and no on one person then there is only really one way to run the vote (the way we do at RFA). the Problem with this method is that it encourages block voteing.Geni 05:10, 27 October 2005 (UTC)


 * In the system I'm proposing, you wouldn't be voting for a candidate for a specific post, so no voting system would be used. It would just be consensus voting. If three people are nominated, you can either object or approve of each one individually. If a consensus supports advancement of all three, then you have three new Arbcom members. If none of them get a consensus, then you have no new ones. So there's no "block voting". – Quadell (talk) (bounties) 13:15, 27 October 2005 (UTC)


 * that is what is done on RFA and block voteing certianly exists there.


 * While the unlimited-size arbcom solution might 'scale' with Wikipedia's growth, I don't necessarily think it will follow that the voting for an unlimited-size arbcom would scale quite as well. -User:Fennec (はさばくのきつね) 21:54, 1 December 2005 (UTC)


 * Overall, an interesting idea, I have a question though. What method would be used for removal of an arbcom member that later (as unlikely as it may be) acts in contrary to the spirit of Wikipedia? --Robert Harrisontalk contrib 07:39, 9 December 2005 (UTC)


 * I'm against this especially at a time when I see more admins squabbling among themselves. The main advantage of elections in one time in one places is that everyone's eyes is on the event.  RfA is not the best watched and I think that has contributed to recent troubles with admins.  Interesting idea and maybe should be taken into account but I'd rather see the size of the ArbCom pool increased first but still have all elections at once.  Maybe a monthly event and you can submit your name for the job whenever you want. gren グレン 03:33, 26 December 2005 (UTC)