Wikipedia talk:Arbitration Committee/History/January to June 2009

Other discussions
The range for the 'Other discussions and reports' list under construction will go from Arbitration Committee Feedback (started 17 December 2008) to Arbitration Role of Jimmy Wales in the English Wikipedia (RFC started 29 June 2009), attempting to cover everything in-between. As mentioned in the executive summary, any help identifying additional substantial discussions not mentioned here would be appreciated. Please list below any items you would like to see included. Carcharoth (talk) 04:08, 22 July 2009 (UTC)

Feedback and questions
Please place feedback and questions here, indicating or quoting the section of the report to which you are referring. Feel free to start subsections here, or ask general questions. Some of the questions will be answered in the full report, but asking here will help improve that report if there are areas where more detail is needed. Carcharoth (talk) 04:08, 22 July 2009 (UTC)


 * You average 80 emails per day, really? That is perhaps the purest expression of ArbCom's workload I've ever seen.  Dragons flight (talk) 07:22, 22 July 2009 (UTC)
 * Yes, but that is only arbcom-l. Then there is func-en, oversight-l, and clerk-l, and most arbs are also on checkuser-l and at least reading wikien-l.  And then there is all the private emails... John Vandenberg (chat) 08:09, 22 July 2009 (UTC)


 * Nothing about the advisory council? Nathan  T 16:32, 22 July 2009 (UTC)
 * That was announced in July, so it will be included in the next report. John Vandenberg (chat) 01:53, 23 July 2009 (UTC)
 * John is quite right. To expand on what he said, I started preparing this report in the last few weeks of June, and spent a fair amount of time completing it while things were happening in July (including the advisory council announcement and RfC). The half-year report was only ever meant to go up to the end of June. I know many people want to discuss the ACPD. I myself have commented in some of the discussions about it, though fairly late on, as I was occupied with preparing this report. But it would be great if people did have questions about what we've done in the first six months of this year, not just in July! I'd be more than happy to answer any questions Nathan or others have about the report or what we did in those six months. I'm also happy to talk about arbitration activities in the last month, but that is better discussed elsewhere. Carcharoth (talk) 00:13, 24 July 2009 (UTC)
 * Sensible answer, should have thought of that. I think you folks accomplish a great deal and do it very well, and I think you've made important strides in making your operations more consistent and professional. I would like to see progress on a process for replacing resigned/removed arbitrators, as I've mentioned before in various places. Any discussion on that subject? How have the subcommittees contributed to the workload/workflow? Is it expected that they will ramp up over time wrt taking some work off ArbCom's plate? A narrative on the impact so far of the subcommittees, significant business they have conducted, and the role they are expected to play in the future would be an interesting addition to this report. Nathan  T 18:49, 24 July 2009 (UTC)
 * Since you folks asked for feedback, I'm still hoping that you might respond to the remainder of mine ;) Nathan  T 01:04, 13 August 2009 (UTC)
 * Sorry. I was going to wait for more feedback and then summarise what has been said and present it to the committee and ask them to comment here. But a quicker way would be to point out this page and ask other arbitrators to respond. In my view, the best way to get issues like the one you raised addressed is to ask for it to be placed on the agenda, and to have arbitrators second your proposal. I think some of what you ask is already on the agenda. But getting the community more involved in prodding arbitrators to discuss the items on the agenda (and add new items) would help. The reason, by the way, for such a bureaucratic process, is to ensure deliberative discussion takes place on agenda items (rather than rapid-fire responses in a day or two with little thought or consideration, and ending up with a "solution" that makes things worse), and to ensure things are prioritised correctly. Dropping everything to discuss a suggestion when it is made would not be the right way to do things. But neither is ignoring suggestions. I hope a balance between the two can be found. The agenda is here (with a link for discussion - to a section at the top of WT:AC/N). The calendar is here. We have tended to slip a bit on the due dates, but I hope we have managed to deal with things in an orderly fashion. The other thing I should mention is that there is nothing stopping a community discussion on particular points being started (though please ensure that a wide cross-section of noticeboards and venues are notified of it), and the results brought to the agenda discussion section. If there is strong community consensus for ArbCom to change something, and no objections are raised, we will almost certainly adopt such changes. It would be nice if a note could be dropped at the agenda discussion section to let us know when such discussions of agenda points are started. Carcharoth (talk) 13:58, 16 August 2009 (UTC)

It would be interesting if the statistical summaries included the average number of bytes and edits to case pages. There were fewer cases in 1H2009 than 1H2008, but my vague subjective sense as a not-very-close observer is that they have gotten messier. 67.117.147.249 (talk) 18:42, 24 July 2009 (UTC)
 * The only one I looked at specifically was the Date Delinking case. The pages, altogether, were as long as War and Peace. I won't comment on the relative merits of the prose. :-) Risker (talk) 19:29, 12 August 2009 (UTC)

I don't think the 80 emails a day is particularly instructive. What would actually be useful (but you'd need discipline in your subjects and a bot to do the work) would be a day-by-day analysis of how many emails are sent about the various cases (is there any reason why this couldn't also include who sent them?). This would go a considerable distance to assuage (or possibly justify, who knows) the complaints that arbcomm ignore cases for long periods William M. Connolley (talk) 23:04, 11 August 2009 (UTC)
 * I don't know that what you're asking for is particularly instructive either. Arbcom-L communication about a case is only one aspect of the work related to it; there is the (often voluminous) reading of case pages, possibly discussions or a draft decision on the arbwiki, commentary on the mailing list, following of contemporaneous relevant events onwiki, discussion on workshop pages and other onwiki pages, the actual voting, etc. There's also little relevance about who sent emails on any given topic; one arbitrator whose task is to motivate others to stay on schedule may make half a dozen posts, while another arbitrator may post only onwiki and a third may post a single but very salient comment to the mailing list. I believe the purpose of providing the "number of emails" statistic was simply to give a small taste of the workload involved in the role. Risker (talk) 19:41, 12 August 2009 (UTC)
 * This rather sounds like excuse making. Of course a breakdown of emails doesn't tell you everything; that is no reason to do nothing. My suspicion is that a break-down of this nature would reveal rather too much about arbcomm activity levels for arbcomm comfort William M. Connolley (talk) 20:16, 12 August 2009 (UTC)
 * Actually, I wasn't hugely in favour of releasing the number of emails either, because it naturally leads to questions like yours and removes the focus on total workload and activity. Several arbitrators have written very extensive or multiple decisions, each of which take many hours to do. First Kirill Lokshin and now Roger Davies have been coordinating arbitrators. The AUSC and Ban/block review subcommittees also add to the workload of individual arbitrators, as they rotate through those roles. Carcharoth invested more than 40 hours preparing this summary, I invested even more time coordinating the checkuser/oversight election, and others have worked on policy draft and review. How is knowing how many emails a particular arbitrator sent with respect to a specific case going to be useful, in any real sense? Risker (talk) 20:35, 12 August 2009 (UTC)
 * It would shed some light on the opaque world of arbcomm. Why doesn't someone try it, for example for the Abd-WMC case. I bet it would be instructive, and for that reason unpublishable William M. Connolley (talk) 21:57, 12 August 2009 (UTC)
 * In what way do you see it as being potentially instructive? If the answer is "this case was mentioned in X posts in Y threads by Z arbitrators (Arb A= Q posts, Arb B = R posts, Arb C = S posts, etc.)....what does it tell you? We both know it tells one absolutely nothing other than that the case was mentioned in X posts in Y threads by Z arbitrators. It is roughly equivalent to my going out to the parking lot and doing a statistical study on how many cars have the number "0" in their license plate number - and so on through all digits 0 to 9. It may be factual, but it's just a factoid. A more useful statistic involving that case would be how many hours each arbitrator spent reading the evidence and workshops, and that cannot be adduced by quantifying mailing list posts. Risker (talk) 22:07, 12 August 2009 (UTC)
 * You really do seem to be determined to find reasons for doing nothing, on the grounds that perfection is unobtainable. More specifically, I suspect that a day to day breakdown would reveal zero activity on a case by most arbs William M. Connolley (talk) 22:28, 12 August 2009 (UTC)
 * See, that's my point. Mailing list activity does not equal arbitrator activity, on this or any other case. Arbitrator activity on a case has too many variables for mailing list activity to be indicative of total activity on a case. So, if Arbitrator C has no mailing list posts on this case, but has posted on several threads in the workshop, and has invested 20 hours reading evidence and other pages, it looks like Arbitrator C is inactive if only the mailing list threads are reported. So no, it's not a useful metric at all. Risker (talk) 23:01, 12 August 2009 (UTC)