I should point out that Kirill Lokshin and I are the only two arbitrators who regularly draft proposed decisions.
Fred
-----Original Message----- From: Kirill Lokshin [mailto:kirill.lokshin@gmail.com] Sent: Sunday, October 14, 2007 12:49 PM To: 'English Wikipedia' Subject: Re: [WikiEN-l] Arbcom
On 10/14/07, Thomas Dalton thomas.dalton@gmail.com wrote:
It would be *more* susceptible to burnout than the current model,
actually,
since there's no provision for the scenario where none of the five randomly-selected arbitrators are willing to do any case work. In a
system
where the full Committee hears every case, a few active members can keep things moving (to some degree) even if the bulk of the Committee isn't actively participating.
(Or, in other words: we can currently close a 5-0 case if we find *any*
five
active arbs, but, in the new model, we would select five arbs first and
then
expect them to be active -- which is a rather dangerous assumption to
make.)
You make a good point. It should work, though, as long as the arbs are honest about when they are and aren't able to be active. They'll be occasions when something comes up and an arb becomes unexpectedly inactive, but in most cases they should know far enough in advance whether they are able to take on a case or not. An inactive arb can be replaced by one of the others if necessary, of course.
"Active" doesn't mean quite what you're assuming it does, though. An active arbitrator is merely one that's participated *in some way* in the recent past; it doesn't constrain them to a particular level of activity.
The basic process of a case is something like this:
1. Case submitted 1a. Delay 2. Case accepted 3. Evidence submitted 3a. Long delay 4. Proposed decision drafted 5. Voting begins 5a. Long delay 6. Case closed
A random panel would essentially attempt to decrease the voting delay (5a) by reducing the number of needed votes, at the cost of increasing the drafting delay (3a) by reducing the chance that an arbitrator would be available to draft a decision (something which, historically, not all arbitrators do a lot of).
You could, I suppose, have a system where arbitrators not on the voting panel could draft a decision for the panel to consider; but that doesn't seem very helpful. Alternatively, you could reduce the voting delay by allowing unopposed cases to pass with smaller majorities, but that would be yet another can of worms.
Kirill _______________________________________________ WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: http://lists.wikimedia.org/mailman/listinfo/wikien-l
On 10/15/07, fredbaud@waterwiki.info fredbaud@waterwiki.info wrote:
A random panel would essentially attempt to decrease the voting delay (5a) by reducing the number of needed votes, at the cost of increasing the drafting delay (3a) by reducing the chance that an arbitrator would be available to draft a decision (something which, historically, not all arbitrators do a lot of).
You could always have a non-random panel. Allocate arbs to cases based on their areas of expertise (interpersonal disputes, certain types of admin actions, certain policies). This eliminates the problems of a random panel, although it would certainly be less practical and could well concentrate work into a few hands.