Historical background on the formation of the audit subcommittee may be found at Wikipedia:Review Board and User:Thatcher/Quis custodiet ipsos custodes and their associated talk pages. Thatcher 15:57, 23 April 2009 (UTC) reply

sent via email just now;

G'day Tzn,

I see you've drawn a short straw, and found yourself on the 'audit subcommittee' :-) (and congrat.s / comiserations to your colleagues, whom I presume to be reading this too).

I have previously expressed serious concerns about the use of checkuser on the english wikipedia, relating to the possibility of abuse occurring within the letter of the WMF privacy statement, but ultimately harmful to the project - specifically fairly random checks, the sharing of username information etc. etc.

I am interested in the audit subcommittee examining checks run on my account to date, in particular to ensure that appropriate rationale's have been submitted.

I would further like to know how many checks have been run, and by how many checkusers.

In addition, it seems to me that there is a general consensus that every candidate for arbcom at the end of last year was 'checkusered' - I'd also like the subcommittee to confirm that this was run with appropriate propriety.

Public examination of checkuser proceedings, and possible ratification / approval from your committee will no doubt increase confidence that the watchers are indeed being watched in some way - in fact, this is the only way! Many thanks for volunteering to take on this task - it's appreciated (if occasionally quietly!)

Happy for this investigation, and all IP information etc. to be published on wiki - indeed, I'd prefer it. FYI, I use iinet, and maybe a few other major australian ISPs (from other spaces).

Happy to answer questions, go into further detail / generally explain my concerns further in a way you'd prefer.

best,

Peter PM.

ps (not in the email)

I just realised that when I say I'd 'like to know how many checks have been run' - it's not immediately clear that I mean 'on me' - sorry about that! Privatemusings ( talk) 21:59, 23 April 2009 (UTC) reply

• Sounds like two requests, actually. Some patience will be required, we have not set up any internal processes for handling requests and Tznkai and I don't have our flags set yet. Thatcher 23:40, 23 April 2009 (UTC) reply

I suppose it does, really - happy to split them off and for you guys to consider the arbcom election thing completely separately (or whatever you think best). I've since heard from Roger and Tzn via email - this system looks like it's heading towards being in good shape :-) - I don't really expect much more word for a week or so, though do get in touch if you'd like any more info / diff.s or whatever. cheers, Privatemusings ( talk) 07:30, 24 April 2009 (UTC) reply

I received a response from Tzn containing the info that 8 checkusers ran checks on my account, and mentioning that in the absence of specific complaints, the sub-committee see no basis for further investigation - I've responded;

Hi Tzn,

Thanks for taking a look, Tzn - I'd note though that it's hard to assert that the rationale's for the many 'checks' against my account may have lacked sufficient merit without seeing them :-)

Could you forward me the rationale's for the checks, in order that I might assess them?

It's certainly my position at the moment that I'm unaware of any rationale for checks to have been run which might not be self-certifying, with the concurrent 'chilling effect' of a rather political agenda.

Dunno if that makes sense really! - perhaps it would be more useful to have an 'on-wiki' brief exchange?

best,

Peter, PM.

ps. I'll drop my note in on the wiki, and am quite happy for all correspondence in this matter to be copied there too. Privatemusings ( talk) 05:37, 5 May 2009 (UTC) reply

• I'm not sure how much value there would be in providing detailed rationales. The subcommittee has examined the issue and is not concerned. We could list the specific rationale ("Privatemusings insulted my mother") but that may only lead to another objection ("It was not an insult, it was constructive criticism"), which takes us into a discussion of whether the checkuser had a reasonable belief at the time that you insulted his mother even if you didn't mean it that way, but ultimately the subcommittee has to decide whether or not the rationale was acceptable even though the parties involved may disagree.

As far as "chilling effects" and political agendas are concerned, let us imagine three types of self-initiated checks:

1. Adam checks Eve because Eve is editing provocatively (or disruptively) in a manner reminiscent of banned user Lilith.
2. Adam checks Eve because Eve is editing provocatively (or disruptively) in a manner that suggests a previously banned user or bad hand account, but Adam does not have a specific suspect in mind prior to running the check.
3. Adam checks Eve because Eve is editing provocatively (or disruptively) in a topic or policy area that Adam is involved with, and Adam hopes to find dirt on Eve to get rid of her.

It is my own opinion (not speaking for the subcommittee) that checks of the first two types are rarely going to be of concern. Many banned users do in fact try to return under the radar and stir up trouble, both overt and subtle. Checks of the third type will almost always be of concern. Checkusers are granted broad discretion to investigate and prevent disruption, but that does not extend to trying to "get" people who annoy them. Having said this, and after examining the contribution histories of your various alternate accounts and of the checkusers who checked them, the subcommittee sees nothing that concerns us that the checks were inappropriate. Thatcher 14:48, 5 May 2009 (UTC) reply

I am a bit surprised that the appointed members were granted Checkuser and Oversight, when even the global Ombudsman are only granted log access. In theory audit committee members could abuse checkuser or oversight and it depend on non-audit committee members or other audit committee members to report it (basically the same situation as before). Wouldn't it make more sense for Audit Committee members to only have log view access and not have the ability to perform checkusers or perform revision deletes? Then it would be impossible for an audit committee member to violate a policy, because they wouldn't have the tools to do so. MBisanz ^talk 02:26, 26 April 2009 (UTC) reply

• As far as I know there is no such thing as log only access to checkuser, unless it was implemented while I was away. In any event, part of our investigations may involve verifying other checkuser results. For example, if we question why Adam checked Eve, and Adam says, "because of something I saw when checking Cain" then we may need to check Cain as well. Thatcher 02:36, 26 April 2009 (UTC) reply

• In MediaWiki the functions are known as checkuser-log vs. checkuser, it can be set either by Brion the way Rollback and Account creator were done or by the Stewards using the global set function of global groups. MBisanz ^talk 03:13, 26 April 2009 (UTC) reply

Overall, CU access I think will prove to be a rarely used, but integral part of our investigation process. Oversight logs only however, if they're segregated would be fine in my opinion.-- Tznkai ( talk) 03:30, 26 April 2009 (UTC) reply

FWIW, I have full checkuser access to every wiki, not just log rights. Note that it is sometimes necessary to have the ability to do a CU query to judge the necessity and validity of actions taken in the original check. [[Sam Korn]] ^(smoddy) 13:45, 28 April 2009 (UTC) reply

I agree that full access to use of the CU tool may be necessary to investigate and monitor CU access. Additionally, understanding how the tools work by using them a few times could be beneficial for someone providing review of the use of the tools. FloNight ♥♥♥ 14:51, 28 April 2009 (UTC) reply

MBisanz makes a good point that there is still some possibility of abuse, but I think we can at least all agree that the current situation is better than the previous situation where there was no review board at all :-) -- Deskana, Champion of the Frozen Wastes 14:53, 28 April 2009 (UTC) reply

I too had some concerns similar to MBisanz, but as long as everyone monitors everyone else I think it should be fine. Tiptoety ^talk 23:26, 2 May 2009 (UTC) reply

Traditionally arbitrators have retained checkuser and/or oversight after their term expires. For individuals who did not have either or both rights before they served on the AuditCom, what will happen to those rights at the end of their term? Will they be removed as Global Ombudsman is handled or will they become normal Checkusers/Oversighters? MBisanz ^talk 02:31, 26 April 2009 (UTC) reply

• Good question, don't know. Thatcher 02:33, 26 April 2009 (UTC) reply
• We don't have an official policy, but since I'm the only person not already cleared for CU/OS, it really falls on me. I've elected to post personal policies in lieu of official policy for the time being, and I will be giving up the permissions at the end of my term.-- Tznkai ( talk) 03:24, 26 April 2009 (UTC) reply

• Well I don't actually think Thatcher had Oversight before (I could be wrong), and of course there will be questions with future holders (if elected by less than 66% for instance), so it probably would be a good idea to make a policy (even if it is copying your userspace). Also, if Arbcom is not going to appoint new CUs and OVs, it might be worth you both keeping the right, given that we've lost several users over the last quarter and now are losing 4 more with this committee. MBisanz ^talk 03:30, 26 April 2009 (UTC) reply

• I will stand for elections (if we're still doing that) if I decided to keep either permission in any case, but I have no problem with an official policy superseding what I've done.-- Tznkai ( talk) 03:35, 26 April 2009 (UTC) reply

I am of the opinion that auditers should have their permissions revoked after their term is up, unless they would otherwise have the permissions anyway (such as people already having the permissions before serving on the board). -- Deskana, Champion of the Frozen Wastes 14:47, 28 April 2009 (UTC) reply

• I agree with Deskana here. Should an ex-auditer want the tools after their term is up they can run for CU/OV election when the time comes around. And personally, I am of the opinion that once an Arbitrators term is up, they too should have CU and OV removed. Just like I said before, should they wish to continuing using the tools once their service as a Arb is up, they can run in a CU or OV election (but that's a discussion for elsewhere). Tiptoety ^talk 23:29, 2 May 2009 (UTC) reply

• I'll add a voice to that. In fact I think Wikipedia should move towards the general principle, that powers received while serving in any post are relinquished at the end of term or resignation. ie. incumbant El Presidente hands off the nuclear codes to president elect on the first day of office. -- Joopercoopers ( talk) 12:07, 18 June 2009 (UTC) reply

Statistics

Is there a policy requiring a logged reason for each use of CU? I note that "Hotel" has provided a reason only 4.4% of the time while other users are in the mid-90s. This lack of reasoning is disturbing to say the least. KnightLago ( talk) 14:16, 28 April 2009 (UTC) reply

I have always thought it should be mandatory to provide a reason. You generally only need to do so once in an investigation. I don't believe I managed a 100% reason rate, but when I failed it was due to a mistaken press of the button, not a deliberate decision. [[Sam Korn]] ^(smoddy) 14:31, 28 April 2009 (UTC) reply

I think I've figured out which I am and I must say my logged reason rate, although high, is lower than I was expecting. I will try to be more diligent in that regard. -- Deskana, Champion of the Frozen Wastes 14:42, 28 April 2009 (UTC) reply

We would be happy to tell any checkuser (privately) who they are. Logged reasons are very important for determining that checks are reasonable and appropriate, both as a Foundation issue (which is long-standing) and for the Audit Subcommittee. We have referred these stats to Arbcom for review. Thatcher 14:47, 28 April 2009 (UTC) reply

It would be fairly trivial to add some JavaScript to require a non-blank reason field (or at least a quick confirm .alert() if the reason was left blank). For what it's worth. -- MZMcBride ( talk) 15:04, 28 April 2009 (UTC) reply

That might be quite useful; I know I set my preferences to remind me to include edit summaries when posting edits, and it's rare that I'll skip the summary now. Risker ( talk) 16:00, 28 April 2009 (UTC) reply

That would be a good start. It should be trivial to require a reason server-side and therefore not to allow someone to override or ignore the requirement. [[Sam Korn]] ^(smoddy) 17:17, 28 April 2009 (UTC) reply

And indeed I have filed a bug requesting exactly that. [[Sam Korn]] ^(smoddy) 17:42, 28 April 2009 (UTC) reply

Of course, what we really want are informative reasons, but any prompt is a good first step. Thatcher 17:46, 28 April 2009 (UTC) reply

Can that same prompt remind the CU to fill out an informative reason? The prompt isn't smart enough to tell which obviously, but a nag will help.-- Tznkai ( talk) 18:04, 28 April 2009 (UTC) reply

Excellent idea, MZMcBride! Thanks for filing the request, Sam. FloNight ♥♥♥ 18:00, 28 April 2009 (UTC) reply

Thanks. Two comments -

1. I'm not able to think of an obvious reason why checkusers cannot be named in this table. Certainly I did in 2008; no problems were suggested arising from it. The respect for privacy is exemplary, but in this case I don't see much benefit. Could consideration be given to giving actual usernames, or at least checking if there are any strong disadvantages.
2. Giving a reason, by itself, isn't necessarily sufficient to confirm independently that the check was appropriate. The "reason" field should somehow give some source, or explanation, why the check was needed - the case it pertained to, a wikilink/checkuser-l/arbcom-l reference, or some actual explanation or common reason. While it's not able to be precise, is there any chance of showing in future stats, a rough idea how many checks have "typical wordings suggesting a good reason"? A due caveat would be needed but this is generally quite helpful.

FT2 ^{( Talk |  email)} 01:16, 29 April 2009 (UTC) reply

AUDSC's report to ArbCom contained more information including more notes about activity levels and wordings in the CU summary log. AUDSC has been in existence for a short time. We are focusing on priority issues now. AUDSC will tailor our work to the feedback we receive from ArbCom, functionaries, and the Community. Likely that something addressing the wording in CU summary will come out later from AUDSC or ArbCom. FloNight ♥♥♥ 18:13, 29 April 2009 (UTC) reply

Just as a note, assuming it isn't reverted, when revision 50638 goes live, the reason field for CheckUser will be compulsory. -- MZMcBride ( talk) 17:31, 16 May 2009 (UTC) reply

Yes, although that doesn't mean people will use informative reasons, which will still be important. Thatcher 22:15, 16 May 2009 (UTC) reply

Ordering

Is there any specific reason for the order in which the stats appear? They overlap (pink and yellow), and I'm wondering if this was on purpose, or if it should be adjusted. Syn ergy 20:24, 28 April 2009 (UTC) reply

Click on the little arrows in the header. They can be sorted alphabetically or numerically by any column. Someone colored the arbs and non-arbs differently, its in the notes. Thatcher 20:32, 28 April 2009 (UTC) reply

Mmm. Thank you for clarifying that. Syn ergy 20:34, 28 April 2009 (UTC) reply

darker?

Can we get slightly darker pastels (I'll fix it myself if this isn't a bot generated table), as on my laptop these two shades are just about indistinguishable they are so washed out. Thanks. PS I strongly agree reasons are goodness. They help other CUs too, not just the audit committee. ++ Lar: t/ c 21:24, 28 April 2009 (UTC) reply

Go ahead. Happy melon added the colors in the first place. I already coded the Arbs and non-Arbs by name, but you can play with color if you want to. Thatcher 22:10, 28 April 2009 (UTC) reply

Done. Some of us are kinda blind. (I'm sure some sockpuppeteers say that about my checks!) ++ Lar: t/ c 02:50, 29 April 2009 (UTC) reply

Completeness

The chart shows 32 checkusers, however, there are currently 40 users with the checkuser right, and if this goes back six months, it should be picking up CUs for Alison, Kylu, Morven, and Rebecca, meaning there should be at least 44 rows, not counting emergency Steward actions. MBisanz ^talk 01:37, 29 April 2009 (UTC) reply

• Our frame of reference here was (1) who is using/not using logged reasons, and (2) who is so inactive that it might be desirable to remove their access. So we worked from the list of active checkusers, and we removed the developers and members of the Audit committee who are inactive as far as logged checks are concerned but who aren't "inactive" for the purposes of thinking about who should have access. Arbcom got a full report, and if they want to release it instead, with or without the real names, they just have to tell us and someone will update the table. I think we just did not think about retired checkusers (as far as I can recall, the other checkuser who retired in the last 6 months was Alison), but listing retired checkusers does not serve either the purpose of getting people to log their reasons or finding out who the active and inactive CUs are, so I guess we just didn't think about it. Thatcher 02:27, 29 April 2009 (UTC) reply

• I understand, I thought it was just a straight pivot table of the checkuser log. MBisanz ^talk 03:20, 29 April 2009 (UTC) reply

I've found the stats for Alison and Morven and the Steward actions. If Arbcom approves releasing a version with the real names I will add it. I know not of this "pivot table" of which you speak. My wife surely knows, but she thinks Wikipedia is foolishness. I could probably extract the checkuser log as a text file and stick it in Excel, but if Excel can automatically count the names of the different users, I don't know how to do it. Hmmm. Thatcher 03:32, 29 April 2009 (UTC) reply

if Excel can automatically count the names of the different users

Yes, that is what a pivot table would do. MBisanz ^talk 03:39, 29 April 2009 (UTC) reply

I have to copy the log to Word first to put in some tabs. I wonder how Word would handle a 30,000 item find/replace... Thatcher 03:44, 29 April 2009 (UTC) reply

Boy MBisanz, you really know how to make a guy feel silly about working by hand.-- Tznkai ( talk) 03:48, 29 April 2009 (UTC) reply

My first boss ever gave me a half hour lecture about how amazing pivot tables were and how I had to use them on every analysis I did. Ever since then I've been spreading the word. MBisanz ^talk 05:18, 29 April 2009 (UTC) reply

(<-)Speaking as someone who has to live and breathe Excel, pivot tables are amazing when used properly (as are data tables, but that is something else). Now start using OFFSET, *LOOKUPS, MATCH, INDEX, and INDIRECT and you're cooking with gas :D-- Avi ( talk) 17:37, 29 April 2009 (UTC) reply

As a software engineer who has had to migrate offices away from managing billions of dollars using spreadsheets whose complexity was only surpassed by the fragility of it, I must voice my objection to any and all spreadsheets. Except maybe when used for shopping lists. And, I have seen all those functions used on shopping lists as well, to good effect mind you, but at least there the worst that can go wrong is the smooth running of a shrewd household budget falls apart. John Vandenberg ^{( chat)} 18:52, 29 April 2009 (UTC) Sorry, I was trying to ignore this thread... reply

For heavy-duty statistical analysis, I will use a platform such as R or our proprietary in-house software. But there is something almost elegant in the ability to quickly analyze and display data in a spreadsheet. Furthermore, depending on the level of managing complexity, a spreadsheet with enough built-in functionality may be sufficient even when managing billions of dollars. Now, if your billions of dollars are broken down into thousands of categories and you are interested in a Variance-Covariance matrix, the spreadsheet is not the proper tool. I think we just brought this discussion WAY out of scope of this page, though :) -- Avi ( talk) 19:09, 29 April 2009 (UTC) reply

I actually did very similar analysis to this myself back in (I think) June last year. I used some pretty simple regexes to put the HTML display of the log into a local database and then did all the manipulation using SQL. I am personally of the opinion that this is a far nicer method than spreadsheets. I don't claim to be a statistics wonk like Avi, John or Matt, but it worked :-) [[Sam Korn]] ^(smoddy) 19:24, 29 April 2009 (UTC) reply

For REALLY heavy-duty data manipulation, I suppose I should stump for the power of DataStage, which can process hundreds of millions of records in applications such as making sure every single package delivered by a particular package company went to the right place, or that every second of cellphone time for a country's national phone system vendor was properly charged, and which does it in a linearly scalable parallel fashion. But I won't. ++ Lar: t/ c 19:32, 29 April 2009 (UTC) reply

I just think its wonderful that we have so many people volunteering to run in the audit committee elections so we can have truly powerful next-generation reports. Thatcher 20:32, 29 April 2009 (UTC) reply

You may not want the more prolific CUs to become audit committee members :). Anyway, we would loveto have you back in the trenches, Thatcher, so once your term is up, we expect YOU to come back to active duty . -- Avi ( talk) 15:37, 1 May 2009 (UTC) reply

I'd like to propose that the audit subcommittee suggest that CU's file reports more often, when it leads to a block. I've noticed that a few CU's already do this, and believe it to be beneficial. Of course, there are checks ran that have lead to no action, or checks that would constitute as privacy concerns, and these are not the types of reports I am asking be filed. Basically, if a check is ran independent of an on wiki request, and it has lead to a block, I would like to see it filed and archived. It would help to alleviate confusion if SPI has a case to look back on in times such as unblock requests and you just can't seem to locate 1) the CU to reconfirm their findings, 2) another CU to reconfirm those findings or 3) the CU has either retired or had their tools relinquished. Also, some CU's, and admins who block per them in private, do not tag properly. This can lead to confusion as to if the account blocked is a sock (of who?) or is a master. I believe it would also aid admins and non admins in filing future cases, where behavioral history is often key to determing an obvious sock. Best. Syn ergy 00:10, 2 May 2009 (UTC) reply

This discussion might be more appropriate at WT:SPI or WT:CHECK. Tiptoety ^talk 23:33, 2 May 2009 (UTC) reply

I specifically asked the audit subcommittee to answer. Syn ergy 21:53, 10 May 2009 (UTC) reply

I'm not sure that's an audit matter. I think for various reasons that when private checks are run and an admin blocks an account, the admin should say in the block message, "per checkuser findings by CheckUser:Smith", so the community knows what is going on. But that may be more an admin "best practices" matter than a checkuser or audit matter. Or rather, the issue of whether private checks are being run for appropriate reasons and documented internally is a matter for audit, but how the admin uses those findings may be more of an admin matter. Thatcher 23:23, 10 May 2009 (UTC) reply

After reading over the Wikipedia:Arbitration Committee/Audit Subcommittee, I have yet to see anything that states when a Auditer may use CU/OV. While I am of the assumption that they are only to be used during an investigation, I think it would be a good idea to have it in writing that members of the subcommittee should not be routinely using such tools. ie: for SPI or OV email requests. Thoughts? Tiptoety ^talk 23:32, 2 May 2009 (UTC) reply

The description on my userpage should give you a rough idea of where I stand, and I believe the other subcommittee members feel similarly (but have no desire to waste precious userspace)-- Tznkai ( talk) 03:32, 3 May 2009 (UTC) reply

I think that those conditions sound about right. I would like to see them (or something similar) a bit more set in stone and agreed upon by the subcommittee as a whole. Tiptoety ^talk 03:42, 3 May 2009 (UTC) reply

I do not intend to use the tools except for subcommittee business. If you want binding rules you will have to take it up with Arbcom. Thatcher 04:07, 3 May 2009 (UTC) reply

Okay, I can do that. Tiptoety ^talk 04:20, 3 May 2009 (UTC) reply

I am a bit curious as to why the subcommittee is performing "rechecks" for the WP:BASC. Seeing as such a check may be the subject of an investigation I feel that auditors should attempt to refrain from using CheckUser at all, unless responding to to a complaint. Instead, it would make more sense for the BASC to perform their own rechecks, or as for other Arbitrators or current Checkuser's opinions. Tiptoety ^talk 23:37, 2 May 2009 (UTC) reply

We were asked by an Arbitrator. Take it up with them. Thatcher 04:08, 3 May 2009 (UTC) reply

I'd like to know how things are going with this. When will these situations be resolved, and if these examples still have these flags, why? Syn ergy 21:49, 10 May 2009 (UTC) reply

I have not brought these up formally, and I'm not sure whether I should. I'll try and start a discussion about it. Thatcher 23:17, 10 May 2009 (UTC) reply

Thank you. I'm aware that a number of editors (mostly admins) already know the names of the cu examples. Basically, this is "somewhat" of an issue that is common knowledge, but I'm hoping formal discussion will begin, and sanctions issued. Regards. Syn ergy 23:59, 10 May 2009 (UTC) reply

I must admit I don't recognize anyone ;) However I fully agree with Thatcher's essay. I hope the AUDSUB will be able to raise to the challenge -- Luk ^talk 08:16, 14 May 2009 (UTC) reply

Statistics

Will these be the stats used to decide who will have the oversight tools removed, or are those included in the next batch? Tiptoety ^talk 23:43, 13 May 2009 (UTC) reply

We are notifying users with low activity from this list. FloNight ♥♥♥ 00:16, 14 May 2009 (UTC) reply

The next analysis will be of oversight-L (who responds, what the delays are like, any structural problems, etc.). Obviously, that will include the active oversighters, but it will only be a subset of actives, since there may be oversight actions that are requested by other means (talk page, ANI, email, IRC, chat, carrier pigeon, etc). There may be users who are not active on oversight-L but who are active in other ways. Thatcher 00:22, 14 May 2009 (UTC) reply

Please don't forget the hamsters who tirelessly carry messages over teh intertubes. -- Avi ( talk) 00:58, 14 May 2009 (UTC) reply

Indeed. The hamsters and carrier pigeons have been most helpful, although my bird seed bill has tripled and my family wonders why the house smells of cedar chips. ;-)

More seriously, the Audit Subcommittee has done some really excellent work here, gathering the statistical information, accompanied by some thoughtful analysis, to help the Arbitration Committee get an informed sense of the use of both checkuser and oversight/revision deletion tools. We have been able to take first steps from the information to date, and I for one eagerly await the analysis of response times and usage of the Oversight-L mailing list. Many thanks to Thatcher for doing a lot of this work and for posting it for community review, and to the entire Subcommittee, which has been working diligently since its inception. Risker ( talk) 02:18, 14 May 2009 (UTC) reply

I second Risker's accolades; well done, all! -- Avi ( talk) 15:28, 14 May 2009 (UTC) reply

Thanks for setting this up, particularly to Thatcher who's been quite tireless as far as I can see. It's not in the form I'd prefer (still rather opaque and still 'by appointment of Arbcom' rather than fiercely independent) but I can live with it for now. What is the current thinking regarding elections? When and in what form? -- Joopercoopers ( talk) 00:42, 17 June 2009 (UTC) reply

• According to the Arbcom agenda posted here, they are discussing an election mechanism and hope to have it in place by August 15. Further questions about elections should probably directed to the talk page there. Thatcher 20:27, 17 June 2009 (UTC) reply

• Could someone give me a little background as to how this has been established? I'm up to speed as far as community discussion regarding a Review Board in about Feb of this year. Am I correct in assuming those initiatives failed due to lack of consensus and this has been set up on a 'top down' basis from ArbCom? Thanks -- Joopercoopers ( talk) 12:11, 18 June 2009 (UTC) reply

That's about it. The original announcement is here and discussion of that announcement is here. Thatcher 12:22, 18 June 2009 (UTC) reply

Thanks for your prompt replies. I note and agree with Tznkai's concerns regarding the 3-3 split. My preference would be to see an entirely elected subcommittee who do not hold other posts, but at minimum, the make up should give a majority interest to elected members. -- Joopercoopers ( talk) 12:37, 18 June 2009 (UTC) reply

Report

• Question: The report lists Jimbo Wales as Staff, however, he is not listed at wmf:Staff or m:Special:GlobalUsers/Staff and was granted Oversight by Arbcom ( diff). Could someone explain this a bit more to me? MBisanz ^talk 12:34, 7 July 2009 (UTC) reply
  • Editors such as (to pick a random example) Dominic, are granted oversight and/or checkuser with the expectation that they will use these tools for the betterment of Wikipedia, and are subject to removal of permission after one year for inactivity. Other types of editors (Arbitrators, Auditors) have the tools but are not expected to use them often, and are not subject to removal for inactivity if the tools are not used, at least not until their "term of office" ends. Developers have the tools for development and testing purposes. Staff, like Cary, have the tools for use at their convenience, when dealing with Office or Foundation matters, or when they want faster action than contacting another user. Jimbo seems to fall into this last category. I added the labels in the first place because I wanted to deter agitation over "Why do Smith and Jones have Oversight permission if they never use it?" which for some users is a fair question but for others is not really relevant. Thatcher 12:54, 7 July 2009 (UTC) reply

We are waiting with baited breath. -- Avi ( talk) 15:23, 21 July 2009 (UTC) reply

• June update to what? Thatcher 14:08, 27 July 2009 (UTC) reply
  • Checkuser and OS stats, of course :) -- Avi ( talk) 14:50, 27 July 2009 (UTC) reply
    • Now that someone has explained pivot tables to me, and I am back from vacation, I can probably do it in a few days. Thatcher 14:52, 27 July 2009 (UTC) reply
  • Pivot tables are fantastic, I agree. Throw in some V/H lookups, a data table or two, and some MATCH/INDEX/OFFSET/INDIRECT calls, and now you're talking :D Welcome back, BTW. -- Avi ( talk) 14:58, 27 July 2009 (UTC) reply

<blank look/> Thatcher 18:03, 27 July 2009 (UTC) reply

• Oh yeah! Go pivot tables. MBisanz ^talk 17:45, 27 July 2009 (UTC) reply

Here and here. Thatcher 18:06, 27 July 2009 (UTC) reply

Thatcher, you are most definitely, "da man". -- Avi ( talk) 18:26, 27 July 2009 (UTC) reply

The Audit Subcommittee is a good compromise in the interim, but in the longer term, I'm sure there would be a strong benefit in creating an independent committee entirely dedicated to the overseeing of the CU and OS permissions. The number in checkusers and oversighters steadily increases, so the need for a stronger overseeing of those permissions. It would also permit for the Arbitration Committee to refocus on dispute resolution, and reduce the workload for arbitrators, while simultaneously making sure a special attention is given to the matter (and for other permissions similarly requiring privacy if we use some in the future, such as local stewardship). We could select auditors at annual elections, and maybe final appointments would be made by Jimbo, like with ArbCom. Specifically, the Audit Committee would have all, some or more of the following responsibilities:

1. investigate complaints concerning the use of the checkuser or oversight tools or alleged breaches of privacy thereof, report to the Arbitration Committee when the investigation is concluded, as described in the procedure
2. remove a checkuser or oversight access as a result of community-agreed processes (such as reconfirmation if we use this) or due to prolonged inactivity
3. hear requests for access and organize regular elections for checkusers and oversighters (in February and August, per precedents ?), subsequently appoint successful candidates in the manner prescribed by policy (as of now, done by the ArbCom, it also makes sense to transfer this responsibility)
4. oversee the use of the oversight and checkuser tools, monitor the checkuser and oversight logs
5. regularly report the committee's anonymized findings to the community, including statistics, discuss and suggest improvements
6. advise (through email) checkusers and oversighters on best practices, point out possible improvements in their use of the tools (such as: make sure to appropriately justify CU and OS actions with informative summaries, see those findings)
7. oversee the use of the functionaries-en mailing list.

The committees should be independent, first I don't see why it needs to be a subcommittee of a committee primarily aimed to resolve disputes, it's not the case anyway since it contains non-arbitrators. A relation of subsidiary to the Arbitration Committee could also hinder the work of AudCom, since they may be reluctant to advise an arbitrator on their use of the tools, and an independence between investigators, regulators or auditors and those 'sentencing' is the norm in the 'real world', for plenty of reasons (due process, etc); and also simply because their roles are, even if historically tied, very much different. Of course, the Audit Committee would communicate with the Arbitration Committee, when concluding investigations, signaling an abuse or proceeding other ArbCom requests, they can do so by email. I don't think AudCom should contain arbitrators specially appointed either, instead individual arbitrators may also seek election for the Audit Committee, so the community would decide if they want or not that the user posses those two hats. This would also change the appointment method for Checkusers and Oversighters, but only because the role of ArbCom in this matter would become the role of AudCom. Cenarium ( talk) 22:10, 1 September 2009 (UTC) reply

Hi Cenarium - I think you have some really good ideas in your proposal above; indeed, a fair number of them are already in place in some form or other (your points 1, 4, 5 and 6). There are a few things I need to caution you about. The first is that access to the Checkuser and Oversight permissions has to be within the applicable Wikimedia Foundation policies ( m:CheckUser policy, m:Oversight); this is not optional, because both of these permissions are operative under the WMF privacy policy. The WMF policy on access to Oversight permission does not explicitly state that Oversight appointments on projects with Arbitration Committees must be made by the Arbitration Committee; however, it does require that an arbitrator post the request for access. CheckUser policy is explicit about the subject of appointments on projects that have an Arbitration Committee:

“

On wikis with an Arbitration Committee (ArbCom) whose members have been elected with the support of at least 25-30 members of the local community, CheckUsers can be appointed by the Arbitrators only. After agreement, a member of the Committee should simply list the candidate on Steward requests/Permissions.

”

The Foundation's policies reflect the fiduciary duty that the Arbitration Committee holds in supporting the appointments to these two permissions. The Arbitration Committee as a whole takes this fiduciary duty very seriously. I would like to think that the hybrid process that we have developed, where Arbcom meets its fiduciary duty by vetting candidates in advance, and then putting their names forward to the community for the community to make the final decision, has found a middle ground where we all share in making the decision. (The alternative, where Arbcom vets people after a community election, is brutally unfair to everyone.)

Both the WMF Oversight and Checkuser policies specify that the Arbitration Committee (on projects where one exists) makes decisions on removal of access (your point #2). This is precisely the reason that the Audit Subcommittee is a subcommittee of the Arbitration Committee. The previously mentioned fiduciary duty is why arbitrators sit on the AUSC; if there are problems with the way someone uses these tools, the Arbitration Committee has a responsibility to act.

One last "caution" and that is that the Functionaries-L mailing list also includes former arbitrators who do not hold either checkuser or oversight permissions, and all of the current sitting arbitrators. However, speaking as one of the current list administrators, I'd be more than happy if someone else took over that role. The Oversight-L list is primarily moderated by non-arbitrator oversighters, and the Checkuser-L list is a multi-project list moderated by checkusers from various projects.

Now about elections: I really like your idea of scheduled, twice-annual elections, although I'd be inclined to do March/October (so that they don't overlap any other project or foundation elections), and I would have no problem with the AUSC overseeing it. I was planning on starting a discussion in a couple of weeks to figure out how to elect community representatives to the AUSC, but it might as well start now. In the discussion about secret balloting for the next Arbitration Committee election, some of us proposed that an AUSC election would be a good place to trial the process; what do you think? Do remember that fiduciary duty I mentioned before, though, and consider how that can be worked into the equation. Risker ( talk) 01:01, 2 September 2009 (UTC) reply

Yes, of course we should make sure to satisfy the wmf policy requirements. If there is a provision that in projects with an Arbitration Committee, only them can appoint checkusers or oversighters, then the Audit Committee could submit the final election results to the Arbitration Committee which would need to validate those, so it would be a process involving both committees and the community. Although we could ask the foundation if they would accept to allow the Audit Committee to appoint candidates, I think they would if the members are selected with requirements similar to enwiki arbitrators, which is the case as proposed. On removal of access, the policy states "On wikis with an approved ArbCom, the ArbCom can decide on the removal of access. On wikis without an approved ArbCom, the community can vote removal of access.", so I don't think it precludes that another entity mandated to do so by the community remove a CU or OS access. But there again, we could ask the foundation for clarifications. Considering this, I don't see what could bar us from having an independent committee with this role. Also, even in its current form, it's not a subcommittee of ArbCom, since it includes non-arbs. Cenarium ( talk) 01:49, 2 September 2009 (UTC) reply

The most obvious problem is that pointed out by Risker - the Foundation explicitly grants to Arbitration Committees the role and authority to oversee, appoint and remove checkusers. Altering this system would require changes at meta, and very likely the direct approval of Foundation staff.

A more significant tick against this proposal, in my mind, is that it doesn't meet my loose criteria for the expansion or establishment of bureaucracy. Complex and bureaucratic systems privilege users who are experienced in the political machines of Wikipedia over other users, and this gulf is most obvious between long-term editors and those new to the community. As a result, any new institutions of bureaucratic authority ought to pass a fairly high bar before they are implemented. A rough outline of the minimum criteria for new endeavors might be:

1. A current and significant actual failure on the part of existing systems to handle the responsibilities proposed for the new mechanism
2. The new mechanism must offer significant benefits in efficiency, transparency and efficacy over existing systems
3. The benefits outlined for the second criteria must outweigh by a significant margin any downsides inherent in the proposed mechanism

In my opinion, while Cenarium's proposal appears to offer benefits associated with institutional independence it fails to meet any of the three criteria I've described above. I would predict that a new, independent audit committee would be composed of editors not sufficiently and consistently active, handle a relatively minimal volume of problems, create the strange situation where the committee that appoints checkusers and oversighters is subsequently prevented from practicing any oversight and most importantly fail to address any actual shortcoming in the current arrangement. Nathan ^T 01:19, 2 September 2009 (UTC) reply

Addressed your first points above. I don't see how it introduces new bureaucracy, this committee already exists, and the elections for the committee are also planned. In fact, it's not a major change from the current situation. Concerning your latest paragraph, as I said, arbitrators could run for the Audit Committee. We could even have arbitrators appointed to serve in the committee, if needs be, it wouldn't be entirely independent, but they could be viewed as liaisons. The Audit Subcommittee is as of now fairly inactive, so those new responsibilities would increase its workload, and make its existence even more justified. As for users selected in the committee, I think we have enough active users who could serve, it doesn't have to be a big committee, 8 members should be fine. AudCom would be precisely responsible for the oversight of the CU and OS permissions, and in case of misuse, would immediately report to ArbCom. In the current arrangement, they can't do much of those tasks so efficiently because of a lack of personnel specifically dedicated. Cenarium ( talk) 01:49, 2 September 2009 (UTC) reply

I'm not particular sanguine about expanding AUSC's responsibilities past CU/OS little-o-oversight and into general Func-L little-o-oversight. For those of you not aware, Func-l gets a fair amount of traffic, most of which is frankly not worth paying attention to. The rare times it actually matters, it either falls within privacy-related problems, already part of AUSC's jurisdiction, or general shenanigans which ArbCom can handle fine without additional involvement. Likewise, while I believe that AUSC is well placed for monitoring elections, I do not want AUSC at all in charge of scheduling elections - the competencies are too different. One is essentially a policy decision - one is a little-o-oversight function.

I disagree, if I understand correctly what you're suggesting, that there is some intrinsic flaw in the AUSC structure that prevents us from working efficiently and effectively. We suffer approximately the same time constraints everyone else does, and we're still trying to figure out what the correct boundaries of our role really is.

Finally, the substance of independence can be gained regardless of AUSC's formal relationship with the Arbitration committee. I have no particular objection to a formally independent committee, (in fact, I generally support a more active, independent role) but I'm not sure if its worth the effort.-- Tznkai ( talk) 04:31, 2 September 2009 (UTC) reply

Any inefficiency that exists is due to not everyone reading their emails as frequently as others and not having as much free time as others. (If we could gather the six of us in Conference Room B for an hour we could get a lot done, but I don't think we are even on the same hemisphere.) This problem will exist with any other form of committee. I'm not sure that an committee that is formally an Arbcom subcommittee and where 3 members are Arbs, will be less effective in getting someone's access removed when (and if) such a thing is recommended, than a completely independent committee. Otherwise, I sympathize with Nathan's "If it ain't broke..." philosophy. Changing the structure will require that some people spend some amount of time and energy that could be used to other purposes ( opportunity cost). Will the benefits exceed the cost? But I have no objections or strong feelings either way, and have already indicated that I won't be running for election, under whatever procedure it is held. Thatcher 05:02, 2 September 2009 (UTC) reply

AudCom wouldn't have to be alone in overseeing func-en, rather a shared responsibility, their work in this regard should be to make sure CU, OS and privacy related matters are properly handled. In regards to their role in CU/OS elections, if we resolve to organize elections for CU/OS every six months, which we should probably ask the community about, they won't have to schedule those. Their role there would be ArbCom's current role, pre-vetting, hearing private feedback, proposing candidacies, etc, and submit the final election results to ArbCom for validation.

In regards to the AUSC structure, I'm certain it works quite well, but maybe it wouldn't work so well if we were to add those new responsibilities (which I meant, I've been unclear), implying more work, since the 3 arbs would make a priority of their arbitration work, so couldn't invest that much time, and the 3 non-arbs wouldn't be enough people in my opinion. While if we have 8 members, and either we allow arbs to be elected for AudCom but no more than 2 or 3, either 2 or 3 arbs are appointed as liaison (and probably one ombudsman too), then we should have enough people.

Considering access removal, I don't think AudCom should make removals of its own (to preserve the purely 'Auditing' aspect), rather recommend them to ArbCom, and I also think they are a good choice to make procedural removals, either due to inactivity, or due to a failed reconfirmation if we ever use this. Cenarium ( talk) 16:54, 2 September 2009 (UTC) reply

I'm not sure I completely follow. You want to create the "AudCom" to replace to the AUSC, and to justify an expanded committee separate from the arbitration committee you'll transfer some (but not all) of the oversight role from ArbCom to AudCom. When asked if there is a current problem with the AUSC, you say no, but argue that the AUSC might not be able to fulfill its duties if new roles were transferred to them from ArbCom. So I'll ask you, is there then a problem with how ArbCom is managing its responsibilities with respect to Oversight and Checkuser? Otherwise, it strikes me as sort of a circular argument and I still don't see the justification for going through the effort of reconstituting the AUSC and partially redistributing authority from one committee to another. Nathan ^T 17:13, 2 September 2009 (UTC) reply

It's not a proposal to replace it, it would be an evolution of the committee currently know as AUSC. Let me clarify:

AUSC is, currently, not a subcommittee of ArbCom, since it contains members who are not arbitrators, so I think it should be renamed to Audit Committee, instead of Audit Subcommittee. It's nothing more than a rename to reflect a reality.

The only authority that would actually be transferred from ArbCom in this proposal are procedural removals, the pre-vetting of candidates and the overseeing of CU/OS elections. The main reasons for this are to reduce ArbCom's workload and because in my opinion, it fits in AUSC's prerogatives, while it doesn't fit so well in the prerogatives of a committee primarily dedicated to arbitrate disputes. The rest of the proposed responsibilities, Risker said AUSC already unofficially had them, more or less, or they are natural extensions. It doesn't remove ArbCom's authority in overseeing the use of the CU and OS tools, it makes of it a shared responsibility.

Then, I propose a relative independence between the committees, it's essentially a matter of appearance and selection of members, and wouldn't require too much changes. They would still work together, AudCom would report to AudCom when finding misuse, ArbCom could request investigations to AudCom, etc. We could have a few liaison agents between the committees, if desirable.

I don't think it would require a huge effort, we would have to organize elections, which is already the case in the current situation, although granted a little more complex that if only three members had to be chosen, to develop guidelines, maybe an audit policy, but we could need this even in the current situation and we have the basis for it, maybe rename their mailing list from arbcom-audit-enlists.wikimedia.org to audcom-enlists.wikimedia.org, and move a few pages. I may have missed some other things, please tell me if there are. Thanks, Cenarium ( talk) 19:20, 2 September 2009 (UTC) reply

The review board proposal attracted a great deal of comment from many people, and eventually stalled without receiving consensus, so Arbcom decided instead to delegate their direct and clear authority to monitor and investigate CU and OS to a subcommittee under rules and a policy that they formed. You propose to convert this to an independent review board with elections. It would seem to me that you need to go back to the community and get consensus for this, far more than the 5 people participating in this discussion. I don't think it will be as easy as you think it will be. Thatcher 19:33, 2 September 2009 (UTC) reply

I'll try to accomplish this step by step, the rename, the new responsibilities, the independence, the policy. The rename and the new responsibilities, we can decide without massive community participation, since the rename is simply to remove "sub", and the responsibilities are already possessed by ArbCom or natural extensions. The independence and the establishment of an audit policy would require more community participation, but I'm sure it's feasible, the situation has much changed since the review board proposal, and I think that proposal was plagued by its name and for being too bureaucratic (e.g. having CU re-access and OS access but not being allowed to use them). Cenarium ( talk) 20:44, 2 September 2009 (UTC) reply

(unindent) I'll separately propose to rename to Audit Committee and grant those new responsibilities, as well as raising to 5 the number of non-arbs in the committee to counterbalance the increase in workload. The rest, later. Cenarium ( talk) 02:16, 3 September 2009 (UTC) reply

Never mind for the rename now, if they are at-large members, this can be considered as a subcommittee. Cenarium ( talk) 02:44, 3 September 2009 (UTC) reply

For clarity of discussion, I propose to discuss AUSC responsibilities in this new section. The only official responsibility of AUSC for now is to investigate complaints concerning the use of the checkuser or oversight tools, and report to the Arbitration Committee when the investigation is concluded, as described in the procedure.

I propose to clarify that investigating alleged breaches of privacy thereof are also within AUSC's scope. There are other responsibilities naturally extending those, consistent with a role of auditing and more or less already in place:

1. oversee the use of the oversight and checkuser tools by monitoring the checkuser and oversight logs
2. regularly report the AUSC's anonymized findings to the community, including statistics (already done a few times), discuss and suggest improvements in the of CU and OS operations
3. advise (through email) checkusers and oversighters on best practices, point out possible improvements in their use of the tools (such as: make sure to appropriately justify CU and OS actions with informative summaries, see those findings)
4. verify that CU, OS and privacy related matters are properly handled in the functionaries-en mailing list

They seem desirable as CU and OS actions are not subject to the usual community review due to the required privacy, and ArbCom alone may not have enough time to invest in this task, while there are definitely checkusers and oversighters who could do so, and since the beginning, they are encouraged to review each other's actions, but a committee specially dedicated to this task would be more efficient.

I also propose to make AUSC responsible for procedural removals, that is, due to an inactivity of over a year (ArbCom passed a ruling on this) and at the conclusion of community-agreed processes, such as failed reconfirmations if we ever do this. It would save time for ArbCom and mark the procedural nature of those, removal of rights by ArbCom are more connoted as sanction, even if they are not; it's unecessary here.

Regarding the selection of checkusers and oversighters, I propose to transfer share ArbCom's role in receiving requests and pre-vetting candidates to with the Audit Subcommittee, which would also oversee elections and submit the final results to the Arbitration Committee for validation. It's also to save ArbCom's time and because it fits to the AUSC's role, though they could also do this in consultation with each other. As to when elections should be organized, it's been proposed to do them every six month, so ArbCom won't have to pronounce on when they should be done, it's a reasonable timeframe and we had the first two CU/OS elections in February 09 and August 09.

Since those new responsibilities imply more work for the Audit Subcommittee, I propose that we use 5 at-large members instead of 3. I'd like to seek comments from the community and the committees on whether those changes are desirable, so as to form a consensus in some way. Cenarium ( talk) 03:33, 3 September 2009 (UTC) reply

Cenarium, it is clear that I have not done an adequate job of explaining that granting and removal of checkuser and oversight permissions is a core and fiduciary duty of the Arbitration Committee. The Wikimedia Foundation has specified that the responsibilities for appointing Checkusers and Oversighters, as well as authorizing removal of these permissions for cause, lies within the sole remit of the Arbitration Committee on any project that has such a committee. It is not a right of the Committee, it is a formal and very serious responsibility. Speaking personally, I believe I would be failing in my duty not to vet candidates for these roles. The community explicitly elects arbitrators knowing that this is amongst our responsibilities. If you wish to change WMF policy, please do so at Meta. Risker ( talk) 04:02, 3 September 2009 (UTC) reply

This is only a particular aspect of the proposal, but I don't believe what I propose hinders that. The overseeing of the elections, the pre-vetting of candidates, may be conducted by both the Arbitration Committee and the Audit Subcommittee. I don't see in any documentation evidence that the WMF wouldn't authorize a subcommittee of ArbCom to request procedural removals or rights, either due to one-year inactivity or a community vote, the arbs in the subcommittee could request those themselves at meta (the wmf policy states that one arbitrator requesting it is sufficient). The role of the other members would simply to point out checkusers and oversighters who didn't use their rights for more than one year. As for pre-vetting, AUSC could first receive applications, then transmit those to ArbCom and give their opinions, then ArbCom would also do their pre-vetting. Cenarium ( talk) 04:28, 3 September 2009 (UTC) reply

Corrected the phrasing to clarify this. Cenarium ( talk) 04:54, 3 September 2009 (UTC) reply

So in other words, we will now have two groups vetting the same candidates, and then the community "vets" them by their votes? Perhaps, before jumping to conclusions about how much work elections are, you might want to ask some questions about it; after all, I coordinated the last one and would be happy to share the information. As part of the last election, a significant number of standard "forms" were created, which will be re-used in future elections, in particular the CU/OS vetting questionnaire, but also standardised emails for various stages of the elections (respond to receipt of request for questionnaire, confirm receipt of questionnaire, offer of nomination following vetting, declining opportunity for nomination after vetting, confirming appointment and providing instructions for identification following successful election, thanking unsuccessful nominees, etc.). This is work that does not need to be redone. I agree with you that standard, twice yearly elections at set periods is a good idea, and recommend March and October so that they will not conflict with any other WMF or Wikipedia elections.

Incidentally, the AUSC already keeps track of the inactivity period as part of its routine statistics, and can certainly flag those who have gone a year without use. As I recall, the first time around an arbitrator member of the AUSC was the one who facilitated the permission removals. There are no checkusers or oversighters on the current list who have gone that length of time without using their permissions.

There is one thing that you've neglected to cover in all of this: how community representatives to the subcommittee should be appointed/elected. This is probably a much more important concern than trying to expand their portfolio. Risker ( talk) 05:07, 3 September 2009 (UTC) reply

There's no difficulty in the selection of at-large members, as in my opinion they should already be checkusers or oversighters, maybe both, for at least six months, because we expect experienced users in this position, and otherwise the committee wouldn't be so efficient. Though at the moment, we wouldn't have enough people if we required both, so we should probably require CU, because it's viewed as the right requiring the most trust, and if we still haven't enough people, CU or OS, and progressively rise the bar. It considerably simplifies the selection. They should also not be arbs of course (maybe not past arbs either), and then they just have to propose their candidacy, and there's the election. We just have to choose the voting method and how we appoint candidates (either it's clear-cut, either ArbCom or Jimbo completes them).

As for duplicating work, I expect the two committees will find a procedure to avoid this. AudCom could make the administrative tasks for example, but ArbCom would review; well, how you see fit.

I also want to come back on the question of the responsibility to oversee, grant and remove CU/OS permissions, and WMF policy. The scope of Audit Committees should stay within the borders of auditing, so anyway they wouldn't be appropriate for removing CU/OS rights outside procedure, as they don't sanction, nor be entirely responsible for granting them, as they don't hire in RL situations. But Wikimedia foundation policies are not set in stone, and they can grant exceptions. So it is possible that in the future, another entity elected by the community, be also permitted to remove CU/OS rights, or add them. Cenarium ( talk) 01:24, 5 September 2009 (UTC) reply

I'd imagine they should already be doing #1, so no quibbles, there, at first glance. #4 likewise seems to be a reasonable extension of the current mission. As far as #2, I'm not sure how that would differ from the current scenario, except for the appended "discuss and suggest improvements" bit -- I'm sure any such suggestions would be taken into account, but please do bear in mind that discussing amongst functionaries and with the community at large is already frequent, and I personally would prefer that we avoid any scenario where some suggestions are "more official", as it necessarily implies that other suggestions are "less official" (I suppose that also goes for #3). Election procedures should probably be discussed separately. – Luna Santin ( talk) 07:17, 3 September 2009 (UTC) reply

I agree, I don't think we should, within Wikipedia, introduce too much 'officialism', and this goes for AUSC suggestions and reports. Although I used the term, it was in the meaning to mark it as such on the description page, so as to recognize this aspect of AUSC. Cenarium ( talk) 16:22, 3 September 2009 (UTC) reply

• Um... process creep much? Stifle ( talk) 08:17, 3 September 2009 (UTC) reply

Audit already does #1, 2 and 3. I suppose it's implied by the charter even if not specifically stated. Thatcher 15:40, 3 September 2009 (UTC) reply

Above Risker points out that we should address the method of election for community representatives to the AUSC; I agree, and I think we should also discuss whether there ought to be a limit to the number of people with access to the CU/OS tools. We don't limit the number of administrators, and as far as I know we haven't had specific limits to checkusers/oversighters in the past. On the other hand, there was a de facto limiting process in that the ArbCom only sought new people for these tools when there was an apparent need. With twice annual elections where all candidates receiving more than 70% support are appointed we have essentially removed any limiter to the number of users with tool access. Is this wise? Should it be "twice annual elections up to X number of active users"? Boiling it down:

1. Should we have more checkusers and oversighters than we need to manage the workload?
2. Should inactive users keep tool access?
3. Should elections (or seats available per election) be limited by workload needs? (partial restatement of 1)
4. Should users who lose access due to inactivity or resign voluntarily regain access on request or seek election?

I think answers to these questions are more important than rearranging committee relationships. Perhaps the answers here are known and I just haven't seen them listed/announced? Nathan ^T 16:41, 3 September 2009 (UTC) reply

I think those are important questions, and AFAIK, open. For AUSC elections see my reply to Risker. This pertains to the diffusion of private information, which should be tightly controlled. This is also one of my problem with functionary-en, whose membership is not enough circumvented imo, and why I request that AUSC oversight of func-en for privacy/etc.

1. No, but this is subjective in many ways.
2. No, it's already wmf policy that inactive users for more than one year with the tool should have it removed, although it's unclear if it's completely inactive or just not using the tool. In any case, there's been an arbcom motion that users who didn't use the tools for more than one year should have it removed. For my part, I would reduce the inactivity period to six months.
3. Maybe, fixing the number of appointed users to three by position by default, and more if necessary.
4. ArbCom didn't clarify how those users could request access back when announcing the motion, I'd like a clarification. I would suggest a restoration on request for six months, then it's by election only, though I'd prefer as I said 6 months of tool inactivity, it's already considerable.

There's also the possibility to make reconfirmations, like stewards on meta, to control the pool of users with access, including past arbs (the clock starting when they leave office), every one year or two years. I'd organize them at the same time as elections. Cenarium ( talk) 01:24, 5 September 2009 (UTC) reply

While it would require more study in depth to come to a firm conclusion, at first sight the idea for CU/OS reconfirmation elections seems reasonable. Elections for new CU/OS privileges could be held in conjuction with these (and, potentially, the desired election for community representatives to AUSC could be held in conjuction as well). — Matheuler 02:08, 5 September 2009 (UTC) reply

I think perhaps we are getting a bunch of separate principles too intertwined in some of these concepts; we started out talking about the role of this (AUSC) subcommittee, and whether or not it should actively be involved in CU/OS elections. The comments above are more closely related to policies for promotion, retention and removal of CU/OS privileges separate from any concerns about the actions of Checkusers and Oversighters, which isn't in the terms of reference for the AUSC.

Cenarium, I do understand your concerns about access to private information. I can tell you that three oversighters will never, ever be sufficient, nor three checkusers. Oversight requests come in 24/7/365, and we are striving to have those requests addressed within a few hours; the longer that an oversightable edit is visible (even as a deleted revision, which any admin can see), the greater the privacy violation. Similarly, checkusers perform thousands of checks a month, and without making it a full time, paid position, there is no way it could be limited to three users. The Arbitration Committee is working to try to reduce the amount of time its members spend carrying out routine checkuser and oversight activity in the hope of reducing workload and, we hope, burnout; however, we routinely deal with situations that involve checkuser data or potentially oversighted information, and we retain final responsibility for ensuring the tools are used properly, so Arbcom members generally need access to the tools, and a degree of competency in their use. As to the Functionaries-L mailing list, only rarely will very specific checkuser information be discussed there (more often it is a discussion of general principles or how to best manage a specific case or request assistance in doing so), only general principles about oversight are discussed (specific cases are discussed on the oversight-L list), and access is closed to checkusers, oversighters, current arbitrators, and a handful of former arbitrators, many of whom continue to do regular checkuser and oversight work. We've done some major pruning of both checkuser and oversight access, despite the new members. Risker ( talk) 02:35, 5 September 2009 (UTC) reply

Misunderstanding, I talked about the number of users who would be appointed by default in an election as long as they have 70%, which Nathan referred to, and which could be risen by ArbCom in case there's a need, not of the total of users with access. If I remember correctly, three CUs have been appointed in August and February, so it makes sense. Of course, ArbCom needs access to the tools. I think I know what I'm talking about, otherwise I wouldn't have come here :) (and I proposed a committee with 8 CU/OS, and in a second time 5 at-large members for AUSC). As for func-en, it's especially for former arbs who don't have CU or OS access I thought. While I trust those in this position, I don't see how their access is justified. Since it involves access to CU/OS information, even if rarely, this should be considered a CU/OS permission. And here comes AUSC, too. I think all this is linked and should be discussed together, with at the end specific proposals. Cenarium ( talk) 03:01, 5 September 2009 (UTC) reply

OK, on the results end, what would be different from what we have now? What tangible benefits are you aiming to create? I am still currently mystified as the content, purpose, and predicted results of what we're talking about? We've got plenty of text here, but I don't think there is a clear point yet.-- Tznkai ( talk) 18:18, 5 September 2009 (UTC) reply

(unindent) Clarify the scope of AUSC. There are a few responsibilities that AUSC informally holds, I suggested a few others that may also be delegated to AUSC, after that it's your's decision. Now that some time has past, I would like the clarification from ArbCom/AUSC on whether they intend to recognize that AUSC has the responsibilities I initially stated:

1. oversee the use of the oversight and checkuser tools by monitoring the checkuser and oversight logs
2. regularly report the AUSC's anonymized findings to the community, including statistics (already done a few times), discuss and suggest improvements in the of CU and OS operations
3. advise (through email) checkusers and oversighters on best practices, point out possible improvements in their use of the tools (such as: make sure to appropriately justify CU and OS actions with informative summaries, see those findings)
4. verify that CU, OS and privacy related matters are properly handled in the functionaries-en mailing list

And if yes, then document this on the page.

As for AUSC elections, I heard they would be planned soon, so I'll open a specific thread on those.

CU/OS elections are not so urging. Cenarium ( talk) 22:44, 2 October 2009 (UTC) reply

As the election comes closer, the actual role and responsibilities of AUSC, both formal and as of now informal, should be clarified to the community. (email to AUSC sent) Cenarium ( talk) 00:48, 10 October 2009 (UTC) reply

I just noticed that User:Azafred has had checkuser rights for a week now for the purpose of doing lookups for the WMF legal counsel. However, he was granted the checkuser right by staff member/sysadmin Tfinc. Obviously staff members and sysadmins can do whatever the Foundation says they can do, both under common sense and the WP:GRU. My question is, why wasn't he granted global checkuser or staff rights, since generally that is how the WMF grants things to employees to avoid confusion at the local level. Also, why was he granted it in the local log instead of the Meta log? Was the AUSC informed or are they aware of it and are monitoring it? I noticed he hasn't been listed at WP:CU, so I figured I would ask. MBisanz ^talk 23:54, 9 September 2009 (UTC) reply

We were not informed, and I was not aware of it until now, and havent seen any public disclosure of who Azafred is. I've asked Tfinc to respond here. John Vandenberg ^{( chat)} 04:14, 10 September 2009 (UTC) reply

tfinc here, aZaFred is an employee of the foundation like myself and we routinely have to use the check user tool in order to comply with subpeona requests. His involvement in this would have been no different then myself or brion fielding this request. My choice to grant him the check user right was simply due to allowing him to be granted the limited privilege level that was required to get the task done. If it makes sense to the community to give him only the staff rights instead of what was specifically needed then I'm more more then happy to switch him over. -- Tfinc ( talk) 05:14, 10 September 2009 (UTC) reply

Of course this leads to my next question of how closely AUSC is monitoring the checkusers if someone can be granted (and presumably use) the checkuser right for over a week without any of the six people charged with monitoring the tool noticing a redlinked username in the checkuser log/list? MBisanz ^talk 04:17, 10 September 2009 (UTC) reply

According to m:User:Azafred, he is a sysadmin for the WMF who was hired in March 2009. Why he wasn't simply given global staff/sysadmin rights like the other employees is my main question (other than what sort of self-regulating the other 38 people with checkuser rights and 6 people charged with monitoring checkusers and the committee of 14 people responsible for enforcing the policy have been doing for the last week). MBisanz ^talk 04:25, 10 September 2009 (UTC) reply

Well, it may be because he hasn't actually carried out any checkusers, so his name didn't show up in the logs at all; while I can't speak for others on the committee, I review portions of the logs a couple of times a week. I reviewed the Checkuser and Oversight rights lists on 30 August, and he wasn't on the list then. I will be patiently awaiting Tfinc's response, and assume that this was a simple error (i.e., direct granting on the wiki instead of the staff right) that can be corrected easily. Risker ( talk) 05:04, 10 September 2009 (UTC) reply

Honestly, it has never occurred to me to monitor the user rights list for rogue checkusers and oversighters. I guess I assumed that people won't be granted the rights outside of normal processes (i.e. the stewards are not going to promote someone without proper authorization). Azafred has 4 logged checks to date, and I can guess from the contribs that it involves a single BLP problem, so everything looks acceptable even if the process was a bit out of the ordinary. Thatcher 11:12, 10 September 2009 (UTC) reply

Well I certainly don't have an objection to Tfinc giving out userrights to employees or any checks Azafred did, I actually do not care what employees do with company resources since that is a management matter, not a community matter.

My point is that we are told time and time again that checkuser is rarely abused since all the checkusers are watching each other, so it would be very difficult for say, you to check someone you are in an edit war with, since someone else would reviewing the log and recognize the COI nature of your checks. But if a redlinked account that no one has ever heard of can make checks and no one flags it as odd, then I have to wonder what kind of self-regulation is going on. I also understand Risker's position that each person cannot review ever check made, but for 38 people to fail to notice those four checks on their random samplings would seem to indicate that the odds of any rogue checkuser being caught for doing something improper are very very low. MBisanz ^talk 12:42, 10 September 2009 (UTC) reply

Ok, I have edited my user page to make sure my Account doesn't show up as a red link anymore. I did ask for the right to run checkuser at the request of our Legal Council. Tomasz was nice enough to grant me the right so that I could gather the information required by a subpoena we received and which was time sensitive. Hope that clears some of the issues. Azafred ( talk) 16:13, 11 September 2009 (UTC) reply

For the record, I have replaced Fred's checkuser flag with the global staff flag. -- Bastique ^demandez 16:35, 11 September 2009 (UTC) (for User:Cary Bass) reply

Okay. That works fine. Thanks. FloNight ♥♥♥ 17:38, 11 September 2009 (UTC) reply

Any news on an update for August and September? Nishkid64 _{( Make articles, not wikidrama)} 19:28, 3 October 2009 (UTC) reply

Frankly its not a high priority for me right now, but I'll keep it in mind for when I have some extra free time. Thatcher 00:11, 5 October 2009 (UTC) reply

Take your time. I'd just like to see how often the new CUs/oversighters are using the tools. Nishkid64 _{( Make articles, not wikidrama)} 03:41, 5 October 2009 (UTC) reply

The process to appoint the three non-arbitrator members of the Audit Subcommittee is underway, with the election itself starting on 30 October. If you think you may be suitably qualified, please see the election pages for the job specification and application arrangements. Applications close 22 October 2009.

For the Arbitration Committee,   Roger Davies ^talk 21:38, 7 October 2009 (UTC) reply

Starting a section here for the AUSC October 2009 election and discussion pertaining to it. I'm currently trying to find a page that describes SecurePoll. The best I can find is mw:Extension:SecurePoll. Carcharoth ( talk) 23:29, 7 October 2009 (UTC) reply

I note that prospective candidates are asked to email privileges2009-at-googlemail.com - a non-WMF address. Might one ask why a secure/WMF-controlled address has not been used for what is, quite clearly, a committee operating under the aegis of the WMF. The use of non-WMF email addresses for 'official' business could make it appear that there is something, ahem, 'fishy' going on and should be, imho, not accepted practice. -- AlisonW ( talk) 20:51, 11 October 2009 (UTC) reply

Neither the Arbitration Committee nor the Audit Subcommittee operate under the aegis of the Wikimedia Foundation. We are organs of English Wikipedia and are responsible to this project alone. The only overlap is that the Wikimedia Foundation grants the Arbitration Committee the responsibility of identifying suitable candidates for privacy-related permissions (CheckUser and Oversight). As we are not agents of the WMF (in fact, are quite clearly not), it is not appropriate to use a WMF (i.e. name AT wikimedia.org) email address for this purpose. Risker ( talk) 23:38, 11 October 2009 (UTC) reply

Might have made more sense, for the purposes of confidence on the part of the folks sending information to this address, to use the arbcom-l list address (which is, worth noting, a Wikimedia address) and route the posts to whatever method you have for handling permissions requests. Nathan ^T 17:25, 12 October 2009 (UTC) reply

I have to say I agree with the Alison and Nathan here. Arbcom uses WMF addresses for it's mailing list, so I don't see why it can't get one for this, too. A random GMail address certainly does not inspire confidence in those sending private data to it. Using a third party like this is not the best way to handle confidential information, either. — Jake Wartenberg 20:53, 15 October 2009 (UTC) reply

What confidential information are you talking about? That address is being used by whichever Arbitrator is coordinating the election. You send an email to that address saying "I want to be on the audit committee." The arbitrator sends you a questionnaire about technical and wiki background. Then you mail the questionnaire back and the arbitrator posts it on the private Arb wiki for the other arbitrators who will discuss whether to accept your candidacy. The only time that private information comes into issue is if your candidacy is approved, and if you are elected, then you must identify yourself formally to the Foundation, by whatever means the Foundation accepts (I faxed a copy of my drivers license to Cary Bass). I don't see what the fuss is about. Thatcher 21:20, 15 October 2009 (UTC) reply

I would hope that people's answers to the private questionnaire will be treated as confidential. This is not a huge issues, but a foundation email address would be more desirable. — Jake Wartenberg 23:54, 15 October 2009 (UTC) reply

Well, I will make inquiries of Cary Bass in this regard, but as this is not a WMF-initiated process, Arbcom does not report to the WMF, and none of the members of the Committee have any specific relationship with the WMF, I do not expect that the Foundation will be eager to hand out Wikimedia email addresses to us for any reason. There is a significant difference between a WMF-hosted mailing list and an email address that ends "@wikimedia.org". Risker ( talk) 00:04, 16 October 2009 (UTC) reply

If an @en.wikipedia.org address could be issued, then that would be ideal. Sadly, I think I've only ever seen @wikimedia.org. AGK 00:34, 18 October 2009 (UTC) reply

A sentence about Oversight reads, "although these are not physically moved but simply flagged as non-public items". Is "moved" a typo for "removed", or are oversighted revisions actually moved from one part of the database to another? Nyttend ( talk) 20:46, 14 October 2009 (UTC) reply

Typo. Thatcher 21:57, 14 October 2009 (UTC) reply

Actually, revisions oversighted using the old Oversight extension are physically moved from the revision table to a private data table. Whereas revisions suppressed using RevisionDelete are merely flagged in the revision table as being nonpublic. But in the context of this sentence, I guess it's all much of a muchness :D Happy‑ melon 22:32, 14 October 2009 (UTC) reply

There was a question asked off-wiki about whether AUSC members would keep the tools after their term. I would strongly urge not; in fact I would actually add a rule at this point: "An AUSC member is granted the tools for AUSC business only, and will be de-flagged at the end of their term of office. They may apply for the tools in the usual way no sooner than [6/9/12] months after their term ended."

Same reason I suggested a 12 month "gap" for arbs on voting at CUOS elections: this ensures there is an enforced break between the roles and helps enforce that AUSC is not expected to merge or flow into CUOS like some kind of shortcut. If they then want to stand, and people trust and value them, there is no reason they won't be appointed at some later time, but it ensures a break and encourages a fair and level field.

For those who are competent and desirous, a break of 6 - 12 months is fine - we won't lose out as a community, the division is useful; if they cannot wait that time, they are probably not good choices in the first place, and it allows better perspective.

FT2 ^{( Talk |  email)} 03:43, 15 October 2009 (UTC) reply

• I would object to such blanket language. It very much depends on whether the person held checkuser or oversight before joining Audit. I see no reason not to allow someone who held one or both privileges before joining Audit to resume freely using them after their term expires. A person who did not hold such privileges before election should could have them revoked when their term expires but I see no reason to bar them from standing in the next election, whenever that is. Elections are so irregularly scheduled that a 6 month limit could result in someone not being able to run for more than a year. Thatcher 11:17, 15 October 2009 (UTC) reply
• Making them sit out for an arbitrary period of time seems to be a poor use of skilled and knowledgeable users. There is acquired knowledge and skills about the use of the tools that the Community should take advantage of by keeping them on as CU and OS. I see no reason to stop an user that the Community supports in the role from keeping the tools. FloNight ♥♥♥♥ 11:35, 15 October 2009 (UTC) reply

As it stands now, Thatcher asked for a community confirmation of his use of OS in the last cycle, Mackensen is a CU, former Ombudsman, and former arb in good standing if I choose not to stand or am not re-elected if I do stand for election, I have already asked for the tools to be removed, and I would be as eligible as anyone else in the next election. It seems to be working fine.-- Tznkai ( talk) 19:11, 16 October 2009 (UTC) reply

• Comment I have put in an application to run for this position, in the non-admin role. I will say that I only expect to be allowed access to tools during the time that my position would run and once it expired I will happily turn them in. It just seems to me that if this is a term position then my tools should expire with the position. Being that this is a very sensitive position and important role, I think that having tight standards is erring on the side of caution. Basket of Puppies 19:46, 16 October 2009 (UTC) reply

The concern here is that AUSC should not be a springboard to tool flagging. AUSC members get the tools for audit purposes only; an enforced turning in and break from them may be appropriate to ensure that aAUSC membership and overshighter/checkuser-ship, are distinct. In general I would be disturbed to see a trend emerge of ex-AUSC members seeking to keep tools as a run-on from their term. I'd feel better if the two were distinct. Point of principle, rather than reflection on any specific individual(s). FT2 ^{( Talk |  email)} 07:45, 17 October 2009 (UTC) reply

• I'm not persuaded that there needs to be a distinct separation. And I'm certain that approval of the Community to retain the tools through an election overcomes any possible concerns. If the person has the time and interest now, and is willing to stand for re-election, then there is not any meaningful reason to turn them away for an arbitrary period of time.
• FT2, this is certainly not a new trend given that former arbs, such as yourself, have kept their tools when leaving ArbCom. In the past, people such as you kept the tools without seeking election to retain the tools. Now people elected for a position (AUSC) are asking the Community to confirm that they retain them. If you feel strongly about this matter, I suggest that you return your OS and CU access and run for re-election to regain them. (Full disclosure:I'm deciding now whether to retain the tools after I leave ArbCom or return them. If I return them and seek re-election later to retain them, I would be the first person to do so. Am not clear that this is something that I want to force on other arbs so I'm reluctant to start the trend. But if FT2 turns in his tools and seeks re-election to retain them, then I definitely will. ) FloNight ♥♥♥♥ 11:41, 17 October 2009 (UTC) reply
• My opinion, FWIW, is that I think that there is a difference, Flo. When electing someone to ArbCom, people know that part of the package is CU/OS, and that traditionally these tools are kept upon the term's end, so that is factored into people's voting decisions. The AUSC is 1) pretty much brand new without any traditions, and 2) specifically meant to be a "line-of-defense" as it were (oversight of oversight :) ), and not a group of people who are CU/OS-lite, engaged in the maintenance work that Arbcom, Arbcom-appointed, and Wiki-elected CU/OSers such as ourselves do. Therefore, it is much more reasonable to consider enforcing a mandatory release to prevent this from being a "back-door" to flags. Anyone who had the flags before, are exceptions, IMO, as they already legitimately had the flags, and their service on the AUSC should not affect their previous trusted status. Just my 2 cents. -- Avi ( talk) 16:47, 18 October 2009 (UTC) reply

Avi, now, maybe people know. But arbs getting CU & OS access and keeping them after their term ended evolved into a custom because Arbs did it. This was not something that the Community agreed should happen. And from reading Community discussions it is obvious to me that many people that elected arbs had/have no idea that this is part of the package. Personally, I'm not concerned about people keeping access to CU or OS after their term ends especially if they are standing for an election. I don't see what the concern is about someone being on AUSC first any more than someone having OS or CU access first before being on ArbCom or AUSC. The Community can look at the people running and decide if they want to have them in the role. FloNight ♥♥♥♥ 17:15, 18 October 2009 (UTC) reply

As I've stated before, I think only users with the CU or the OS permission, or both, should be allowed to stand for AUSC election. First because we need experienced people in this position, a user with no experience with those permissions on enwp will probably be of little use, at least initially; second, we also obviously can't judge their own use of the tools and related business if they haven't had them, I'd prefer that people standing for auditing have demonstrated a good use of the permissions. It would also remove the concern of tool retention and use for AudCom members (especially when we'll have enough people to require both permissions). We're different than real world organizations; they can hire auditors who have experience in a specific area, they can judge of their experience, etc; while we can only judge based on what users did on en.wikipedia, the role is also novel and peculiar to enwp, this is not auditing in the strict sense and should not be viewed as such; we should leave that to the WMF.

Concerning tools retention, I think regular reconfirmation, as with stewards on meta, would solve the problem nicely.

Also; I was disappointed to see that ArbCom didn't seek community feedback on how the election should be made, this should have been discussed beforehand. Cenarium ( talk) 21:47, 18 October 2009 (UTC) reply

Oddly enough I'm fairly certain there has already been discussion in a few places on how to proceed with AUSC elections - and running another poll to determine how to poll is the sort of thing satirical cartoons are made of. The fact is, AUSC matters don't attract a lot of wide community attention and input - we'll be lucky if we get a solid showing at the elections (and I hope we do).-- Tznkai ( talk) 00:42, 19 October 2009 (UTC) reply

There's been no significant public discussion I'm aware of. This is not strange or unusual at all, ArbCom specifically requested community feedback for the CU/OS elections, and significant changes ensued; we always discuss AC elections in advance. For the present election, there are points certainly deserving discussion, for example the election as described seems to be straight-to-voting. I think it should be in two phases, questions/debate for one week, then the proper one-week voting phase, or if in one phase, allow for debate on-wiki.

Indeed it doesn't seem to attract much community attention, but then proceeding with elections is putting the cart before the horse. We'll see how it turns out... Maybe the solution would be to broaden the scope, but it couldn't be realized as a subcommittee of ArbCom, but you'd need community support for this... So we're turning in circles, unless it becomes part of a greater plan. Cenarium ( talk) 00:54, 24 October 2009 (UTC) reply

(unindent) There's a difference. Until 2008 Arbcom began to open up CU/OS to significantly more non-arb users (Aug 08), almost all the active checkusers and oversighters were ex-arbs. Experienced, knowledgeable on the issues, trusted during office, and (with exceedingly rare exception) trusted after office, ex-arbs are well suited to take the brunt of those jobs off the active arbitrators.

AUSC members take on a short term role to (in Thatcher's words) "watch the watchers". That's very different. There should be a clear line drawn, so that a "double check" role doesn't become a "feeder" role. We want AUSC members to have no vested interest in the tools, other than to watch others' usage and be willing to speak (collectively) if it's at all questionable. There should be a kind of tension between the AUSC and CU/OS teams, and they should not be "too close", precisely because AUSC is not doing Arbcom or ex-arb's work. It's reviewing their work for impropriety. It's a regulation role. For that reason AUSC should not merge into a "feeder" role for CU/OS applicants. FT2 ^{( Talk |  email)} 14:02, 20 October 2009 (UTC) reply

I agree AUSC shouldn't be used to 'feed' CU/OS, but for persons in those positions, you need an experience on the matter, and it's a unique experience you can get only on enwp as CU or OS. Otherwise we have now way to know if the user would be fine in this role, contrary to RL organizations who can base their choices on past experience, CV, etc, we have only the experience on Wikipedia at our disposal. That's why I'd require at least one of the CU or OS permission, so obtained through the CU/OS selection process, thereby removing the concern of feeding the CU/OS pool. However, I agree that it would be good if AUSC members didn't use their CU/OS tools during office, for the reasons you mention, we could do this if we have enough CU/OS. Though to fully realize what you suggest, you'd need more independence from ArbCom - a committee on its own. Cenarium ( talk) 00:54, 24 October 2009 (UTC) reply

• "CheckUser is used primarily to fight vandalism... " Primarily strongly implies that there are other, unmentioned legitimate uses. All legitimate uses need to be mentioned, and "primarily" should be deleted.
• "Unlike deletion, which can be performed and reversed by any administrator but leaves an accessible version in the transparent page history, oversight removes expunges page revisions, leaving no history" The words in red need to be inserted and the substitution made for clarity.-- Wetman ( talk) 06:58, 16 October 2009 (UTC) reply

Be aware that this is only intended as a brief summary to put the subcommittee in context. The full descriptions of oversight and checkuser are at the linked pages. Thatcher 12:27, 16 October 2009 (UTC) reply

Indeed. Clarity at every stage, since there's nothing to hide. Any objections to these edits to the text?-- Wetman ( talk) 08:19, 17 October 2009 (UTC) reply

The process to appoint the three non-arbitrator members of the Audit Subcommittee is underway. If you are suitably qualified, please see the election pages for the job specification and application arrangements. Applications close 22 October 2009.

For the Arbitration Committee,   Roger Davies ^talk 19:15, 17 October 2009 (UTC) reply

Time is rapidly running out. The closing date for completed applications is 23:59 (UTC) 22 October 2009. If you are interested in becoming one of the three non-arbitrator members of the Audit Subcommittee, see the election pages now for the job specification and application details.

For the Arbitration Committee,   Roger Davies ^talk 17:35, 20 October 2009 (UTC) reply

The site notice shown to all users at the moment says "The process to appoint three non-arbitrator members of the Audit Subcommittee is underway". Could there possibly be a more turgidly offputting and bureaucratic way of phrasing that? Are you people seriously using the word "subcommittee" in a non-ironic sense, and just what the fuck is a "non-arbitrator member"? I've been using Wikipedia for over six years very, very actively, regularly participating in acronym-laden policy discussions, and that message made no sense at all to me, instead just provoking a visceral gag reaction (and this rant). Do you think the average user has a hope of understanding it?

And oh, before some nitwit attempts to explain in tedious detail, my questions above are rhetorical and I do perfectly understand what it's attempting to say.

Here's some plain English: "Wikipedia is electing users to review the use of 'CheckUser' and 'Oversight' administrative powers", linking to a page that explains what Wikipedia:CheckUser and Wikipedia:Oversight are and why anybody should care. And please, for the love of Jimbo, rename the "Audit Subcommittee" to something that doesn't sound like a Herbert Kornfeld editorial. This is the kind of shit that is destroying Wikipedia. Jpatokal ( talk) 13:01, 22 October 2009 (UTC) reply

• The term "audit subcommittee" and "election" are both linked, and both the election and audit subcommittee pages have a description of what the committee does with links to other important pages like checkuser and oversight, among others. The election announcement was also made at the admin noticeboards, arbcom noticeboard, and village pump, and the main enwiki mailing list. It is somewhat difficult to encompass a number of possibly unfamiliar concepts in a message that won't piss people off for being too long, and it is rather unfortunate that you chose 12 hours before the close of nominations to complain. (On the other hand, someone who has no idea what checkuser and oversight are will probably not make a good candidate for the position.) If you are concerned, you should probably watch MediaWiki talk:Watchlist-details, as this announcement will be removed shortly but an announcement about voting in the election will probably be made once voting opens. Thatcher 13:51, 22 October 2009 (UTC) reply

Since this apparently wasn't sufficiently blindingly obvious: my concern is not the minutiae of the process of appointment of non-arbitrator members to the subcommittee. My concern is that Wikipedia is getting strangled by vast tangles of bureaucracy that are completely and totally impenetrable to the community you are supposed to be serving. Jpatokal ( talk) 15:23, 22 October 2009 (UTC) reply

The tone isn't particularly helpful, neither is the vague insinuation about "some nitwit" misunderstanding you but your criticism is duly noted. If you have a better wording, a combination of {{sofixit}} or a specific section are the go.-- Tznkai ( talk) 23:23, 23 October 2009 (UTC) reply

I'm not clear on whether you object to the existence and mandate of the subcommittee, the election process, or the wording of the notice. The first two issues should be discussed with Arbcom, although there may be discussion in the archives of this page that might prove informative. The wording problem would be handled at the mediawiki talk page for the edit notice. Thatcher 17:09, 24 October 2009 (UTC) reply

My reading of his complaint was that it is both one that concerns the third issue and also a more general one about the unnecessarily waffling language that is often observable in Committee texts. AGK 18:08, 24 October 2009 (UTC) reply

• Jpatokal: You're a little late to the party. The Committee has been using "turgidly offputting and bureaucratic" wording for many a year now. Going into great detail when writing case decisions reduces the scope for parties to wikilawyer, and I suppose the style has carried over. If you don't like it, you're free to use the handy "edit" button and simplify the wording yourself. AGK 16:22, 24 October 2009 (UTC) reply
• As AGK points out, this is general and persistent; I had complained on this at MediaWiki talk:Watchlist-details#August checkuser and oversight election, and similar complaints have been made at WT:AC/N. I created an editnotice for MediaWiki:Watchlist-details (placed at Template:Editnotices/Page/MediaWiki:Watchlist-details, feel free to edit - needs improvement). For the general matter of wordings for arbitration-related business, I was planning on making a RFC on policy enforcement and dispute resolution, this could be discussed there - I remember the controversy around the use of Latin words, after being discussed in a RFC, ArbCom stopped using them. Cenarium ( talk) 19:37, 27 October 2009 (UTC) reply

(od) Coming late to the party, the blame for the turgid text lies squarely with me. It's mostly cobbled together from legacy stuff and I didn't have enough time to re-write it more accessibly. The other election stuff though is probably better.   Roger Davies ^talk 05:09, 30 October 2009 (UTC) reply

Don't worry about it, Roger; and don't worry about the barnstar below, either. AGK 01:11, 1 November 2009 (UTC) reply

Roger got his own barnstar the other day; I'm evenhanded. ;) Durova ³⁵¹ 01:17, 1 November 2009 (UTC) reply

Barnstar

		The Rosetta Barnstar
		Awarding the Rosetta Barnstar to Jpatokal for translating bureaucratic jargon into plain English and for a much-overdue injection of common sense. Durova 351 01:08, 1 November 2009 (UTC) reply

Shouldn't the report subpage and maybe Wikipedia:AUSC#Statistics and reports be updated in light of this ? Also, I have asked a question here about the use of revision deletion and a review possibly needed, considering the upcoming roll out of revision deletion to admins. Cenarium ( talk) 15:01, 24 October 2009 (UTC) reply

• Everything is up to date on the statistics and reports pages. When I post the october numbers I'll add a footnote about Raul's resignation (and Alison's resumption). I'll leave the bugs and such to others. Thatcher 17:07, 24 October 2009 (UTC) reply
  • Apologize I had misread the report page; I see the revision deletion vs suppression issues are handled. Thanks for your dedication Thatcher, Cenarium ( talk) 19:39, 27 October 2009 (UTC) reply

This is a follow up to Wikipedia talk:Arbitration Committee/Audit Subcommittee/Archive 1#Use of the "tools". Tiptoety seems to be inactive otherwise I would ask him directly. Has anything official been set down about AUSC members and whether they should or should not use the tools as would someone who was elected specifically as a CU/OS? – xeno ^talk 22:37, 27 October 2009 (UTC) reply

There is no unified standard at this time. I have elected to abstain almost entirely, but others have not.-- Tznkai ( talk) 23:29, 27 October 2009 (UTC) reply

I don't think it's appropriate for the tools to be used by the auditors. Since there is no standards, I will be voting much the same as I would vote for someone standing for the CU/OS position proper. I think this dearth of policy should also be made clear in the election pages. – xeno ^talk 23:30, 27 October 2009 (UTC) reply

To elaborate, but only slightly, the basic argument for restricting tool use to AUSC members is the increase in objective distance, the argument against is the inefficient use of resources. As it stands, we are relying on the individual judgment and discretion of AUSC members, as we do in all other AUSC business. Speaking only for myself, I prefer the distance to avoid even the whiff of possible groupthink. There is plenty of legitimate disagreement to go around. -- Tznkai ( talk) 23:33, 27 October 2009 (UTC) reply

I will be voting much the same as I would vote for someone standing for the CU/OS position proper. That is the only sensible position, in my opinion. Even without ever using checkuser or oversight, an auditor has full access to the checkuser log, from which much can be deduced, and full access to view suppressed revisions, including the alleged real names, addresses and phone numbers of editors and admins. Checking the logs by itself does not create a log entry, and much mischief could be done by someone without ever leaving a trail. Thatcher 00:01, 28 October 2009 (UTC) reply

My own opinion is that it is very difficult to make an honest assessment of tool usage if one does not have sufficient experience to understand how the tool works. I would not expect an AUSC member to carry out SPI investigations, but I do expect them to understand why there might be multiple checks to ensure that the narrowest range possible is being blocked, for example, and that may not be obvious to someone who has not carried out checkusers themselves. I would hope that if an AUSC member is aware of an edit that needs to be suppressed, and none of the "regular" oversighters responds in a timely way, that the AUSC member will have the judgment to assess whether this is a time-sensitive request and, if so, to ensure it is done. (Not all suppression requests are time-sensitive.) Depending on the thoughts of other members of the subcommittee, I would like to see the AUSC take on more of a guidance/education role as well, to become an effective resource for checkusers and oversighters, and probably to be actively involved in developing best practices for the use of the tools. My own use of checkuser since joining the subcommittee has mainly focused on (1) rechecking results at the request of a newer checkuser, (2) being the only CU available during a significant cross-wiki vandalism outbreak, usually at the request of a steward or (3) Arbcom-related checkusers. Most of my suppressions have been related to the same points. Risker ( talk) 00:01, 30 October 2009 (UTC) reply

The current iteration of the AUSC description page refers to the three community seats as being filled by "at-large members." The meaning of at-large is unclear. Might be retitle the section to "community members", to eliminate any scope for confusion? (This probably has been discussed in the past, so I apologise if I cause previous debate to be duplicated.) AGK 01:13, 28 October 2009 (UTC) reply

Wouldn't this denomination lead to believe that arbitrators are not members of the community ? I don't like this opposition ArbCom vs the Community, that I have found at several places and alluded in various comments. Arbitrators are members of the community and ArbCom is part of the community (specifically not beyond or above it). In this case, I'd prefer arbitrator member vs non-arbitrator member. Cenarium ( talk) 02:37, 28 October 2009 (UTC) reply

non-admin candidates

I think it may also need to be clarified that this position is not open to non-admin editors. I make mention of this as my application was rejected due to not being an admin. So it's a lot more limited that appears. I had thought it was open to any community member and was a bit confused. Basket of Puppies 03:50, 28 October 2009 (UTC) reply

That's something that has never been raised before. It should go on the elections page for qualifications for the next election. I'd rather that Roger or another arbitrator added it, though. They could change their minds whenever they wanted to. Thatcher 04:10, 28 October 2009 (UTC) reply

The election blurb described the "ideal candidate" as an admin, but it didn't specifically exclude non-admins. We've started looking into how permissions might nead to be split up to allow for non-admins to fully participate, in part because of non-admin candidates for this role, and the possibility that there could be a non-admin appointed to Arbcom at some point. I don't know how the community as a whole would vote if presented with a non-admin candidate, but it would be good to be in a position where such a selection wasn't limited by technology. Risker ( talk) 04:33, 28 October 2009 (UTC) reply

I noticed the "ideal candidate" aspect, but felt that having a non-admin member would actually help in creating broad representation on the Subcommittee. I'm not necessarily upset as I understand the technical requirements, but I was told in private email that I could not be considered for the position if I was not an admin already and informed I needed to withdraw my candidacy. If the full community decides that they want only admins as members then they achieve that during the open voting process, but I am just confused by the seemingly unpublished prerequisite. Thoughts? Basket of Puppies 04:52, 28 October 2009 (UTC) reply

It is a timing issue. The full Arbitration Committee voted to explore making the changes to permissions that would allow for non-admins to be placed in an Audit position. This needs to be managed with the developers and with the Community. We voted to move on this after the election because there was not time to do so before the start of the election. So, we did not vet you for this election but that does not close the door on the possibility that non-admins would be vetted for these seats in the future. Does that make sense and answer your concerns? FloNight ♥♥♥♥ 13:10, 28 October 2009 (UTC) reply

I certainly don't take it personally as it's a restriction that applies to all non-administrators. I would just like to see more broad representation on the Subcommittee, which would include non-admins. I understand the timing issue and hope that the next round of elections will be open to non-administrator applicants. Basket of Puppies 17:40, 28 October 2009 (UTC) reply

A non-admin being elected to this position would certainly be a feat. Kind of like a non-admin bureaucrat. There's a natural progression. Would a non-admin AUSC member be able to view oversighted edits? But not deleted edits? Peculiar. – xeno ^talk 17:44, 28 October 2009 (UTC) reply

There are a couple of ways to think about this. a) if you have not proven yourself trustworthy enough be elected an admin (either because you failed or because you never tried), you are not qualified to apply for advanced permissions like CU and OS. Or b) if you win a competitive election for advanced privileges such as CU and OS, you are obviously trusted and respected and should automatically be given admin status as well. Thatcher 18:06, 28 October 2009 (UTC) reply

I'll admit that I view serving on this type of subcommittee as being separate and distinct from an administrative position in the traditional sense. Committees of this sort are meant for investigative and consultative purposes. Thus, any member of the community in good standing ought to be able to run for the position. If this necessitates a temporary addition of permissions then I don't see the issue. Being able to view deleted and oversighted revisions is not necessarily or exclusively an admin permission, but can be applied to others who the community approves of. Basket of Puppies 18:53, 28 October 2009 (UTC) reply

Originally, many people thought that any user in good standing should be an admin, too! But somewhere along the way additional criteria was added. Loads of it. :-) Personally, I'm of a mixed mind about whether someone should be on AUSC without being an admin. The skill set needed to be on the AUSC are not identical to being an admin but their is certainly some overlap. And having admin-ship, CU-ship, OS-ship does give evidence of on wiki commitment and supplies knowledge and experience for for the work. There will be a large learning curve for someone that has never used any of the tools. But, having an non-admin, non-functionary on the subcommittee gives a fresh perspective. That part would be good. FloNight ♥♥♥♥ 19:08, 28 October 2009 (UTC) reply

I wonder if anyone has even tested the behavior of the oversight permissions when used by an account that does not have admin permissions. Certainly being able to access deleted revisions is an important part of performing the oversight function, I wonder if oversight even works without the admin buttons. I could support the idea of granting a non-admin temporary sysop status to go along with their temporary CU and OS status. but it's a pretty radical idea and I'm sure there would be concerns raised by others. Something to think about between now and the next audit committee election in 12 months, maybe. Thatcher 19:19, 28 October 2009 (UTC) reply

• The reality is that a candidate who is not an administrator is not likely to be elected by the community to a seat on the subcommittee, and any editor who is thinking of running whilst not a sysop ought to be in touch with our norms enough to know that such is the case. (That could, however, change, given a suitably trustworthy and competent candidate.) Explicitly prohibiting non-administrators from running would debar a type of editor whose inclusion on the subcommittee could be very advantageous. Policy, ideally, should reflect (and in theory is just a codified version of) the practical reality of operations on the project, so there would also be advantages to changing the AUSC policy/description page to state that only admins are eligible for election to a seat. I suppose it really does depend on whether the community would ever want a non-admin to be elected to a seat. It seems to me that a discussion to evaluate what the consensus on the issue of non-sysop AUSC members is would be a sensible next step. AGK 20:49, 28 October 2009 (UTC) reply

• • I tend to agree that a discussion should take place in order to gauge the community's consensus on the idea of a non-admin member of the AUSC. I would be happy to be involved in this. Would a central discussion be the best starting place or an RFC? Or a different venue? Basket of Puppies 20:59, 28 October 2009 (UTC) reply

• • • RFC is the standard choice of method for gauging community consensus on a topic, but really, the discussion could be held anywhere (so long as it was on-wiki and well-advertised). AGK 21:02, 28 October 2009 (UTC) reply

• • • • AGK, I think I may begin a draft RFC in my userspace just for this purpose. Do you think the RFC should wait until after the elections, just in order not to stir the pot so-to-speak? Basket of Puppies 21:03, 28 October 2009 (UTC) reply

• • • • • Kudos for taking the initiative to start the discussion. I think it would be sensible to wait until the elections have been completed, because (as you say) doing otherwise might adversely affect the election process. AGK 22:00, 28 October 2009 (UTC) reply

• • • • • • The very rough draft of the RFC is located on my user page. Please feel free to help out. :) Basket of Puppies 22:44, 28 October 2009 (UTC) reply

Speaking as a non-administrator, I'm not sure I see too much value in the presence of non-admins on either the arbitration committee or the audit subcommittee (both are currently under discussion). I think as much as possible of the work of both groups should be publicly available and reviewable by all, but it doesn't necessarily make sense to have people with no experience using a set of tools (or even any tools) be responsible for reviewing their appropriate use by others. Our hierarchy isn't so exclusive that we need a "civilian" presence on any group that has a review role, there is generally enough natural diversity. Of course, we could consider election to the AUSC and ArbCom the practical equivalent of an RfA... But we'd need quite a bit of community discussion before that would step could be taken. Nathan ^T 21:05, 28 October 2009 (UTC) reply

Nathan, that's a very, very interesting viewpoint. I can and do appreciate it. I guess all the more reason to have an RFC to gauge if the community would be willing to trust a non-admin AUSC member. As well, would they trust someone who is a former admin and turned in their mop under non-controversial circumstances? Additionally, the issue gets a bit thicker. What about admins who have been blocked for various rules violations or formally admonished (but not desysopped) by the ArbCom? I don't know if it's necessarily so cut and dry. Basket of Puppies 21:50, 28 October 2009 (UTC) reply

My own opinion is that this is a decision (select a non-admin to sit on the subcommittee) is one best left to the community. The reason why we disqualified the non-admin candidates this time around is that they simply would have been unable to do the job if elected given that deleted revisions would have been beyond their reach (and are much more numerous than suppressed revisions) and often necessary to understand the context. Adding the right to see admin-deleted pages and revisions to the oversight bit could not possibly have been done in time for this election, but remain on our to-do list. —  Coren  ^(talk) 23:08, 28 October 2009 (UTC) reply

Coren, I agree with you. After the election I will introduce an RFC to try to gauge the community's consensus on 1) having non-admin editors at all and 2) possibly requiring a non-admin editor. I hope the results of this RFC will then be used to make (if necessary) technical changes. Basket of Puppies 23:21, 28 October 2009 (UTC) reply

I saw the page header inviting votes for candidate "users"; a user is not necessarily an editor and an editor is not necessarily an admin. I am disappointed to find that in fact the candidate pool at this time is limited to admins. -- Una Smith ( talk) 03:59, 1 November 2009 (UTC) reply

I'm not going to vote in an election in which all of the candidates have been preapproved by the ArbCom. Furthermore, none of the candidates appear to represent my viewpoints, which I suppose is no accident—if they did, they wouldn't be approved to stand as candidates. I would, however, be interested in hearing someone attempt to justify holding an election under such restricted conditions. Everyking ( talk) 02:51, 30 October 2009 (UTC) reply

IMHO I have no issue with this arrangement because:

• (a) this committee is a subcommittee of ArbCom, was created by ArbCom, and has not ever been a free-standing Wikipedia body
• (b) the model is moving from a "direct appointment by ArbCom" to "some degree of community involvement"

If the community wants this body to be freely elected, divorced from ArbCom or whatever, then that can be resolved by RFC in plenty of time for the next election. Manning ( talk) 02:57, 30 October 2009 (UTC) reply

Wouldn't it be nice if the community had the option of voting for people who wanted to divorce this thing from the ArbCom? That would be much better expression of community feeling than an RFC, which as we all know usually conclude ambiguously and resolve nothing. As it stands now, it looks like we have a list of identical candidates. If there are differences between them, I wish someone could point them out to me. Everyking ( talk) 03:07, 30 October 2009 (UTC) reply

Well I haven't heard *anyone* making a case for divorcing it from ArbCom, and it's been over six months since the October election was announced. This is a SUBcommittee which has to work very closely with ArbCom and was only created to exercise powers previously exclusive to ArbCom. Also because ArbCom still has ultimate authority over CU/OS auditing, in principle these powers can be reabsorbed into ArbCom and the subcommmittee dissolved at any time. So there is no obvious reason to make provisions for the subcommittee's divorce.

If however your real issue is that you feel ArbCom should simply NOT possess ultimate authority over reviewing CU/OS actions, that is a completely different thing. That WOULD need to go to the community, as not even ArbCom has the authority to overturn years of established practice without community consensus. Struck in light of Thatcher's comment below. Manning ( talk) 03:22, 30 October 2009 (UTC) reply

Cenarium has been a proponent of divorcing it from ArbCom. See his many posts on this page. ;-)

As you point out above in (b), Arbcom has been devolving and delegating its role of appointing and monitoring OS&CU, while at the same time ensuring that these tasks are done satisfactorily. The OS&CU elections have been part of this process, and this audit subcom election was planned from the moment when the audit subcom was first established.

In time this subcommittee could become detached from Arbcom, but I think a restructure should wait until it has a few runs on the board, or until there is some catastrophic failure that acts as a catalyst for change. John Vandenberg ^{( chat)} 05:34, 30 October 2009 (UTC) reply

Maybe, through our collective experience with the ArbCom, we've become so familiar with catastrophic failure that we have trouble recognizing it for what it for what it is. Yes, the tornado destroyed the house—but the old shed out back is still standing, you see. One out of 100 ain't bad! Everyking ( talk) 05:44, 30 October 2009 (UTC) reply

I look forward to your well considered opinion here. :P John Vandenberg ^{( chat)} 06:31, 30 October 2009 (UTC) reply

Everyking, Arbcom has a responsibility to the Foundation to only grant oversight and checkuser permissions to people whom it feels are trustworthy. Until someone persuades the Foundation to change the rules that it (the Foundation) follows, then Arbcom will always have the last word on checkuser and oversight appointments. If they don't pre-approve candidates, they would have to review them after the election, and possibly deny someone who otherwise received enough votes, not to mention wasting the time of the candidate and the community. Now, there may be good reasons to grant checkuser and oversight to people who are not considered "trustworthy" by Arbcom, but you'll need to work that out with the Foundation, probably meaning Cary and the Stewards. Thatcher 03:43, 30 October 2009 (UTC) reply

Isn't the community better placed than the ArbCom to determine who is trustworthy? After all, it's the community that elects the ArbCom. I think it would be much better to have this "approval" business done after the election: then people might have a genuine choice, and if the ArbCom chose to act against the community's wishes by vetoing a successful candidate, then it would have to face community disapproval which in turn could facilitate some changes in the arrangement. Wouldn't that be better than having this sham rubberstamp process? Everyking ( talk) 03:55, 30 October 2009 (UTC) reply

I don't think there's much difference, just whether the accusations of collusion or cabalism or whatever are front-loaded or end-loaded. Thatcher 04:06, 30 October 2009 (UTC) reply

(od) Everyking: Every admin candidate who submitted a completed application went forward to the short list, i.e. none were "vetoed" by ArbCom. That none of the candidates is advocating your particular viewpoint (whatever that view might be) may simply be because it is not widely held.   Roger Davies ^talk 05:00, 30 October 2009 (UTC) reply

Do we have a place where people can raise bugs and usability problems for discussion? Onwiki discussion would be better than flooding bugzilla with duplicates. We need to discuss which issues should be fixed before the software is used for a bigger election. John Vandenberg ^{( chat)} 22:38, 30 October 2009 (UTC) reply

I've just created Bugzilla for this purpose.   Roger Davies ^talk 00:40, 31 October 2009 (UTC) reply

An alphabetically-sorted list of editors who had voted by 23:52 (UTC) 30 Oct 2009 has now been posted. This will be replaced by updated completed lists every day or so during the currency of the election. Community scrutiny is invited and concerns about double-voting should be raised on the Voting log talk page. No editors have thus far been disenfranchised.   Roger Davies ^talk 00:35, 31 October 2009 (UTC) reply

The election, using SecurePoll, has now started. You may:

• Visit the election "home page" for an overview;
• Review the candidate's statements: Dominic • Frank • Jredmond • KillerChihuahua • MBisanz • Tznkai;
• Or go straight to your personal voting page: here to cast your votes.

The election closes at 23:59 (UTC) on 8 November 2009.

For the Arbitration Committee,   Roger Davies ^talk 07:26, 1 November 2009 (UTC) reply

There's only one day to go! The Audit Subcommittee election, using SecurePoll, closes at 23:59 (UTC) 8 November. Three community members will be appointed to supervise use of the CheckUser and OverSight tools. If you wish to vote you must do so urgently. Here's how:

• Visit the election "home page" for an overview;
• Review the candidate's statements: Dominic • Frank • Jredmond • KillerChihuahua •

MBisanz • Tznkai;

• Or go straight to your personal voting page:

here.

For the Arbitration Committee, — Rlevse • Talk • 17:07, 7 November 2009 (UTC) reply

Now that the elections have concluded and the new AUSC members have been installed, I wonder if you all think this is the appropriate time to initiate an RfC regarding non-admin editors being part of the AUSC. I have a draft located here. Please let me know if you think it needs change or expansion. Once it's tidied up a bit I'll go ahead and file it as a formal RfC. Thanks! Basket of Puppies 03:55, 11 November 2009 (UTC) reply

Go for it, although you might get drowned out by that other election-- Tznkai ( talk) 04:05, 11 November 2009 (UTC) reply

Oh my, you're totally right! I'll likely wait until the entire election season is over, then. Thanks for the headsup! Basket of Puppies 04:13, 11 November 2009 (UTC) reply

I was going to post the following question on another page, but I figured since the voting is over, this might be a better place. Why was this page created?-- Rockfang ( talk) 09:42, 12 November 2009 (UTC) reply

If you mean the voting log discussion page, then that is made clear at the top: "For discussion of the voting logs, and possible instances of double-voting." If you mean the voting log, then I would presume because an on-wiki log is preferable to relying on an external extension to log important details of the election. We might have to re-examine the logs in six months' time (if, for example, evidence of sock puppetry came to light). An on-wiki page is less likely to be corrupted or removed than an extension file. AGK 14:23, 12 November 2009 (UTC) reply

I indeed meant the actual log page. Thanks for the reply.-- Rockfang ( talk) 18:40, 12 November 2009 (UTC) reply

The Audit Subcommittee has decided that the alternate member (currently user:MBisanz) of the Audit Subcommittee will not receive list membership and should not be given the Oversight or Checkuser permissions as an alternate member. Barring any unforeseen events, the Arbitration Committee will appoint the alternate member in the event of a vacancy.

For the Audit Subcommittee, -- Tznkai ( talk) 06:03, 19 November 2009 (UTC) reply

Fantastic news. :-) Thank you for the heads-up. -- MZMcBride ( talk) 09:59, 19 November 2009 (UTC) reply

Thank you. I understand that having the responsibility of making that decision was a burden, but we felt that the members of the subcommittee itself were the ones in the best position to evaluate what method of inclusion would best help. —  Coren  ^(talk) 11:53, 19 November 2009 (UTC) reply

I noticed under Special:ListGroupRights that admins have the new power of click tracking. I don't know how to use this power, but has AUSC ascertained what user information is being disclosed as a result of this module? I could imagine some people being dismayed if it turned out that any admin could figure out what pages they are viewing through a simple API query. MBisanz ^talk 01:19, 27 November 2009 (UTC) reply

o_O That does, err, worry me a little bit. I wouldn't panic if it had been assigned to checkuser, but absent details on what exactly this right entails I think having it on to admins is... well, let's just say I, for one, will be looking into it as of this minute. —  Coren  ^(talk) 01:29, 27 November 2009 (UTC) reply

It would appear it has now disappeared from ListGroupRights. It was between

 # View detailed abuse log entries (abusefilter-log-detail)
 # Add groups: Rollbackers, Account creators, Edit filter managers, Autoreviewers, Confirmed users and IP block exemptions

MBisanz ^talk 02:15, 27 November 2009 (UTC) reply

Yes, the extension was disabled (even if the right existed) but I've asked Tim to remove the right from the sysop groups anyways out of justifiable paranoia.  :-) —  Coren  ^(talk) 02:23, 27 November 2009 (UTC) reply

Interested editors are invited to participate in the SecurePoll feedback and workshop. SecurePoll was recently used in the Audit Subcommittee election, and has been proposed for use for the upcoming Arbitration Committee election at this current request for comment (RFC). Your comments, suggestions and observations are welcome.

For the Arbitration Committee,
Dougweller ( talk) 09:08, 12 November 2009 (UTC) reply

As has already been noted on this page, someone has deleted revisions on Wikipedia talk:Arbitration Committee/Noticeboard, Wikipedia:Arbitration Committee/Noticeboard, and User talk:David Gerard. I cannot possibly imagine the levels of stupidity required to think this was a good idea to try to un-ring such a loud bell. If mistakes were made, trying to delete them from existence is the worst possible solution. -- MZMcBride ( talk) 00:22, 2 December 2009 (UTC) reply

I said it on functionaries-en, and I'll say it again in public. To refuse to retract the very serious accusations made against David, instead opting to use brute force to pretend that they were never made in the first place, shows an incredible lack of intellectual honesty, integrity, and accountability. It can only diminish the community's trust in the arbitration committee. — Werdna •  talk 00:25, 2 December 2009 (UTC) reply

Can you expand on what has been oversighted? Viridae ^Talk 00:30, 2 December 2009 (UTC) reply

How is this justifiable within policy? Basically, only Oversighters can see it. Wouldn't it be more appropriate to retract or amend? Durova ³⁷¹ 00:31, 2 December 2009 (UTC) reply

This is akin to a newspaper trying to go around and collect copies of a bad story from every household. Print a damn retraction, if necessary. But this revision suppression (from admins, even!) needs to be immediately reversed. Unfathomably stupid, this whole affair. -- MZMcBride ( talk) 00:37, 2 December 2009 (UTC) reply

Totally agree, it's like instead of referring to the moon as the moon, you just call it a big round rock in the middle of space and orbits the earth. Just changing the name of what has happened, doesn't change what happened. Someone needs to lose the bit on this one. -- Coffee // have a cup // ark // 00:39, 2 December 2009 (UTC) reply

Lets not go buy our pitchforks and torches quite yet and give ArbCom (or whoever made the suppression) a chance to comment / explain. That said, this does appear troubling. Tiptoety ^talk 00:40, 2 December 2009 (UTC) reply

Can anyone offer a general description of what led to this, for those of us who aren't aware? Reso lute 00:42, 2 December 2009 (UTC) reply

If somebody explains, their edits might be suppressed! :-| Killiondude ( talk) 00:46, 2 December 2009 (UTC) reply

• No, I think we are supposed to speculate. Surely, the clue to all of this - is has he been regranted his checkuser and oversight rights, well rights is probably a misnomer, but you know what I mean.  Giano  00:42, 2 December 2009 (UTC) reply

• He has not been regranted his checkuser and oversight permissions (at least, there is nothing in the meta logs to indicate as such, and no formal request to the stewards has been made). However, a public explanation is definitely deserved, and should have been given as soon as anything was suppressed. NW ( Talk) 00:48, 2 December 2009 (UTC) reply
• This is where to check. Not currently. Dragons flight ( talk) 00:51, 2 December 2009 (UTC) reply

I'm just going to remind everyone that suppression is reversible, and ask everyone keep as cool a head as possible.-- Tznkai ( talk) 00:44, 2 December 2009 (UTC) reply

My head is perfectly cool, and on top of my shoulders, which I suspect will not be the case for someone close by very soon.  Giano  00:46, 2 December 2009 (UTC) reply

Well, please supply a correction if this is mistaken: doesn't the Foundation's Oversight policy apply? And if so, what clause was this done under? It seems obvious that an attempt to erase the problem draws far more attention than a simple retraction would. Clarify, please? Durova ³⁷¹ 00:49, 2 December 2009 (UTC) reply

Courtesy link: Foundation's Oversight policy. Durova ³⁷¹ 00:54, 2 December 2009 (UTC) reply

Can someone summarize the general nature of what was oversighted? It would help to know why this is such an issue. People who know what is going on seem really worked up, but those of us that are unfamiliar with the details could use a primer to decide if we should get worked up too! -- Jayron 32 00:52, 2 December 2009 (UTC) reply

David Gerard had his Checkuser and Oversight flags removed involuntarily. The announcement and discussion by the community has been suppressed. ⇌ Jake Wartenberg 00:57, 2 December 2009 (UTC) reply

If my advice and that of some others is acted upon, a revised announcement will be posted that will also permit the suppressed content to be restored. If not, there's always Citizendium.... Thatcher 01:01, 2 December 2009 (UTC) reply

Now you're just being cruel... :P MastCell  ^Talk 01:15, 2 December 2009 (UTC) reply

Citizendium tends to deal with criticism somewhat suppressively too, I believe. *Dan T.* ( talk) 01:20, 2 December 2009 (UTC) reply

(rhetorical tone) You're kidding! (normal tone) It's events like this that prove the value of certain Sites Which Shall Not Be Named. -- Seth Finkelstein ( talk) 01:16, 2 December 2009 (UTC) reply

How long does it take to post a message explaining this?

You don't accidentally suppress this many revisions. Try as you might, you can't just trip on something and censor a few dozen revisions. This was clearly planned (though clearly not thought-out). Why is it taking so long to get a statement from the person who did this? Surely they could've written one up before embarking on this stupidity. -- MZMcBride ( talk) 01:18, 2 December 2009 (UTC) reply

As long as it takes to get it right, I would imagine.-- Tznkai ( talk) 01:20, 2 December 2009 (UTC) reply

I'm afraid that we don't have that kind of time. - Rjd0060 ( talk) 01:22, 2 December 2009 (UTC) reply

Only on Wikipedia could one be cross-referenced from a discussion about possibly inappropriate censorship to an article about apocalyptic predictions. — Pathoschild 05:27:29, 02 December 2009 (UTC)

It's only a rush for the Arbitration Committee when they're defaming people? Hah. -- MZMcBride ( talk) 01:25, 2 December 2009 (UTC) reply

You know when someone's waiting for the elevator, and they keep... hitting... the button over and and over? It's annoying, and it doesn't make the elevator doors open any more quickly. MastCell  ^Talk 01:32, 2 December 2009 (UTC) reply

The elevator analogy is cute, but not really fitting. This is more like someone making a giant mess and waiting for a shitstorm to brew as they slowly write a justification for doing so. Do you have a good analogy for that? I imagine there might be some good examples from the George W. Bush presidency if you get stuck. -- MZMcBride ( talk) 01:38, 2 December 2009 (UTC) reply

• Put it this way, if you post something in haste that you later find necessary to revise, it will take you far longer to deal with the cleanup than if you take the time to get it right the first time. Thatcher 01:33, 2 December 2009 (UTC) reply
  • You're too smart for your own good. -- MZMcBride ( talk) 01:38, 2 December 2009 (UTC) reply
  • And, hey, if they end up putting up a statement they don't like, there's always more suppression. That's how this stuff works, right? -- MZMcBride ( talk) 01:40, 2 December 2009 (UTC) reply
    • Eh, why always with the drama? Patience, my dearies. Presumably an answer will be forthcoming in the goodness of time. I don't see why people need to set their heads on fire and run around in circles right now. I submit that this isn't THE MOST IMPORTANT THING happening here right now. I'm sure Scibaby is awakening some sleeper socks, or possibly Grawp is preparing to wield his massive cock as we speak. Crafty ( talk) 01:41, 2 December 2009 (UTC) reply

A brief note

There will be an explanation about the suppression coming shortly (hours, not days). —  Coren  ^(talk) 01:39, 2 December 2009 (UTC) reply

This takes hours? It appears you're at keyboard now. What part of Meta policy is this action compliant with? Help! Help! I'm being suppressed! Durova ³⁷¹ 01:43, 2 December 2009 (UTC) reply

I fail to understand why the ArbCom didn't make a post about this decision, before they did it. It's ridiculous to pretend that they wouldn't know that this would be brought under extreme question; and it's even more ridiculous for them to act like they couldn't have taken the extra time to write out at least some small comment on why they were going to do this, and how their reasoning applied to the meta oversight policy. -- Coffee // have a cup // ark // 01:52, 2 December 2009 (UTC) reply

If they had done so, we would have all rushed to familiarize ourselves with the information that they wanted no-one to be familiar with.  Skomorokh  01:54, 2 December 2009 (UTC) reply

In all honesty it would have been perfectly fine, (albeit not exactly how I would like to have seen it happen), if they had posted a few minutes after the oversight why they did it, but hours afterwards? That is unacceptable. -- Coffee // have a cup // ark // 03:25, 2 December 2009 (UTC) reply

Please be sure to include how the actions were justifiable under the oversight policy (assuming of course that little checkbox labeled "Suppress data from administrators as well as others." was checked). - Rjd0060 ( talk) 01:44, 2 December 2009 (UTC) reply

There had better be. What the heck is going on here? - Tbsdy lives (formerly Ta bu shi da yu) ^talk 01:55, 2 December 2009 (UTC) reply

I look to forward your statement with great interest (phrasing is deliberate) -- Seth Finkelstein ( talk) 01:58, 2 December 2009 (UTC) reply

I'd like to add my name here as one who is quite confused by this whole ordeal and is looking forward to a coherent and comprehensive explanation. Glass Cobra 03:18, 2 December 2009 (UTC) reply

The question that should be answered first was the one immediately obvious to most people who read the original announcement; whose idea was it to write something like that about someone under their real name, without any evidence of any sort, and then post it publicly under the official imprimatur of the arbitration committee? Particularly if the evidence used was the evidence later supplied by someone else, namely a single blog post that contained no private information obtained via use of either oversight or checkuser? I'd like to hear that explanation first, before the relatively minor issue of later revision suppression is discussed. Nathan ^T 02:25, 2 December 2009 (UTC) reply

A guess

Since Arbcom is being slow, I'm going to throw out a theory of their motivation. The announcement they suppressed contained an HTML link to an off-wiki page containing an email apparently published without its sender's permission and the IP details of the specific identified sender. It was the off-wiki posting of that information which motivated stripping David Gerard of his rights. Frankly, I found it weird that Arbcom called attention to that off-wiki page at all when making their original announcement. I would guess the point of the suppression was really to kill that HTML link (perhaps the individual who wrote the email even complained to Arbcom). If that's the case, then I would suggest that Arbcom really ought to repost the rest of the announcement and unrelated discussion minus the offending HTML link. Of course, that's just my guess about what's going on. Dragons flight ( talk) 02:32, 2 December 2009 (UTC) reply

Sensible analysis. If that's all that happened, why would the suppression go so far beyond what was necessary? See Streisand effect. Durova ³⁷¹ 02:45, 2 December 2009 (UTC) reply

I don't believe that the orginal anouncement contained a html link. Seddon ^talk| ^WikimediaUK 03:04, 2 December 2009 (UTC) reply

It didn't - and that refusal to be specific caused others to hunt up specifics, and put them in a mind to be outraged over some damn thing, which made the information more public rather than less. Actions have consequences, and the Streisand effect is a known part of the consequences of suppressing information - so of course they've made the same mistake again. — Gavia immer ( talk) 03:31, 2 December 2009 (UTC) reply

My recollection is that a link was included in the announcement by way of explanation. (Obviously I have no direct way of checking.) If I'm mistaken about that, then I'll gladly offer my apologies to Arbcom. Dragons flight ( talk) 03:35, 2 December 2009 (UTC) reply

What? No. Thats silly. The prior page was sloppy in its claims with respect to David's actions. (For example— publishing on your blog an email someone sent to you personally, as is mentioned above, is in absolutely no way in violation of the Wikimedia privacy policy). While I don't know the specifics of the oversighting here, in the discussion which occurred on the checkuser list I argued that the carelessly chosen words used by arbcom were themselves not only inappropriate, but libellous towards David. It would seem to me to be completely appropriate to remove them per use (2) of the oversight policy while a more carefully considered statement was drafted, so I think that is the most reasonable explanation in the absence of further information. An overly hasty explanation which itself would have to be amended or removed would not be helpful, so the loud demands for an instant response seem like a lot of unjustified drama to me, the world doesn't end if it takes a day or so. I seriously doubt anyone intended to remove any of the discussion beyond which was strictly necessary to remove the libellous claim(s). -- Gmaxwell ( talk) 03:22, 2 December 2009 (UTC) reply

If this stands, shall I start creating a list of pages where other people, such as Mr. Kohs, have been possibly libeled, and thus years of revisions will have to be deleted? -- NE2 03:27, 2 December 2009 (UTC) reply

Unfortunately, other stuff exists. Not all other things are equal though, but if there is other stuff that really should go away, then perhaps you *should* be making a list. -- Gmaxwell ( talk) 03:37, 2 December 2009 (UTC) reply

Wikipedia:Administrators' noticeboard/Mantanmoreland has some stuff about a certain Mr. Bagley that is significantly worse than what was said about Mr. Gerard. Please suppress every revision that contains these statements, as well as any revisions of Wikipedia:Administrators' noticeboard that may have contained them before the discussion was split out. Other examples are at Wikipedia:Articles for deletion/Judd Bagley and User:Anticipation of a New Lover's Arrival, The/Overstock. -- NE2 03:47, 2 December 2009 (UTC) reply

That's a very fair comment, actually. I'm going to courtesy blank those - Alison ^❤ 11:27, 2 December 2009 (UTC) reply

In a nutshell, Arbcom was largely in agreement that the prior wording should be revised or replaced, but is not in agreement about what should replace it. Thatcher 03:29, 2 December 2009 (UTC) reply

And by removing entire announcements and threads without any explanation at all, they guaranteed the action would attract more attention. Could it have been too hard to have substituted a placeholder message? Durova ³⁷¹ 03:39, 2 December 2009 (UTC) reply

..and that justifies oversighting everything to do with the announcement how, exactly? I'm thoroughly confused. -- Conti| ✉ 11:03, 2 December 2009 (UTC) reply

Looks like a childish powerplay to me, with packs of little buddies ganging up and fighting with each other, using oversight like a toy. The initial arbcom announcement was spot on, Gerard doesn't rise to the level of decorum expected of functionaries and checkusers. He used his position of checkuser to ridicule someone on his blog and call him a 'waste of skin'. I don't think it is too much to expect the people with these special rights to behave like adults. -- I am not Paranoid ( talk) 03:54, 2 December 2009 (UTC) reply

Delinking that per the contributory copyright infringement clause of the copyright policy. It's easy enough to find; let's not add to the problem. Durova ³⁷¹ 03:58, 2 December 2009 (UTC) reply

Is Gerard's blog a BADSITE now? *Dan T.* ( talk) 04:20, 2 December 2009 (UTC) reply

Set down the drum before you hurt your knuckles beating it, Dan. ;) This is exactly the same principle we use to delete some but not all YouTube links (depending on whether they're legit hostings or copyvio). Durova ³⁷¹ 04:34, 2 December 2009 (UTC) reply

hang on a minute... I think you're suggesting, Durova, that David Gerard's blog is infringing copyright law - that David Gerard is breaking the law. That assertion could be considered defamatory, and presumably this conversation could now be hidden? I dunno if that's true or not, but it's the silly season on wiki, so it might be... Privatemusings ( talk) 04:54, 2 December 2009 (UTC) reply

Of course that's not what she is suggesting. Don't be more absurd than necessary (which I admit is pretty absurd to keep up w/ this thread). Protonk ( talk) 06:02, 2 December 2009 (UTC) reply

no biggie, protonk, but doesn't asserting that a link would be 'contributory copyright infringement' mean that the post infringes ( / breaks) copyright ( / the law)?? - agree on the absurdity of this all though, so p'raps lets leave it there :-) Privatemusings ( talk) 07:18, 2 December 2009 (UTC) reply

Eh, reading up it seems that you aren't jumping too far in order to reach that conclusion. I still think this is a case of your speculation on her speculation and the compounding is causing us to get (more) out of hand. Protonk ( talk) 07:37, 2 December 2009 (UTC) reply

Did any of the folks at this thread other than myself look at the account's edit history? Gems like this? The account is indeffed now. Beat me up for defending Giano's honor at my own user talk, please. ;) Durova ³⁷¹ 17:52, 2 December 2009 (UTC) reply

I can't argue against a possible basis in decorum— but the claim of some kind of checkuser misuse is spurious. Full agreement on this being petty politics, but what can we expect from a process which has basically become political. Everyone is elected via popularity contests these days, and that breeds all kinds of stupid games. (As opposed to the different stupid games which come out of other modes of operation). In any case, it's important to not confuse the pretext for purpose. If the English Wikipedia community, or its delegated agents, feels that David hasn't acted with appropriate decorum then it should act on that.

Though I feel compelled, from years of professional respect, to give David's decorum a little defence: He has tended to be somewhat scathing and bombastic at times, but at times being bombastic, blunt, or using satire is the *most* adult response. At least if your notion of an adult interaction is a straightforward, appropriately respectful, and honest response. (I've been listening the the house of commons on CSPAN recently, and David has *nothing* on those folks!) Many Wikimedians prefer a very introverted and navel gazing notion of decorum, and I think sometimes erring too far that way. At times David has been the only person willing to stand up and tackle a public issue with forceful words when forceful words were required. By no means do I think his actions have always been perfect, but it amuses me to consider that David's interactions with the press have always seemed to produce less blow-back and loss of face than the foundation's own interactions. Perhaps David is just crazy enough that the press happily quotes him directly rather than making crap up, perhaps the press is more likely to believe someone who isn't afraid of being a little critical, I'm not sure.

In any case, some finding regarding decorum isn't what was there. I'm sure that it'll all be sorted out in the fullness of time. -- Gmaxwell ( talk) 04:13, 2 December 2009 (UTC) reply

If you're going to hold an authority position, you really need to conduct yourself in a civil and respectful manner, even in the midst of disagreements. And we're not talking about an occasional lapse; he's had a consistently combative tone for as long as I can remember. Everyking ( talk) 06:53, 2 December 2009 (UTC) reply

Is Gerard a symbol of Wikipedia's past or its future? - your choice.

Instead of demanding answers, which will probably not be forthcoming from the Arbcom - one should just analyse the facts. The statement announcing the removal of rights and condemning Gerard's behaviour was not signed by a couple of newbie Admins, but by a full flush of Arbs, including, very interestingly, Wikipedia's own famed lawyer Brad and one or two of the other more intelligent ones. So they were sure of their facts and actions. So rather than demand answers of them, demand an answer of Gerard, what has he done or said to cause the oversight. The answer to that you will find more readily deductible and less taxing to the imagination. Of, course, having gone so far, the Arbs should have stuck to their guns and lived of died together, their position is now untenable. So having rendered the Arbcom impotent, let those that have interfered (and interference has happened) now set up their own arbitration committee, and I can assure you they are keen to do so. Then ultimately lets see how some of those shouting here like that, in a few month's time. Wikipedia is regressing and its not for the better.  Giano  08:17, 2 December 2009 (UTC) reply

It's been regressing for a while now. - Tbsdy lives (formerly Ta bu shi da yu) ^talk 08:54, 2 December 2009 (UTC) reply

• If what you say is true and this was directed from outside ArbCom, a statement from those people would be nice. We're not a democracy, but we sometimes play one and pretend that ArbCom is our elected government. If ArbCom is not in charge in this situation, they need to say so, to make it clearer what authority (if any) they have, and if it is even worth electing new people for next year (depending on who is really in charge). If ArbCom is in charge, I am utterly puzzled by this attempt at self-destruction. Wouldn't just collectively resigning have been easier? — Kusma ^talk 11:48, 2 December 2009 (UTC) reply

On this thread

Content is great stuff. I've got a new camera and I've started adding video. (I haven't gotten sound to work properly in ffmpeg2theora, will need to fiddle with that.) - David Gerard ( talk) 11:54, 2 December 2009 (UTC) reply

And there you have it. David Gerard gets it. Wikipedia is about content, not this stupid meta crap. - Tbsdy lives (formerly Ta bu shi da yu) ^talk 13:14, 2 December 2009 (UTC) reply

David, if you come up with a good process for video, please post it somewhere. The video instruction I've found so far is scattered and confusing. Wknight94 ^talk 15:03, 2 December 2009 (UTC) reply

That's 'cos it basically is confusing! I might try Firefogg again - David Gerard ( talk) 18:05, 2 December 2009 (UTC) reply

David, if you've got files that do not convert correctly please let me have them. Email works. I'll fix ffmpeg2theora, or (if it's an ffmpeg bug) at least make sure someone does. It's worth testing the nightly builds, also if your audio problem is a synchronization problem try giving it the --sync option. Wknight94, seen commons:Help:Converting_video?-- Gmaxwell ( talk) 15:26, 2 December 2009 (UTC) reply

If you listen to that video, you'll hear the background noise - which should be steady - chopping in and out in a strange manner. It was done by feeding the camera MJPEG to the Ubuntu 9.10 repo ffmpeg2theora at the command line. I'll try again with Firefogg, and also try again with a compiled version of the latest ffmpeg2theora source - David Gerard ( talk) 18:05, 2 December 2009 (UTC) reply

Lets get back to the pitchfork branding please. Seddon ^talk| ^WikimediaUK 15:54, 2 December 2009 (UTC) reply

We can upload the videos! - David Gerard ( talk) 18:05, 2 December 2009 (UTC) reply

Background information

I see that many people have asked about the background story that led to this Oversight abuse incident. You will find full details about this here: [1] Supreme Dog-King of Dalmatia ( talk) 23:45, 2 December 2009 (UTC) reply

Given that it may take some further time for an official explanation to come given the constraint of availability, timezones, and the complexity of the underlying matter (I may have been overly optimistic in my initial estimation) I will make a brief statement myself in my own capacity. This is not a statement by the Committee.

A few days ago, the Committee made an announcement on its noticeboard relating to the advanced permissions of David Gerard (specifically, his access to checkuser and oversight, and his access to the functionary's mailing list). As is usual in such notices, it was a simplified summary of an otherwise complex matter.

David Gerard expressed objections to the nature of the announcement and that, as stated, it could cause significant harm to himself in the real world. While the committee did not believe the statement to be erroneous, they agreed that it was strongly worded and that it could easily be misconstrued to mean more than it did to uninformed bystanders. There was an attempt made to amend the statement to be more general, but that was insufficient to allay the concerns of David because multiple references to the previous version could still be found and linked to.

In every case the committee speaks publicly, we are torn between the necessity of transparency and the equally compelling necessity of avoiding undue harm to everyone involved. Given the possibility of real harm to David's reputation and career by having our announcement taken out of context, his objections are entirely understandable. In the day following, Mike Godwin (acting, it is important to note, in his private capacity and not in any way as the Foundation's legal counsel) brokered and agreement between David and the Committee that the statement should be removed entirely to avoid the possibility of further needless harm, as sufficient to all involved to resolve the matter.

Given that the concerns of the Committee had been already entirely addressed, and that the continued existence of the notice was neither necessary nor useful, the possibility of real harm to a longtime contributor dominated its side of the balance. I, and the other arbitrators, therefore agreed to remove the revisions of the notice and of the surrounding discussion of its contents. The reason why no statement was immediately offered is that explaining matters in a way that did not simply create the same problem all over again to the satisfaction of all involved requires significant coordination, while the continued existence of the announcement was doing demonstrable continuing harm; as is usual in the case of revision suppression, time is of the essence and the action precedes the explanation.

The reason why I am now stepping out of line and providing this statement — which I hope will help comprehension without causing further drama — is because the continuing speculation about the motives behind the suppression undermine the resolution of the incident and cause undue strife within the project. I reiterate here that I am speaking in my own personal capacity as an informed functionary, balancing privileged information with the understandable desire of the community for an explanation, but emphatically not in the name of the committee in any way. —  Coren  ^(talk) 15:57, 2 December 2009 (UTC) reply

Thank you Coren. I have yet to read the substance of what you just wrote but at least its a start. Why this couldn't have been said sooner I do not know. I will let you know what I think soon. Seddon ^talk| ^WikimediaUK 16:06, 2 December 2009 (UTC) reply

Thanks for this. Although I think that oversight was highly unwise, and unjustified. Many of us saw the initial statement and have copies of it, and indeed there's nothing at all to stop us quoting from it. The usual response to a statement with an unfortunate wording is a retraction, clarification and apology, not a suppression. Further, many users have accused arbcom statements of being unfair - are all such cases to be reviewed by Mike Godwin in future, and lead to a negotiation between the committee and the other party?-- Scott Mac (Doc) 16:08, 2 December 2009 (UTC) reply

• • I also note that David Gerard has not removed the original blog posting which resulted in the removal of his tools.-- Scott Mac (Doc) 16:11, 2 December 2009 (UTC) reply

I'm having difficulty understanding how Mike Godwin, not acting as a Foundation employee yet using his access to the private mailing list (which only comes from him being a Foundation employee/legal counsel), did not abuse his access and remit. Thank you for the statement, Coren. -- MZMcBride ( talk) 16:09, 2 December 2009 (UTC) reply

Mike's access to the list, I believe, is similar to that of former arbs, in that his institutional memory and advice are helpful when discussing these kinds of issues. This in no way means that every time he speaks, he does so as legal counsel--that should be obvious. ArbCom decides who has access to that list; once given, how can responding to e-mail to the list be viewed as abuse? I'm sorry, MZM, I'm not understanding your concern. -- Avi ( talk) 16:21, 2 December 2009 (UTC) reply

As of January 16, 2009, the former arbs had their access removed from the list as not being required for the functioning of arbcom. There is a strong argument (that you make) for Mike's inclusion on func-en, but it does not apply to his inclusion on arb-l. MBisanz ^talk 16:46, 2 December 2009 (UTC) reply

Almost all the discussion took place on functionaries-en, at least up until last night. It's quieted down considerably, suggesting the discussion has since moved to Arbcom-L. Thatcher 16:53, 2 December 2009 (UTC) reply

I have no idea if he is on arb-l; I have no access there. I was speaking solely about func-en. -- Avi ( talk) 16:54, 2 December 2009 (UTC) reply

No, Avi, there's a fundamental misunderstanding here. Mike's access to functionaries-l is not the same as most of the other people on the list. He was never an Arbitrator or given CheckUser or Oversight by the Arbitration Committee or Jimmy Wales. He is on the list because and only because he is the legal counsel for the Wikimedia Foundation. Coren's statement explicitly states that he was not acting as Wikimedia counsel here (and I imagine, though I haven't seen so publicly) that Mike would say exactly the same. So my question is, how is it appropriate for him to act in the English Wikipedia's internal matters specifically not as Foundation staff when the only way for him to do so uses the access granted to him specifically for (and only for) his job? To me, it seems like a gross misuse (and possible abuse) of access and remit. If he were acting as the Foundation's legal counsel, he would have a reason to post to the list, act as a broker, do whatever it is he would do in that role. But he specifically seems to have stepped out of the role, in which case there's a lot of confusion as to why he was participating in any of these private discussions at all. Something does not seem kosher here. -- MZMcBride ( talk) 18:41, 2 December 2009 (UTC) reply

Strongly agreed. Brilliantine ( talk) 19:26, 2 December 2009 (UTC) reply

MZM, even if that were to be true, that is irrelevant, I believe. One of the the purposes of the list is for ArbCom to be able to tap into the institutional knowledge and advice of the members of said list. So if Mike sees something that he thinks is a bad idea going on, you are telling me that you think he has to refrain from giving advice because he lacks overt CU and OS bits (although he is a member of "Staff", does that give him bits?) even when he expressly says he is speaking from experience and not as "/the lawyer/"? I think that is a bit of a stretch. Aside from being the WMF lawyer, he is also someone with a lot of personal, professional, and academic experience with these kinds of issues (having argued cases, written about cases, written articles, and just dealt with this topic for decades) and if he has some personal advice to give ArbCom on a forum to which ArbCom expressly invited him, I see absolutely nothing wrong with that. Of course he stepped out of his role as WMF lawyer; he was donning the role of "invited list member friend of ArbCom who wants to share some advice/experience." I see no misuse of any access; the access is their for each invited member to be able to help ArbCom handle the mass of issues they face. -- Avi ( talk) 20:19, 2 December 2009 (UTC) reply

I think any attempt to present Godwin's professional experience as particularly relevant to the issues at hand here is a red herring. I could understand his input if Arbcom went and specifically asked him whether he felt their actions were justified, but that doesn't appear to be the case here. From my reading of the emails, he appears to making direct demands of the committee, waving what appears to me to be some kind of vaguely threatening 'I know best' legal flag. More to the point, I haven't seen the community ever give Godwin or Arbcom the remit to enter into negotiations of this kind with regard to sanctions applied to an unrelated party. Notice Coren's use of the phrase "brokered an agreement". Brilliantine ( talk) 20:39, 2 December 2009 (UTC) reply

Avi is correct. The people on the functionaries list are there partly because of role, and partly because their input is valued. AC doesn't have to explicitly ask, each time, "Hey, anyone have anything to say?". It is expected that if people have insight, they will share it. ++ Lar: t/ c 21:06, 2 December 2009 (UTC) reply

Ok, I think that helps me understand how this works. This makes MG's actions look even more indefensible to me. While he does say in these alleged emails that he is commenting in a personal capacity, he also makes much of his professional experience. I see this as essentially waving a big stick. There are no 'legal' criteria attached to the removal of checkuser/oversight, as far as I know. Moreover, while I'm not a lawyer, I doubt you would find many who would think DG could successfully bring a libel case based on the Arbcom statement, as MG implies in a round-about manner in one of the alleged emails. I believe he is using his position to make disingenuous comments that further a specific end. Brilliantine ( talk) 21:32, 2 December 2009 (UTC) reply

I respectfully disagree, Brilliantine. When I saw him refer to his "professional experience" I understood it to mean something like: "Guys, I'm not speaking out of school here. I've been involved with this kind of stuff on a very reguklar basis for 20 years, and with that experience, I'd suggest…" -- Avi ( talk) 22:43, 2 December 2009 (UTC) reply

Indeed, we will have to disagree. I can't bring myself to see these alleged emails as anything other than intimidatory, no matter how hard I try. It's the forceful and insistent tone that does it for me. Brilliantine ( talk) 22:59, 2 December 2009 (UTC) reply

But, at the least, you should appreciate that others may not have taken it so, and that Mike may not have meant it so. And thus, even if one can think that Mike was guilty of poor phraseology (about which we will agree to disagree), I think saying that he "grossly abused his access" is a bit of an overstatement in this light, in my opinion. -- Avi ( talk) 23:04, 2 December 2009 (UTC) reply

Avi, it isn't invited-list-member-friends-of-ArbCom-who-want-to-share-some-advice-or-experience-l, it's functionaries-l. Mike is on the list because of his job. When he leaves his job, he will (presumably) be removed from the list. (Though, seeing the trend with the Arbitration Committee, they may just appoint him as a 19th Arbitrator; sense and logic left years ago.) With all due respect to Mr. Godwin, if you were trying to genuinely establish a list for people who "share some advice and institutional memory," there is a long list of people who would be higher priority to be added (Kelly Martin and Doc glasgow have more institutional memory between them than half the list's current membership combined). Mr. Godwin is on the list because of his job. Period. If he steps out of his job, it's not his place and it's not proper for him to engage in private dealings concerning the English Wikipedia, esp. for the purposes of manipulating on-wiki action (in this case, a massive use of RevisionDelete). -- MZMcBride ( talk) 21:22, 2 December 2009 (UTC) reply

I just want to say that I've been part of online communities, and have studied their norms and procedures, for more than a quarter of a century now. What's more, I've been doing privacy law, as well as other Internet-related law, for 20 years. Now, you may take the view that I should just excise that part of my brain when something happens on Functionaries that seems precipitous and at great risk of undermining Arbcom's authority as well as of doing harm to a publicly named individual, but I feel morally that it would be inexcusable not to speak out. You may have a different notion of ethics than I do, but I'm comfortable with the choices of (a) expressing my concerns when the issue popped up, and (b) acting as a mediator, taking into account all sides, to try to de-escalate disputes and return things to a more rational level. I believe Arbcom's removal of the original language was the right thing to do, not because oversighting it was going to keep it a secret (I'm not that naive), but because it would signal that Arbcom itself had second thoughts about the wording it had used, which is a good message to send and a lesson for the future for everyone. I'm perfectly aware that a number of Wikilawyers, such as MZMcBride, apparently, imagine that the world is properly organized into pigeonholes, and that good little soldiers stick to themselves in the little pigeonhole that the Wikilawyers think is appropriate for them. I'm sorry, but I strongly disagree -- I think that notion is morally vacant. And, it must be said, I'm a member of the community too, have been editing Wikipedia for years before I even bothered to register a username, and want the projects and the community to succeed with a minimum of grief. One of these days, I imagine, even MZMcBride will want to see more people willing to step up and try to fix a problem when procedures have gone off the rails. Societies work better when people are willing to do that. MikeGodwin ( talk) 05:01, 3 December 2009 (UTC) reply

Normal members of the community operate on-wiki. I don't think it's a particularly radical notion that you keep out of routine wikidrama and only involve yourself in the affairs that directly relate to the reason you're a member of the list (Foundation-related legal issues). But I say this realizing that there's little to be done about it. Oh well. -- MZMcBride ( talk) 05:18, 3 December 2009 (UTC) reply

While Mike’s post makes sense, it is a pity he couldn’t do it without labelling someone a Wikilawyer, a pigeonholer, a good little soldiers, and a holder of something morally vacant. -- SmokeyJoe ( talk) 10:55, 3 December 2009 (UTC) reply

• Did Mr. Godwin consider the possibility that it was David Gerard and his actions that brought about "significant harm to himself in the real world" ? The notion that one takes responsibility for one's own actions seems to be a foreign and frightening concept to some these days. Tarc ( talk) 16:38, 2 December 2009 (UTC) reply
• Has David Gerard made or implied legal threats? To the Arbs, Wikipedia or the Foundation. Only, the truth please.  Giano  16:57, 2 December 2009 (UTC) reply
• Thank you for shedding some light on this situation, Coren. However, I still see no satisfactory explanation why the usual thing wasn't done (retract what needs to be retracted, then courtesy blank). It is weird and worrying to see this done by oversight. Another thing: Can we have an official statement that David Gerard's checkuser privileges have been removed? The link at m:Steward requests/Permissions would like a target that explains it. — Kusma ^talk 17:02, 2 December 2009 (UTC) reply
• This doesn't appear to fall within the correct use of the oversight tool. Can I assume that the Audit Subcommittee will be reviewing this, in a manner well separated from ArbCom? Tony Fox (arf!) 17:17, 2 December 2009 (UTC) reply

<giggle/> Thatcher 17:25, 2 December 2009 (UTC) reply

Well given that one of the AUSC members has resigned and two are arbs who are possibly involved in the situation (isn't clear exactly what Arbcom has done as a body or as individuals here), I'm not sure how the other three members would handle a case. Also, it would take someone emailing AUSC to formally file a request I assume. MBisanz ^talk 17:27, 2 December 2009 (UTC) reply

Thank you Coren for providing that explanation. It provides a reasonable reason for using the oversight tool in this instance. The people clammoring for arbs heads don't seem to have any personal interest in the matter besides a desire to dig into a matter that really does not affect them. I don't really need to read everything that ever was written here, and I don't see the need to continue to punish David Gerard, in perpetuity, for his trangressions. He has committed a wrong. His wrongdoing led to an appropriate response from the ArbCom, which was to remove his access to advanced tools. To maintain that the evidence of that wrongdoing be availible forever as a scarlet letter that David must wear, for all in the world to see, seems to be the only purpose in maintaing the now oversighted information. For those clamoring for its return, or asking for the heads of those responsible for its removal, let me ask these questions:

• What possible purpose does this information serve towards improving encyclopedia articles?
• What possible purpose does this information serve towards preventing David Gerard from committing further wrongdoings given that his access to the tools has already been revoked?

Seriously folks, take things under perspective for a bit. Yes, ArbCom and the other high functionaries of Wikipedia should not act capriciously or arbitrarily in the administration of their duties at Wikipedia. But neither should they be bound slavishly to regulations which prevent them from responding appropriately to unforeen problems. No one should deviate from expected norms without just cause. However, there will always be unique situations which existing norms do not cover. This was a unique and unprecedented situation (given the possibility of real-world harm to the accused) and needed an unprecedented response. It is not ArbCom's job to do exactly what will not make anyone angry. It is their job to make the hard decisions when the community cannot. Should it be a normal course of action to use Oversight in this manner. Absolutely not. However, does a single instance of this sort of use, in a unique and unprecedented situation, represent a capricious dereliction of duty? No. Lets all get back to stirring up random drama at ANI writing some encyclopedia articles. -- Jayron 32 17:52, 2 December 2009 (UTC) reply

Thank you Coren. The silence was deafening. Hopefully in due course we will have a joint statement by all parties explaining the gist of what happened and why it took a long time for the joint statement to be made. I don't expect anything complicated or detailed, a 100,000-foot-level statement along the lines of "x alleges y did z, y does or does not concede that he did z or that if he did he does or does not concede that it was improper" along with "after investigating, the arbitration committee decided...." Also, a statement from any party including ARBCOM regarding "lessons learned" would be helpful but not necessary - lessons learned are most useful as internal documents anyways. Details of exactly what transpired, especially details that would do WP:HARM, IMHO need not be made public.

"In due course" should be days or maybe a week or two, not months. davidwr/( talk)/( contribs)/( e-mail) 02:41, 3 December 2009 (UTC) reply

I should point out that the committee since published a statement on the noticeboard; I would not expect David Gerard will wish to make one of his own given that much of the fuss centered around his desire to have the matter removed from public view. —  Coren  ^(talk) 03:06, 3 December 2009 (UTC) reply

"I love everyone. I suggest you all go write something. Assume better faith too. Be excellent to one another." - David Gerard ( talk) 13:51, 3 December 2009 (UTC) reply

I am 100% in favor of writing something, as long as it entails expanding the wikipedia project and not calling others "to be a waste of skin." I'm no saint, but if I were a public representative of Wikipedia, I would not for a single second think about publicly insulting someone and topping it all off with a blatant and severe privacy violation in regards to user data. I sincerely hope that you will take your newfound position to expand Wikipedia to help make it a more comprehensive online encyclopedia while ceasing the drama mongering, flaming, baiting, and passive aggressive posts entirely. I would also address Mr. Godwin and the Wikimedia Foundation, but I would rather not be permablocked today or receive another baseless and unprovoked threat from an administrator for dissent. Vodello ( talk) 14:53, 3 December 2009 (UTC) reply

The Register has just published a story which sheds a lot more light on this issue and provides much detail that Coren does not mention. The gist is that the Arbitration Committee was strongly advised by Mike Godwin that its intervention against David was unsupportable. Mike's comments to Arbcom-L, quoted by The Register, are devastating for the ArbCom's credibility. Since the e-mails are now in the public domain, I'll quote them here. Mike said in his first comment:

I've been following this discussion, and it seems to me that the case for removing David Gerard's checkuser and oversight functions has not been made in any way that meets what I as a lawyer would characterize as due-process and evidentiary standards... Please communicate to all involved my strong personal and professional preference that they reconsider this decision.

And in a later comment he wrote:

My strong suggestion is that ArbCom reconsider its decisions, which seem more like arbitrariness than arbitration.

That is simply devastating for the ArbCom. Many people, myself included, have complained about the arbitrary decision-making of the ArbCom. Now we have the Wikimedia Foundation's general counsel making the same complaint. How can anyone regard the ArbCom as credible now?

It's time that the current members of the ArbCom realised that they have lost all credibility; the ArbCom itself needs major reforms to eliminate the blanket of secrecy and its lack of accountability. We don't need an ArbCom election; we need an ArbCom recall to clear out the people who have brought us to this point, and we need to to start afresh without the taint of scandal that's dogged this ArbCom throughout 2009. -- ChrisO ( talk) 18:06, 2 December 2009 (UTC) reply

What a heap of o!buggeration. Or as Shakespeare put it "oh, what a tangled web we weave....." FWIW, in English law, Mike Godwin could protest all he liked that he was just giving advice as a good ole boy, but it would never hold up in an English court under presumption of agency. Not that it matters, except that Arbcom should state publicaly that WMFs lawyer has access to their mailgroups, or take him off it. Indeed, if they had run this past him AS their lawyer BEFORE springing it, it would have been a lot better. As it is, they have lost an enormous amount of face, and will have fun getting it back. However, at least the oversight was done according to policy, as libel is one of the authorised use of oversight. Way to not handle things guys. Next time, if it concerns off-wiki behaviour and it's a person who is prominent in RL, get some advice from the brief first. Elen of the Roads ( talk) 18:17, 2 December 2009 (UTC) reply

Two thoughts:

1) I agree entirely about "getting some advice from the brief" but it indicates one of the major problems with arbitration that we've never satisfactorily addressed: the ArbCom is making decisions that may affect editors off-wiki without considering the possible legal repercussions of that, including the damage to those editors' reputations. It's only a matter of time before someone sanctioned by the ArbCom sues and subpoenas its discussions. Given Mike's comments, I strongly suspect that those discussions would not do it much credit if the blanket of secrecy was ever lifted.

2) Note that Mike said in his first e-mail that his advice was his strong personal and professional (my italics) preference. -- ChrisO ( talk) 18:28, 2 December 2009 (UTC) reply

This new evidence shoot many holes in my previous understanding and concerns me on several levels. MBisanz ^talk 18:19, 2 December 2009 (UTC) reply

New evidence? Everything Cade Metz says about Wikipedia has to be taken with a full pound of salt, it is always fed to him by axe-grinders hoping to help him grind his own axe. Guy ( Help!) 09:19, 4 December 2009 (UTC) reply

Are you suggesting Cade Metz forged Godwin's emails?! Guy, you're like a court jester sent in to diffuse the tension. Thanks for the chuckle. Manuel Labor ( talk) 04:11, 6 December 2009 (UTC) reply

Disagree. I'd say this backstairs interference by Goodwin is unacceptable. All of us saw, and still can see, David Gerard's inappropriate blogging. I'd have thought the community would be right behind his immediate removal as a checkuser. Broadly this boils down to David Gerard is part of the old guard and he's able to bring out the big-guns to attack the community-elected arbcom. Unacceptable. Arbcom stand your ground.-- Scott Mac (Doc) 18:23, 2 December 2009 (UTC) reply

In short, everybody is in the wrong, and everybody keeps digging themselves in deeper. Reso lute 18:27, 2 December 2009 (UTC) reply

No, no. I can see what Godwin's afraid of (this is the same reason most English local authorities don't allow their elected members to have blogs - because they'll do something just like this). The right answer would have been an arm-up-the-back request for David to stand down (at least while the unpleasantness was sorted out) coupled with a nondescript announcement. That's how these things are done in England. Elen of the Roads ( talk) 18:29, 2 December 2009 (UTC) reply

Bear in mind that the ArbCom and those with access to Arbcom-L are really the only people who have the full range of information on this issue. The rest of us can only see David's blog (which I haven't read). Mike's clearly-expressed concern is with the process by which the ArbCom came to its decision: "[it] has not been made in any way that meets what I as a lawyer would characterize as due-process and evidentiary standards." Bear in mind that, with only one exception as far as I know, the ArbCom's current members do not have any formal training as investigators or analysts. This is the kind of thing that lawyers and detectives do only after years of training; we give the job to untrained amateurs with no grasp of due-process or evidentiary standards. People are elected to the ArbCom on the basis of being politically acceptable, not on the grounds of any expertise in what they are being asked to do. That is a recipe for trouble, as we're now seeing. -- ChrisO ( talk) 18:35, 2 December 2009 (UTC) reply

Epic win for David Gerard. He lost his oversight and checkuser bits but at what cost to the ArbCom? This has turned into yet another PR disaster, 2008 style, where the arbs are exposed as disorganized and not aware of the consequences of their own actions. To make matters worse, we now see that the leaks of e-mails sent to mailing lists under the control of ArbCom are now once again happening. The lessons to be learned here are: Don't mess with David Gerard and don't e-mail any ArbCom controlled mailing lists unless you want the rest of the world to read your mails as well. Vyvyan Basterd ( talk) 18:30, 2 December 2009 (UTC) reply

"devastating for Arbcom" ? Inasmuch as they should have displayed a backbone and said "butt out" to outside meddling, sure. Otherwise that is a rather bizarre take on these events. Tarc ( talk) 18:33, 2 December 2009 (UTC) reply

With all due respect to Mike Godwin, WP:NOTLAW. Wikipedia is not a moot court. Insofar as David's blog indicates clear problems with his use of the tools, ArbCom acted correctly. Mike Godwin is not David's lawyer, and ArbCom is not a court of law. We elect ArbCom to make the hard decisions, and when they get faced with one, there is 100% certainty that more than half the people will find fault with it. Ask yourselves this question: How many of the people currently claiming ArbCom abuse here would have made the same case for abuse Had ArbCom not acted in the first place. The more and more I follow these discussions, the more I am convinced that certain people at Wikipedia are only interested in convicting ArbCom first, and THEN looking for something to complain about afterwards. If ArbCom had not sanctioned Gerard, we would be faced with many of these same people claiming that ArbCom was protecting him in some way. If they had not oversighted their decision, which contained sensitive information, these exact same people would be asking for ArbCom's head for irresponsibly exposing Wikipedia to potential litigation. It's more and more a case of blasting ArbCom regardless of what they do... -- Jayron 32 18:43, 2 December 2009 (UTC) reply

FWIW, the firm consensus on the checkuser list appeared to that there was no evidence for misuse of the tools, and in fact no evidence that there even *might* have been a misuse of the tools. Please don't continue to repeat the claim. It's wrong and unjustified, and it has even been retracted by those who originally made it. Go ahead and call David the cancer that is killing Wikipedia, if you like... but you don't need to accuse him of breaching trust in order to express your disapproval. Of course, the arbcom is free to twiddle user rights however they like, community tolerance permitting, but not everyone is going to or has to agree with it. But that isn't really news. -- Gmaxwell ( talk) 18:55, 2 December 2009 (UTC) reply

While I agree with ChrisO generally about competencies, I'm not sure that any significant expertise was required. I think this is more down to inexperience in protocol. David told the world the outcome of an SPI and linked it to the real live individual he believed to be socking. That's a clear breach of Wikipedia rules - never mind about use of tools. He should have been "suspended" and a statement could have been drafted that fudged something up about an investigation. The English local government does that rather more times than it would really like, and it's a standardized procedure. Everyone says nowt, the guy disappears into the background, and the report rolls in two years later. In this case, any statement issued should have said the minimum and strictly related it only to wikipedia policy. If David has breached the UK Data Protection Act, the other guy can make a complaint to the Information commissioner. Elen of the Roads ( talk) 18:59, 2 December 2009 (UTC) reply

Er. You're saying these things, but they aren't true. Please point me to Mr. L's edits. Surely you must know the account if David disclosed it. No?

Moreover, the confidentiality afforded to people actually guilty of site misuse is practically non-existent. Checkusers routinely disclose people who have been sockpuppeting, in the process indicating which accounts are linked and linked IPs (where necessary) and such. Without the ability to disclose necessary information in cases of abuse checkuser would be entirely pointless… (or would you prefer we made all blocking actions secret?) though this is all moot in this case because no /checkuser/ derived data was ever disclosed. If you're reading that checkuser was used and the output disclosed then you're reading too much into it. Cheers. -- Gmaxwell ( talk) 19:10, 2 December 2009 (UTC) reply

I am not talking about the tools, and I am not talking about Wikipedia accounts. I'm talking out in the real world. David's tweet said that "<Real live person> carried out fraudulent edits to their Wikipedia article - I know it because I investigated." That's on David's bloody blog - it's not up for debate. What the names of the sock accounts are isn't important. Are you not able to see this? Elen of the Roads ( talk) 20:08, 2 December 2009 (UTC) reply

I'm interested in knowing why you think so lowly of your own position that you took the additional effort of making up words that David never stated rather than the simple and expedient measure of copy and paste. Not only does your contrived version not match what David actually said on twitter, it's longer. Not up for debate, indeed.

In every ruling the arbcom says the equivalent of "<Real live person> edited wikipedia in <random bad ways>. We know because we considered <these facts>." (dead/fake people don't make bad edits). Every administrator on Wikipedia, per the blocking policy, provides an explanation of wrong doing whenever they block. Perhaps I'm misunderstanding you... If you are only taking the position that David's actions were ill-considered, poorly stated, crude, rude, or otherwise socially unacceptable then we're arguing past each other. My only purpose in commenting there was to refuse the clearly wrong statement that private data and/or the checkuser tool was used/abused in this case.-- Gmaxwell ( talk) 20:27, 2 December 2009 (UTC) reply

That's fine. I'll have my wikilawyer talk to your wikilawyer. Meanwhile, in the real world, intimating that (say) a journalist can't (say) keep their trap shut about their sources might have some impact on their reputation...but only in the real world of course. Elen of the Roads ( talk) 20:51, 2 December 2009 (UTC) reply

Interjecting here to make the point that the Register's speculation about WP is always to be taken with a grain of salt. Protonk ( talk) 18:48, 2 December 2009 (UTC) reply

Agreed and I don't have much time for Cade Metz, but the quality of his reporting isn't relevant here; the issue is what's said in the e-mails he quotes. -- ChrisO ( talk) 18:52, 2 December 2009 (UTC) reply

Yeah, that's obvious. Unless we assert that those emails are faked, the issue is that Mike Godwin was involved (something not mentioned in the explanation from Coren above) Err, Coren mentioned it, sorry. Protonk ( talk) 18:56, 2 December 2009 (UTC) Protonk ( talk) 18:56, 2 December 2009 (UTC) reply

Well, it's more than just Mike's involvement, it's the fact that a highly experienced and high-profile lawyer thought the ArbCom's decision in this matter was arbitrary and unsupportable. People seem to be advancing quasi-conspiracy theories about why Mike would say this but I would have thought the simplest explanation is the most likely - that what he believes what he says. I don't see any reason to imply he's a liar or is saying it just for political reasons. I've seen individual arbitrators being reckless and haphazard with evidence, so I'm not at all surprised by his comments; they seem quite justifiable. -- ChrisO ( talk) 19:04, 2 December 2009 (UTC) reply

He is a lawyer, not a judge. Lawyers think what whoever pays them wants them to think. In any case, I don't see how legal training has any reference at all to a arbcom action based on their collective judgement as to whether DG was acting in a matter compatible with his duties. I thought this was a consensus-based project?

With regard to the notion that "Lawyers think what whoever pays them wants to think," who is paying me right now to think that you are wholly ignorant of what it means to be a lawyer? Or to be an ethical human being who also happens to be a lawyer? I'm not a cartoon figure just because you have cartoon notions of what lawyers are like. MikeGodwin ( talk) 05:09, 3 December 2009 (UTC) reply

Moreover, what standing does Godwin have to discuss this with arbcom in a private capacity? MzMcBride's comments in the section above some up my view on this quite succinctly. And - did the community ever give arbcom a mandate to enter into secret deals with third parties?

I do not wish to impute anything by this, but I would very much like to hear a statement, purely for the avoidance of doubt, as to whether Jimbo and Godwin discussed this prior to the comments Godwin is said to have made. This is purely in the interests of full disclosure - I really do not mean to imply anything, but it is a question I would like to know the answer to, and I get the feeling this question may be a bit of an elephant in the room.

Jimbo and I discussed the matter only after I had expressed my concerns on the Functionaries list. (We also discussed them in the office today. MikeGodwin ( talk) 05:09, 3 December 2009 (UTC) reply

Right, that does it - I'm introducing Nazis into this thread and calling Godwin's law on Godwin. Chill out everyone! - Tbsdy lives (formerly Ta bu shi da yu) ^talk 12:33, 3 December 2009 (UTC) reply

Lastly, I am having a hard job seeing as how the cause - i.e. this mess - leads to the effect - i.e. John's resignation. Brilliantine ( talk) 19:25, 2 December 2009 (UTC) reply

Why do you think that that particular effect is a reasonable response to causes? People frequently resign positions, delete their user pages, etc. based on their own motivations which don't always make sense to others. -- Gmaxwell ( talk) 19:31, 2 December 2009 (UTC) reply

Exactly. I don't think it's a reasonable response and would appreciate if he outlined his motivation in greater detail. But the issue of what on earth Godwin and arbcom are doing getting involved in negotiations like this is the most troubling. This is precisely the kind of issue that should be thrown to the community for consensus. If the community decided arbcom's original statement contained errors of fact, it would surely be simpler to agree - following discussion in a public forum - to change it. Brilliantine ( talk) 19:38, 2 December 2009 (UTC) reply

Hm. I think I agree with you, mostly. But the normal course of review for private data tools is the ombudsman, the board appointed body that deals with private data issues, and the checkuser mailing list. One of the reasons that a wiki is not permitted to only have one active checkuser (zero is okay) is precisely because the checkusers are expected to police each other, and the ombudsman exists to catch the cases where self-policing fails. As far as discussions regarding David's Wikipedia intersecting behaviour, I absolutely agree that thats up to the Wikipedia community. But hasn't the community delegated that kind of activity to arbcom at its own prerogative? ...In my day... arbcom only dealt with perennial issues that the community couldn't agree on, like blocking Plautus satire... and it only took cases after the community efforts had failed. It seems odd to me that you're complaining that the community should have handled this, if that is what the community wants why is arbcom even touching all these cases where the community hasn't failed to demonstrate an ability to reach consensus? -- Gmaxwell ( talk) 20:37, 2 December 2009 (UTC) reply

But once the announcement and the sanction is out there, the situation is the same as that following any non-checkuser-related Arbcom case or motion or announcement. And if someone gripes about an arbcom sanction following a case, it's generally left for the community to deal with, no? I don't see how this should be any different. If DG had just piped down and taken the decision, it would have got about one-hundredth of the publicity it has now. This just looks to me like a classing toys-out-of-pram issue. Brilliantine ( talk) 21:19, 2 December 2009 (UTC) reply

Sure. I mean, this whole issue has been blown out of proportion, but that is LARGELY the result of the use of rev. delete to cover up the problem (and only somewhat to do with the community's proclivity for blowing things out of proportion). Obviously the arbcom was reluctant to simply say "we fucked up, mike godwin said our dismissal was out of line, and we retract it". The tools given to them solely to arbitrate cases allowed them to do so secretly. If this were a public mailing list or they did not have the capacity to oversight (basically) the revisions, this debate would not have happened. Protonk ( talk) 19:26, 2 December 2009 (UTC) reply

Disagree with Protonk. The issue is that David is regarded as en.wp's "UK press officer". This creates a presumption of agency (ie it could be argued that he is somehow doing it 'on behalf of' the organisation). Hence the panic. The rest of Wikipedia isn't "real" (in a legal sense). Taking away sysop tools isn't something that a real court in the UK would ever look at - the whole thing is voluntary, there is actually nowhere he could take his case. Elen of the Roads ( talk) 19:02, 2 December 2009 (UTC) reply

The issue raised is the reputational damage. That certainly is actionable, if it can be shown that decisions made by one party adversely affect the reputation of the other party. -- ChrisO ( talk) 19:13, 2 December 2009 (UTC) reply

Well, I mean the added issue. What interests my about the Register article is the suggestion of added centrality to parties in the discussion. Obviously the impropriety of the whole mess is not solely related to that. Protonk ( talk) 19:21, 2 December 2009 (UTC) reply

@ChrisO - quite right. The guy edited under his real name, and apparently with some standing. If en.wikipedia discipline him and then make it public, it will cause reputational damage to him - particularly as the allegation is breach of confidentiality, which is hugely significant in some professions. No one in the real world gives a flying fart about whether or not he used sysop tools, and if the whole thing had been handled more quietly and kept focused on Wikipedia sysop tools, this wouldn't have blown up like this. Elen of the Roads ( talk) 20:12, 2 December 2009 (UTC) reply

Which goes back to your point about getting the brief involved at the start. It shows a regrettably cavalier approach to reputational issues. -- ChrisO ( talk) 20:25, 2 December 2009 (UTC) reply

Um, I have to say that's just too goddamn bad, then. Editing under one's real name is not a shield for bad behavior. Tarc ( talk) 21:45, 2 December 2009 (UTC) reply

Everyone, Wikipedia insider or not should be able to have the expectation that they won't be smeared by *inaccurate* statements on the wikimedia sites. Sometimes fairness comes to the privileged first, but thats all the more ammo in arguing that it should be extended to others. As a result of this discussion here some accusations about Judd Bagley have also been removed. -- Gmaxwell ( talk) 00:35, 3 December 2009 (UTC) reply

• This all amounts to nothing. Were I to be banned and go running to a lawyer who then started emailing the Arbcom with his professional opinion - that would be tantamount to a legal threat. Goodwin's opinion was clearly not sought or required. Brad is a lawyer, they clearly felt a second opinion was not unecessary. This is nothing more than heavy handed bullying and intimidation of the Arbcom because Gerard is one of the chosen few. Well tough luck, he has been fired, if he wants to make legal threats let him leave Wikipedia and pursue the matter in the law courts of London, and good luck to him - he would need it. Nobody would touch such a ridiculous case.  Giano  20:42, 2 December 2009 (UTC) reply

Giano, I believe you are quite right as to the actual merit of any potential legal case. However, lawyers are notoriously cautious, and always willing to find an opportunity to settle. Elen of the Roads ( talk) 20:53, 2 December 2009 (UTC) reply

• In this instance I am 100% correct.  Giano  20:59, 2 December 2009 (UTC) reply

No one has stated that David went to Godwin. I know that I gave my opinion that the claim appeared to be wrong and harmfuly defamitory before hearing about this from anyone, much less David. I too can make up random 'facts' and then get angry over them, but thats no way to have a discussion. -- Gmaxwell ( talk) 20:57, 2 December 2009 (UTC) reply

• No they have not stated it have they. Goodwin often opines on arbcomn decisions and forces them to reverse, doesn't he?  Giano  20:59, 2 December 2009 (UTC) reply

• You haven't had much interaction with Godwin, have you? If I ever see him pass up a good argument I'll be phoning his doctor in concern for his health... (I don't mean this as a criticism— I think his input is usually excellent, except when he disagrees with me, of course) But really... he coined a notable 'law' about arguments on the Internet, strangers to the subject usually don't write widely circulated laws on it. Godwin arguing in my internets? It's more likely than you think. -- Gmaxwell ( talk) 21:10, 2 December 2009 (UTC) reply

Lessons learned?

How sad this all is, and how easy it would have been to avoid. Questions to ask:

1. Was it necessary to post the original announcement in haste? A more carefully written wording might have headed off problems.
2. What exactly was the basis for the deflagging? Confusion on that point is a breeding ground for FUD.
3. Would it have been too difficult to have posted a placeholder at the same time the revdelete occurred? The disappearance of whole threads certainly attracted attention and anything that has the appearance of coverup is going to raise questions.
4. Would the members of the Committee please read and memorize Streisand effect?
5. Would the members of the Committee please be mindful that thoughtful and reasoned criticism is essential to our project?

These last several months there has been too much of a tendency among arbitrators to brush off feedback with a they'll complain no matter what we do dismissal. It is important to remember that there are plenty of Wikipedians--even among the Committee's vocal critics--who find value to ArbCom as a body and wish to see it succeed. The trouble this week is that one of the Committee's core missions is to handle sensitive situations wisely and it obviously made a mess of this one.

As I wrote at the end of l'Affair Orangemarlin, it would be a very useful thing if ArbCom adopted a standard practice of the United States Navy: holding a "lessons learned" session after an operation to see what went right, what went wrong, and what can be done to improve funtioning in future. The Committee neither adopted that suggestion nor responded to it.

One symptom of a dysfunctional organization is that its members depart during crises. Respectfully requesting that the remaining Committee members consider the "lessons learned" model as perhaps yielding better results than its current operational habits. Durova ³⁷¹ 20:47, 2 December 2009 (UTC) reply

• I think the above sounds reasonable. er. Well, what doc says below with respect to NOT elaborating is agreeable to me too. The normal practice is to speak circumspectly and then only respond to statements made by the person whom you're finding against. "We found his behaviour at the company Christmas party to be unbecoming." "Nonsense! I merely gave a speech like everyone else!" "You gave the speech naked." <crickets>. -- Gmaxwell ( talk) 20:59, 2 December 2009 (UTC) reply

• • I suspect that arbcom would do better NOT always to elaborate on their judgements. The minute they got into describing what David had done wrong, they got into trouble. Had they merely said "in view of David's blog posts, we don't have confidence in him retaining the tools" and admitted only their own subjective decision, then they'd have be fine. If David wanted elaboration on their reasoning, he could have been sent that, and it could have been made public only had he wished. That would minimise any inference that arbcom were commenting on the character of a real person - they'd be commenting only on their own willingness to have him as a checkuser - which is a wholly subjective thing. I also think (perhaps it happened) that opportunity to resign tools without any finding should be offered in such cases. Again, I say this with hindsight - not as a criticism of arbcom - who I don't think behaved improperly. With hindsight, they could have done things better here, but all actions have been perfectly understandable and (presumably) in good faith. I think their initial decision to respond to David's blog post was certainly both correct and bold. (I also think the community doesn't so much "always criticise" so much as it works in real 24-7 immediate response time - and forgets arbcom can only act with the consensus of 18 people in different timezones. I do think Mike's role needs some further discussion, but mainly to make sure there is clarity on his position vis arbcom going forward.-- Scott Mac (Doc) 21:34, 2 December 2009 (UTC) reply
• The suggested focus here would primarily be on the Committee's own internal decisions in handling and communicating this decision. Perhaps Mike Godwin ought to have been brought into the dialog at an earlier time or a more skilled communicator could have been tasked with drafting the announcement. Almost certainly, a placeholder statement ought to have gone up at the same time the revdelete was done. And with all the confusion and cross-talk I'm still uncertain about the policy basis for revdeleting all of the discussion threads. It stretches the imagination to suppose that every single one of those posts could be considered defamatory. So much cross-talk is occurring that it's hard to make sense of fact v. rumor, and that's exactly the kind of dilemma that no one but a troll enjoys. Durova ³⁷¹ 21:55, 2 December 2009 (UTC) reply

These are my answers:

1. No. Yes.
2. The original statement was ambiguous to the extent that it could reasonably convey an impression that we did not intend it to do. In issue is a very fine, though very significant distinction. Care (and consequently, time) is necessary to avoid repeating the original mistake.
3. Yes. See the previous answer.
4. Our superpowers are somewhat immune from that effect.
5. Love some. Know where we can get any?

On the topic of revision hiding, yes, it is necessary when hiding a revision to hide all subsequent revisions that contain that revision. -- bainer ( talk) 22:40, 2 December 2009 (UTC) reply

Thank you very much for the replies, including the barbed wit of no. 5. Durova ³⁷¹ 23:22, 2 December 2009 (UTC) reply

Does this mean that one of the duties of the committee is to scoop a giant Wikipedia litterbox when someone else with a lot of stroke around town takes a huge, fowl and downright atrociously toxic dump that has a stink of mammoth proportions (without covering it back up with Wikitty litter! Sigh, men..), then replace the Wikitty litter entirely, spray a ton of air freshener, and make it look like no "bombs" were ever dropped? I apologize that I couldn't think of a better analogy, but this whole thing stinks. Vodello ( talk) 04:32, 3 December 2009 (UTC) reply

• While ArbCom erred on the wording of the motion, I think it is important to remember that ArbCom did what the Wikimedia Foundation asked them to do: in clear, the WMF put ArbCom in charge of the CU and OS permissions on enwiki, so you can't blame them for enforcing standards. Contrary to what do courts - ruling on whether such and such law has been violated, ArbCom makes the CU/OS standards up and enforce them altogether, we have no other way to proceed - because the WMF policies are too incomplete and our policies are descriptive, so you can't compare with real courts, thus neither evidenciary standards or due process apply in a judiciary sense.

Another important point: ArbCom didn't make any arbitration here, they have responsibilities other than arbitrating on-wiki disputes, and the one of overseeing CU/OS is distinct from arbitration. So the process to summarily remove advanced permissions (in evidencing, due process and so on) is not comparable to the one employed in formal enwiki arbitration cases. Cenarium ( talk) 19:02, 3 December 2009 (UTC) reply

• If I recall correctly, many of the "lessons learned" from incidents in the Navy were interchangeable and forgettable. If I had a dollar for every incident report I read which listed "lack of situational awareness" (or "mental attention", that phrase was popular for a while) as a root cause and "improve training for watchstanders" as a "lesson learned", I'd be rich (or at least have a few hundred dollars. :P ). Arbcom is in a pretty stressful position here. The suggestions Durova makes are helpful, but may not point to the source of that stress. The community clamors for haste, demands discussion but (effectively) expects unanimity. Its not hard to imagine a scenario where DG had the tools removed and no announcement was made (because it was being crafted) which would have resulted in angry speculation and froth on these boards. What I worry about is the transition between closed discussion and open presentation. Explosions like this one should push arbcom away from closed lists (especially closed lists with leaks to the Register, hah), but it won't. I suspect (as do others) that the sturm and drang over such a minor issue will provoke a retrenchment regarding openness. The public face of the committee will have to be more composed and scripted because we (the community) obviously don't tolerate delays and slip-ups. Please, please, please fight back against this temptation. Transparency is ugly, painful and uncooth. Most (as bainer points out) of the feedback will be...unhelpful. But it is a cornerstone of the project and it is the only way to avoid these silly disputes and flare ups. Protonk ( talk) 19:28, 3 December 2009 (UTC) reply

Seconded. Arbcom's remit is always going to include things that need private discussion, and I strongly suggest that they establish a venue for private discussion rather than the present venue that's only accessible to trusted functionaries and drama-mongering shit-stirrers. For anything that doesn't need privacy, you can disarm the shit-stirrers by depriving them of the ability to post your business before you do. That's a tough thing to do, and in the real world private channels always gradually wear away at promises of openness (they are just easier), but if Arbcom will take seriously the need to be as open as possible, it can avoid explosions of this type. — Gavia immer ( talk) 20:45, 3 December 2009 (UTC) reply