Wikipedia talk:Arbitration Committee/Audit Subcommittee/Archive 2

Archive 1

Archive 2

Archive 3

Archive 4

}}

Request for discussion on how to handle sockpuppet investigations involving large institutions

Hi, I was pointed here by User:AniMate. Recently I was the victim of collateral damage involving a sockpuppeteer who was editting from my university. That fact being unknown to me, I commented in an ANI thread started by the user, who claimed to be the victim of a jewish conspiracy or some such. My (deleted) edit can be seen here. Regardless, when he was banned, a c/u was run and my account was confirmed as a sock, due to the fact that we were both editting in the same IP range. While I don't fault User:J.delanoy for the inital block, as the checkuser did confirm me as a sock, it became quite a stressfull ordeal trying to get it overturned. Much of the difficulty arose from the high esteem in which c/u requests are rightfully held. While they are almost always accurate, there are situations such as this where they can return false positives. It is for this reason that I wish to discuss how SPIs can be conducted where large institutions with thousands of users on a limitted IP range are involved. Specifically, I'd like to see if we can find a way to both protect the integrity of the project, as well as reduce the instances of innocent editors being caught in the crossfire. I admit, most of this arose from simple bad luck, with me posting in an AN/I thread started by a fellow student. However, my experience leads me to believe that I can't be the only one this has happened to. Would it be possible, upon ascertaining that a suspected sock resolves to a large institution (a major university in this case), to suggest that greater weight be given to circumstantial evidence, such as edit history and positions taken? I'd be grateful to here everyone's opinion on this. Thank you, Throwaway85 (talk) 03:15, 4 December 2009 (UTC)

I wonder if the IP in question was marked as a shared IP. We do have templates to mark IPs that we know to be shared, this helps us make decisions when blocking based off of IP to avoid this type of collateral damage. However, only a checkuser has access to that IP for privacy reasons. If there is a checkuser in the house, was that IP in question marked as a shared IP? ^Chillum 03:44, 4 December 2009 (UTC)

That would be helpful, but it would be the IP range itself that would need to be marked as shared, as there are hundreds that all get lumped into the pool. If it helps, I can ask the network admins for the exact list, and provide it here. I'd still like to find a more general solution, however, for future occurrences of similar situations. Perhaps a closer relationship with network admins might be in order when dealing with large institutions. It may be possible to get Abuse Investigations departments to limit the user themselves, rather than block an IP that is shared by thousands. I specifically recall a case of vandalism on Gordon Campbell where the IP resolved to the BC parliament buildings. When I contacted their abuse department, they informed me that the IP in question belonged to a specific school district, and that they would follow up themselves. This might be a more appropriate course of action than simply blocking the IP.Throwaway85 (talk) 04:26, 4 December 2009 (UTC)

I'm going to reiterate what I said on my talk page. I very sincerely believe that there is no systematic way to remove or lessen the number of false positives. Every checkuser case is absolutely and utterly unique; even if the same case is run by different checkusers, they will often come to different conclusions, or some may find things that others do not. With so much variance, there is practically zero chance that any coherent method to approach such things could be found. And let me assure you that Throwaway's case is far, far less complicated than many that I deal with. Your university has, including IPs, maybe a few hundred editors within the last three months. I very, very frequently deal with ranges that had upwards of five thousand editors in three months. And it is not unheard of to find places with more than five thousand in one month. With regard to the case at hand, I am still not 100% convinced, but the opinion of the other checkusers is that I was wrong, so I will accept that. I have dealt with at least two similar cases in the past, and in both cases, my gut feeling was wrong.
I neglected to do so earlier, so I would like to offer Throwaway85 my apologies for my block and implied accusation. I guess a botched CU/block is a fitting way to end this day of blunders. J.delanoy^gabs_adds 04:46, 4 December 2009 (UTC)

No worries J, apologies accepted and appreciated. I don't blame you for trusting your gut in the slightest. I was wondering, however, if you had any comment on my suggestion of closer collaboration with network administrators in these situations, rather than blocking shared IPs outright, especially in cases where no ongoing disruption or vandalism is occurring? Throwaway85 (talk) 04:58, 4 December 2009 (UTC)

As far as the general case is concerned, it would be difficult to do more than issue a general reminder to the Checkusers to use caution when dealing with large institutions. There are so many different permutations involved. Some schools and businesses run all their web traffic through a single server. Some have stable IPs, some have dynamic IPs. The response of IT departments is highly variable; most will not give us the time of day, or answer someone who is only a "volunteer", no matter how highly placed. Thatcher 05:06, 4 December 2009 (UTC)

Fair enough. I certainly am not trying to change any particular policy, as the incident of false positives is likely too low to warrant putting up with all of the gaming and wikilawyering that could result. I'm merely trying to see if we can think of some ideas to both protect the interests of the project, and ensure that as few people are incorrectly labelled as socks as possible. Thanks for your input. Throwaway85 (talk) 05:16, 4 December 2009 (UTC)

Checkusers are often requested to find the "real account" behind IP vandalism or disruption (see here for a recent example I was involved with). Sometimes we get lucky and it is obvious, most of the time it is not, which is why template:Fishing exists. Unless there is a direct IP match or certain obvious "tells", there is some element of risk in connecting an IP to a user. I do think that checkusers should be mindful of those risks and consider whether the IP edits are disruptive or damaging enough to be worth taking whatever level of risk exists in a particular case. Thatcher 05:21, 4 December 2009 (UTC)

As a regular patroller of CAT:UNB, I have been somewhat passively following this case to see how it turns out. My only contribution to the question at hand about false-positives is that, in most cases, WP:DUCK is actually a better test than pure checkuser would be. In general, many disruptive socks are easy to spot because they have favorite topics, OCD-like adherance to often rediculous ideas or behaviors, gramatical "tells" or other such easy to spot stuff. No endeavour is likely to have 100% success, and the checkusers involved should not be faulted in any way, but care must be taken that behavioral AND technical evidence is analyzed, and I would posit that in many cases, the behavioral is the better and more reliable part... --Jayron32 05:30, 4 December 2009 (UTC)

I concur. Technical data is quite easy to manipulate, but it takes a truly dedicated socker to successfully alter their mannerisms and interests. That's the main point I was making in my defense: That I and the user in question were on opposite sides of the debate, that I openly derided him as a conspiracy theorist (bad me), and that I had never encountered the users he had an issue with or editted on topics he editted. My frustration arose from how readily those arguments were dismissed in favour of the c/u result, which I openly admitted would be positive given that we were editting from the same location. I'm unsure of how what action to take in regards to this, however, as I am wholly unfamiliar with the process and don't wish to make an ass of myself by proposing something silly. Throwaway85 (talk) 05:40, 4 December 2009 (UTC)

The reason I dismissed your edits was because I have seen many cases where someone deliberately argues against their own socks in an attempt to throw off suspicion. I thought you were attempting to do this. Actually, the fact that you had posted on the thread made me more sure that you were socking. J.delanoy^gabs_adds 05:46, 4 December 2009 (UTC)

Fair enough. Like I said, the main culprit here seems to be bad luck. I only posted in that thread in response to Crafty's tongue-in-cheek claims that they were in a cabal. Just my luck that the OP was a couple hundred feet away from me at the time. Had I known I might have slipped a laxative in their drink. Throwaway85 (talk) 05:51, 4 December 2009 (UTC)

J, are you able to tell us anything about what led you to investigate Throwaway85? Perhaps some information about what made you suspicious would be helpful to all and give a little insight into a CU process that seems like high magic to a lot of us.--Wehwalt (talk) 05:36, 4 December 2009 (UTC)

I saw an IP come out of nowhere and post an attack against an established user on a noticeboard. the IP was obviously not a newbie, as he evidently had a grudge against the user in question, I checked the IP and its range to see if I could identify a master. The results of my queries, combined with circumstances surrounding the issue, led me to think that TA85 was behind the attack. J.delanoy^gabs_adds 05:44, 4 December 2009 (UTC)

No worries. I'm really less interested in assigning blame in my particular case than I am in seeing what future improvements we might make. My case has been settled, and I fully accept that everyone involved was acting in good faith and with the best interests of the project at heart. If anything, cases such as these shed light on how two honest people can approach the same situation with wildly different results. Throwaway85 (talk) 06:00, 4 December 2009 (UTC)

After reading through the relevant discussion on AN/I and the talk pages, I am impressed that this did finally get resolved amicably - but in the spirit of reviewing our processes to improve them, I can identify the following specific failures:

• Excessive reliance on browser information: [1] [2]; it needs to be recognised in the future that there can be a plausible, common and innocent explanation for this sort of behaviour, and it is not conclusive.
• Lack of a true block review in these cases: this is not a review at all, merely a reconfirmation. If admins are going to quasi-automatically reject checkuser-based unblock requests (i.e. with no review of either technical or behavioural evidence), then it makes it incredibly difficult to reverse false positives - and false-positives will always happen, as J.delanoy points out.
• Too much weight was then given to the first quasi-automatic block review: more a personal mistake rather than institutional, but this is worrying - given that the first unblock requests were rejected without any real re-examination, discouraging further appeal is a little unfair. When the block is indefinite and it is agreed false positives happen, at least one genuine re-examination of the behavioural evidence is in order at some point.

I'm pretty sure any checkuser reading this case will remember it next time they have similar evidence, so I doubt this over-reliance on browser data will reoccur often if at all. However, admins automatically treating checkuser decisions as utterly infallible truth is a slightly harder attitude to adjust, and I'm not sure how it would be best to do this. They somehow need to be reminded that mistakes can happen, and that they can always review behavioural evidence themselves. Knepflerle (talk) 10:28, 4 December 2009 (UTC)

Those are salient points, Knep. In the absence of a reliable mechanism to reduce false positives, should we instead focus on ways of ensuring that reviews are full and fair? I realize that the much higher population, and turnover, of admins as opposed to checkusers means that the message is harder to get across. Aside from the lower reliance on browser information, how can we find ways to ensure that a) fewer users are incorrectly identified as socks, and/or b) the process of appealing a checkuser block is more accessible and open? I get the feeling that, had I not had at least a basic understanding of how the network operates at my school (although I was mistaken on several counts), and where to go to seek proof of my innocence, that my block would never have been overturned. How can we better aid editors who don't have that knowledge? Throwaway85 (talk) 11:06, 4 December 2009 (UTC)

One thing we could usefully do is dispel the Wikipedia urban legend that the checkuser tools allow checkusers to see personal information, and so the checkusers knows for certain who editors really are. They do allow the user to see technical information that is not normally published by mediawiki - but I was a moderator on a browser support forum a few years ago, and we ran software that showed IP and browser details for every post - heck, we used to add the browser details as part of the post as I recall. We should be clear that while in the course of a sockpuppet investigation the checkuser may sometimes figure out the rl identity of an editor, the checkuser tool does not reveal the editor's name and address. J.delanoy's line above even if the same case is run by different checkusers, they will often come to different conclusions, or some may find things that others do not should be included in the page on checkusers, and be part of the admin training. This is not to lessen the regard for checkusers - it's a painstaking detective job and few people would put the effort in - but so that it is clear that at the end of the day, the judgement is that 'in the opinion of the checkuser' this person is a sock of X. Elen of the Roads (talk) 13:05, 4 December 2009 (UTC)

Another good point. At the end of the day, I question how many admins, especially those from non-technical backgrounds, truly understand both the powers and limitations of a checkuser investigation. For that matter, I myself am unaware what they have access to. Obviously there is browser and IP data, in addition to good ole' sleuthing, but what else is there? Do they regularly perform traceroutes? Do they have access to the actual site access logs? Can they tell, for example, the MAC addresses used (or purported to have been used) when edits were made? If not, the building of functionality into their toolset to check for that could be usefull. While I am vehemently opposed to giving anyone actual access to that information, having the ability to run a tool that checks against it, perhaps with a server-side request, could prove usefull. Throwaway85 (talk) 13:20, 4 December 2009 (UTC)

Perhaps another possibility is look into how a checkuser's judgment is implemented. In this case, Throwaway85 was blocked without warning, indef. There might be useful discussion as to whether it is better to inform the the individual that a checkuser investigation has found such and such and ask for his response before making a determination and before blocking. Additionally, WP:SOCK does not mandate an indef block for the supposed sockmaster, there is considerable discretion. Just some thoughts.--Wehwalt (talk) 13:13, 4 December 2009 (UTC)

I concur. Speaking from experience, it is incredibly frustrating to go to make an edit only to find that you have been blocked, and your only contact with the rest of the community is through your talk page and the request for unblock notices. Perhaps, in cases where there is not ongoing disruption/vandalism, checkusers (and even blocking admins for that matter) could be encouraged to notify the editor in question and seek comment before blocking. This however, brings up my previous point about people who are not tech-savvy not knowing how a checkuser determines their results, and therefore not knowing how adequately to defend themselves. Throwaway85 (talk) 13:20, 4 December 2009 (UTC)

Hard cases make bad law, as we lawyers say. However, I concur with Throwaway85. I am not tech savvy, and if a checkuser determined I was a sock (now, that would be a scandal), I don't know what I would do next except protest my innocence. Perhaps thought should be given to informing the user directly of the right to appeal to ArbCom, and also of not blocking if there has not been disruption pending the outcome.--Wehwalt (talk) 13:28, 4 December 2009 (UTC)

(edit conflict)Dear god no. ArbCom has enough to do without having to deal with every sockpuppet ban out there. I could see some serious potential for problems there. Perhaps, however, a template could be placed on the user's talk page, explaining that a checkuser has determined that they are a sock, what a sock is, how these things are determined, and how to go about appealing it. Throwaway85 (talk) 13:36, 4 December 2009 (UTC)

Perhaps better then to have a link to the relevant language under WP:BLOCK.--Wehwalt (talk) 13:44, 4 December 2009 (UTC)

Except, in most cases, the sort of evidence that the checkusers saw WOULD have been taken as clear abuse. "Good hand-bad hand" accounts are quite common, and it is not unusual to see one user having his two accounts argue with each other to cover up his actions. I don't think we can (nor should we) do anything from a policy standpoint. Just a general "hey, we screwed up a bit here, lets do better the next time" sort of lesson needs to come from this. --Jayron32 13:33, 4 December 2009 (UTC)

And that may well be the case. If, however, we can come up something that can do some good while not introducing more rule creep and still protecting the project, I think that could only be good. Throwaway85 (talk) 13:36, 4 December 2009 (UTC)

Yes, Throwaway85's case is unusual. No harm in discussion though. Incidentally, I applaud the good work the Checkusers do.--Wehwalt (talk) 13:44, 4 December 2009 (UTC)

How about this: Edit the block templates so that clicking on "guide to appealing your block" takes you directly to the appropriate section, in this case [3], and edit this particular section to include a very short, basic rundown of how the match is made, ie IP, user history, browser data, etc. At least then editors would know what this mysterious "checkuser" business is all about. It also would streamline the appeals process slightly, and hopefully lead to fewer people making stupid unblock requests (although I know those are the easiest to deal with). Throwaway85 (talk) 14:00, 4 December 2009 (UTC

@Jayron32 - there is a valid issue over the unblock requests. If admins defer to the checkuser, how do we ensure unblock requests are considered correctly in the unusual case of a mistake. A lot of admins would have disabled T85's ability to edit his talk page because he wouldn't shut up. In the end it was only because an admin was prepared to conclude that his explanations sounded plausible that he was unblocked. This feels like there is a sense in which he was lucky. Elen of the Roads (talk) 14:39, 4 December 2009 (UTC)

Several checkusers were involved in checking out the unblock request and looking over the original findings. Thatcher 14:45, 4 December 2009 (UTC)

And I note that the unblocking admin is a checkuser. I think the following are worth considering:

1. Wherever possible, ask for a response before blocking the supposed sockmaster.

2. When extended consultation among checkusers is needed before a determination of sockpuppetry can be made, or an unblock request dealt with, a note to the effect of "We're on it. Please be patient." would be appropriate.

3. The preceding suggestions regarding alterations to the block template seem worthy of consideration.

Thoughts?--Wehwalt (talk) 15:02, 4 December 2009 (UTC)

It is an unusual case in that an SPI wasn't opened by a third party (ie a non-checkuser ) - in the majority of cases, someone else has noticed the sock-like character of the posting - and there was no quacking to point at, and it features the kind of organisational setup where a user might legitimately edit from a number of different machines, and different IPs in a range. I can't see it happening very often. It does clearly show that checkusers are prepared to confer amongst themselves, and offer and accept second opinions, which is a heartening conclusion. I think there is merit in your suggestion of an 'under investigation' note, rather than leaving the user feeling that they are being ignored. I also think that reviewing admins should be prepared, in the very small number of cases where the blocked editor isn't obviously quacking, continues to civilly protest their innocence and is offering some kind of technical explanation of this kind (not my kid brother dun it) to recuse and hand it off to the checkuser cabal to look over again.Elen of the Roads (talk) 16:07, 4 December 2009 (UTC)

It is a fair point. If it is entirely initiated by one checkuser, we need a bit more to allow for the possibility of error, and also for the sake of appearances, to have it looked at independently if there is objection by the alleged sock. I think the system worked well, but ... we could easily have had an annoyed admin block Throwaway85's talk page and we would have very likely lost him to the project. We are too much indebted to the event for our acquittal, to paraphrase Jane Austen. I think checkusers have taken it to heart, and I think Elen's suggestions are excellent.--Wehwalt (talk) 16:21, 4 December 2009 (UTC)

If a block is based on evidence that only a checkuser can see, then perhaps we should have an unblock template that only other checkusers investigate. Perhaps this is not workable, but it would help ensure that users blocked on unavailable evidence get a proper hearing. ^Chillum 16:47, 4 December 2009 (UTC)

Agree. In fact, in this case, it's not just that only checkusers could see it, but I'm not sure how many people would understand what the discussion was about (eg what was the significance of Firefox 2.2) as it did need a level of technical knowledge. The vast majority of socks are detected by quacking, and usually quack even louder in their unblock requests. It's only where the suspicion of socking is based more or less only on checkuser evidence - and the technical end of it, not just "well, you're posting from the same IP as User:Iquacklikeaduck, that it would be needed Elen of the Roads (talk) 17:00, 4 December 2009 (UTC)

Alternately, just wait a bit, and a checkuser patrolling RFU (I can't be the only one) will find it and take care of it? --jpgordon^{::==( o )} 18:34, 4 December 2009 (UTC)

As indeed you did, and as any checkuser would. The risk to the editor in these very unusual situations is that they have their talkpage blocked by an admin who thinks they are just trying it on like any other socks. It's finding a balance between protecting an editor in these circumstances, and overloading another layer of bureaucracy onto the system. Elen of the Roads (talk) 18:46, 4 December 2009 (UTC)

This case is, at the most basic level, nothing more than me being a complete idiot. There is no systemic problem with checkusers at large that I know of, with regard to checks being made on editors who edit from schools and so forth. This is, quite simply, me not paying attention. I can't do much more than apologize for what I did, but I would like to say that, for the record, this case should reflect solely on me, and should not be viewed as a problem with other checkusers. If anything should be done, it should not affect the others, it should affect me, and me alone, as I am the only one who is at fault here. J.delanoy^gabs_adds 18:50, 4 December 2009 (UTC)

• I wouldn't worry about it. When I edit from work, my IP address is the gateway of an educational internet provider which is shared with about 600 other institutions (many of which are 1000+ student schools), so if you CUd me I'd probably have sockpuppets numbering in the 100s. Happens. Black Kite 18:56, 4 December 2009 (UTC)

I don't even think there's any fault to be placed. Those of us working CU and RFU have to make judgment calls frequently. Every once in a while -- quite rarely, actually -- we make a mistaken call. But: perhaps protecting a user page because they've questioned a checkuser finding repeatedly shouldn't be done unless at least one other checkuser has verified the finding? --jpgordon^{::==( o )} 19:07, 4 December 2009 (UTC)

That seems reasonable. Keep in mind, however, that J had requested comment from another c/u, who agreed with his findings. In my mind, this wasn't a question of a bad c/u result, but rather of a somewhat flawed appeals process that severely limits the ability for an accused editor to properly defend themselves. Much of the difficulty in addressing this problem arrises from its very rarity, as we must ensure that no corrective measure for these few cases jeopardizes how the system is normally meant to work. I think the general visibility of this case amongst the c/u community should help reduce future instances of wrongful conviction, and the changes to the unblock template, etc, should make appealing the process easier. If anything, perhaps a reminder to admins not to step too deeply into c/u decisions might be appropriate. Also, what would an appropriate course of action be, had my third appeal been denied, and my talk page locked? Should I have come back, evading my block as an IP editor, in order to bring the matter to ArbCom? Throwaway85 (talk) 21:17, 4 December 2009 (UTC)

"J had requested comment from another c/u" -- perhaps I missed that; who? --jpgordon^{::==( o )} 23:20, 4 December 2009 (UTC)

Not sure, see [4].

I think you'd still be able to email ArbCom in that case. -- Atama頭 23:19, 4 December 2009 (UTC)

If a talk page is locked after several denials, then, should a link be provided to email arbcom in the locking comments? Throwaway85 (talk) 23:26, 4 December 2009 (UTC)

Definitely; Arbcom loves these problems! ;-)

John Vandenberg ^(chat) 23:42, 4 December 2009 (UTC)

I generally make it a point to give the unblock e-mail address for anyone whose page I lock while they are blocked. ^Chillum 23:46, 4 December 2009 (UTC)

A prudent move. I think the problem we're dealing with is when people forget to do so, or when an editor otherwise falls through the cracks. I'm not familiar with the process, is there a template issued or just page protection? Throwaway85 (talk) 01:16, 5 December 2009 (UTC)

Kinda depends on how annoyed we are by the user... --jpgordon^{::==( o )} 02:25, 5 December 2009 (UTC)

I laughed, I cried, I peed a little. Throwaway85 (talk) 09:58, 5 December 2009 (UTC)

I agree with J. Delanoy's assessment. ;) I feel a little ashamed because I noticed the block, thought it was a bit dodgy but didn't say anything (partially out of trust in checkusers and partially because the editor who got blocked seemed like s/he might have fit the bill). In retrospect, I should have said something. LARGE (>1000 users) institutions queried over a LONG (> 1 month) period are likely to have multiple edits to wikipedia. The possibility of a collision is high enough that we ought to exercise an abundance of caution when performing checkuser blocks. I'm sure the checkusers know this, but I felt the need to apologize/bloviate. Protonk (talk) 02:42, 6 December 2009 (UTC)

I appreciate the mea culpa, but there's nothing to apologize for. As I've stated before, the optics of the situation were decidedly against me. Rather than assigning blame, I'd like to discuss how we can avoid it in the future. Your point about large institutions is salient, and it's that case that I'm looking to discuss. The fact that I commented in the same thread as the blocked editor did me no favours, but there must be a way to reduce the occurrence of these situations in the future. Throwaway85 (talk) 03:05, 6 December 2009 (UTC)

Regarding disputed oversights for an Arbitrator.

How long does it take to ask one simple question? [5].  Giano  08:05, 17 November 2009 (UTC)

Tick-tock. Hipocrite (talk) 13:24, 24 November 2009 (UTC)

It is the job of the Audit Subcommittee to investigate what happened, and that means agreeing on a set of facts and conclusions. Now, we all already know that no outing occurred and that all of the revisions were unsuppressed within hours, which means that while speed is important here as always, this case has no more urgency than the other, older ones that AuSc had before the incident occurred. In addition, we are (or at least I am) currently in a transitional period. Having said that, there has been significant effort put into this already, and I will work on resolving this case within a few days. Dominic·t 07:20, 25 November 2009 (UTC)

Do you need some help with this? What can us perfectly normal users do to get some clarity here? Are any of the implicated individuals running in an election, currently? It's been weeks - almost a month - at this point. Hipocrite (talk) 16:01, 1 December 2009 (UTC)

• Since this popped up on my watchlist again, I'm changing the section title from "illicit" to "disputed". There is nothing about the actual suppression by DerHexer that was questionable; when in doubt, removing edits first and then discussing them is the appropriate and desired model. Suppression can be reversed, and I'm sure you and everyone else would like the benefit of the doubt if you ever make a request for suppression. There are things about the nature and timing of the request by the Arbitrator that need to be sorted out, but the word "illicit" prejudges the incident and really bothers me. Thatcher 16:27, 1 December 2009 (UTC)

I don't care about the header. When will we get a report? Can we have a timeline of events? Is there a dispute? If there is, at this point, weeks, closing in rapidly on a month, perhaps the audit subcommittee should issue a majority and minority report. Or three reports. Or 6! Or disband itself. An Arbitor told Giano "The audit subcommittee is looking at this right now. You need to be patient and wait for the report (even if it takes days or weeks)." Was he lied to? Weeks is about to come due. Hipocrite (talk) 17:08, 1 December 2009 (UTC)

• Thatcher is wrong. Quite a lot was "questionable" which is why I questioned. It seems the Arbs always want to be very silent and quiet where some people are concerned. Just look at how much D Gerard has been allowed to get away with over the years and anyone questioning him being firmly stamped on. Disgraceful behaviour should never have been allowed to continue so long. However, better late than never; such just and needed retribution certainly won't happen next year when the Arbcom returns to Jimboesque form. People would do well to remember that when voting this year. All in all the Arbcom currently have a lot to deal with, so I am prepared to cut them some slack, but while I have no doubt the report, when it finally comes, will be stupefying in its banality and denial, I would still like to see it - eventually.  Giano  19:34, 1 December 2009 (UTC)

No one is wrong about their own personal opinions. However, I think my opinion on the general desirability of suppressing marginal edits immediately, and then discussing them later, is more defensible than the alternative. A great many marginal cases are reported to the oversight mailing list. If there was a requirement that all marginal cases were discussed and approved first, then some marginal cases, that were eventually deemed appropriate for suppression, would be in the public view for much longer than needed, possibly resulting in preventable harm to editors and private citizens. DerHexer may have acted on partial, incomplete or even incorrect information provided to him by others, but I do not believe his actions can be faulted. If you read my above comment in its entirety, you will notice that I said there are issues still to be addressed. Thatcher 19:42, 1 December 2009 (UTC)

We are, in fact working on it. Our arbitrator members appear to be a bit distracted with their docket, and I've been distracted by family and the elections, but I think we should see some movement soon.--Tznkai (talk) 19:52, 1 December 2009 (UTC)

Not to be a pest, or anything, but when is "soon?" Deadlines, and what not, at this point, seem apropos. It is timley and important given our current election. Hipocrite (talk) 19:56, 1 December 2009 (UTC)

I honestly don't know. Just speaking for myself WP:AUSC is high on my wikipedia priority list, but low on my overall priority list, and I believe rightfully so.--Tznkai (talk) 20:08, 1 December 2009 (UTC)

Look, that's fine. I'm not assigning work. I'm asking that the AUSC either shit, or get off the pot. We can go back to public yelling if there's not going to be an answer - that was the old way that checkusers and oversighters were not held accountable for their actions. To be clear - there are candidates in the arbcom election that are either being unfairly tainted by association to this boondogle, or that are being given an unfair free pass. The comittee pledge to "keep the complainant ... informed of its progress and expected date of completion." Unless you missed it, I sent in an official complaint. I have not been kept informed of your progress and expected date of completion. We are rapidly aproaching the three week hard-deadline that you set for yourselves. My complaint was submitted 2 weeks, 4 days and 45someodd minutes ago. Hipocrite (talk) 20:19, 1 December 2009 (UTC)

<--Unless Giano has information that I am not aware of, this issue does not involve any of the current Arbcom candidates. The intermediary who passed Rlevse's request to DerHexer on IRC is not a candidate, and I am unaware of the involvement of any other candidate. Thatcher 20:22, 1 December 2009 (UTC)

Is he not a candidate Thatcher? How do you know this? I was told it was a closely guarded secret. I would prefer yo know for sure, not from heresay, and you appear to be hearing and saying. Secondly, Coren is a candidiate, and his bullying and trying to hector and threaten me into silence is certainly something that needs airing and investigating.  Giano  22:52, 1 December 2009 (UTC)

I do indeed know who it is and it is not a current Arbcom election candidate. Thatcher 23:46, 1 December 2009 (UTC)

Thanks. I don't think it would be appropriate to play 20-questions with this process, given that I've been yelled at for playing 20-questions the last time I was looking for some accountability so I'll stop now and wait the 2 days 23 hours before asking you all what the next step is. A complaint to the audit auditors comittee? Hipocrite (talk) 20:27, 1 December 2009 (UTC)

The user who acted as intermediary is bigger than a breadbox. Does this help? Thatcher 20:40, 1 December 2009 (UTC)

I'm reporting you to the audit comittee for admitting that it was Breadcrab. Hipocrite (talk) 20:45, 1 December 2009 (UTC)

Using that method of analysis, it could have been Breaded Spider, Thatcher or even Hipocrite. J'accuse! Thatcher 21:26, 1 December 2009 (UTC)

Oh no Thatcher, you are wrong again, it was a Beaurocrat! That's why the oversighter beleived all he was told. I thought everone knew that.  Giano  22:56, 1 December 2009 (UTC)

I do know who the IRC intermediary was. Thatcher 23:46, 1 December 2009 (UTC)

Oh Good, well now go and find out why every edit the Arbcom has recently made concerning David Gerard's behaviour has just been oversighted. That would be far more interesting. I of course, already know, but others will want to know too.  Giano  00:08, 2 December 2009 (UTC)

I do know. I want to see if it plays out further before I say anything. Thatcher 00:14, 2 December 2009 (UTC)

My advice, would be for them to tell the truth why they have oversighted the entire David Gerard affair (cancelled a day of their lives) you see, otherwise there will be speculation. The only good thing is that speculation is always less mundane than the truth. Which in this case is more....how can I put this....."delicate."  Giano  00:20, 2 December 2009 (UTC)

I agree that in this case, the mundane explanation is preferable to unanswered speculation. It is not my place to make such an explanation, except perhaps in a dire emergency. Better to let Arbcom figure out what to do on their own than to be forced into a course of action by an unrelated third party. Thatcher 00:26, 2 December 2009 (UTC)

Does this subcommittee have a mandate? Are they fufilling it? Pings sent to all members. Hipocrite (talk) 19:49, 8 December 2009 (UTC)

The mandate of the subcommittee is clearly stated at Wikipedia:AUSC#Purpose. From what I have seen, this is being fulfilled by the subcommittee. Vassyana (talk) 22:37, 8 December 2009 (UTC)

The subcommittee is discussing all pending matters and I have reason to believe that we will issue decisions or reports on all of them to the parties soon, hopefully by the weekend. Newyorkbrad (talk) 14:57, 9 December 2009 (UTC)

I hope you all had a great weekend. Would you like to propose disbanding this committee, or do I need to do that? Hipocrite (talk) 05:46, 14 December 2009 (UTC)

I do find it ironic that you want us to rush to a conclusion about whether or not somebody was hasty in their use of the tools. - Jredmond (talk) 18:23, 14 December 2009 (UTC)

"Randy in Boise" suppressions - AUSC report - Discussion

Original report

Well, I've just read the report and I honestly don't know whether to laugh or cry. The report says absolutely nothing that was not already known, and fails to address any of the important issues raised by this incident. Just a few issues that I notice:

1. ...an editor posted a comment that included a reference to a Wikipedia critique [6] from some years ago... -- this certainly sounds like an obscure reference. Might it not have been more balanced to mention that the critique in question is quoted on an on-wiki page and is quoted at the top of the talk page of editor 1? How about it being included and illustrated in WikiSpeak? Isn't the fact that the critique was published three and a half years ago relevant to the issue of its obscurity? What about the fact that the critique is referenced by other editors in places like Village Pump on 14 October 2009 or at ANI on 26 May 2009, or WP:AE and is certainly known to a reasonably-sized group of editors?
2. ...this reference was incorrectly but reasonably interpreted by another editor as being an attempt to reveal personal details about the second editor. -- which immediately raises the question of whether this "second editor" is likely to be aware of the critique in question. Now, for a random editor it might indeed be reasonable to assume that the reference was obscure. However, that is harder to accept when one considers that the second editor is a member of ArbCom, and the first is a noted critic of ArbCom. Has editor 2 really never had cause to visit editor 1's talk page? Never noticed any of the references on ANI? Has there really never been a reference in any place this long-term editor and Arbitrator has seen - not even in Wikipedia:Governance review/Proposals#Discussion of Direct Election of Arbitration Committee which is a topic surely of interest to an Arbitrator?
3. ...an attempt to reveal personal details... -- which presumably relates to editor 2's first name. Of course, the fact that this name has been useed on wiki before - including by other Arbitrators on case pages - suggests the name is not secret. Surely otherwise the AUSC should be accusing ArbCom members of outing? Editor 2 (aka the Arbitrator) must have known his or her first name was known on wiki and has been used, so why did she or he think this use of his or her name was any form of outing?
4. ...The second editor, aware that there is a significant time factor involved in suppressing edits to AN/I, alerted both the Oversight mailing list (Oversight-L) and contacted another administrator on IRC to try to locate an oversighter for suppression of the edit that was mistakenly thought to be an outing attempt... -- but editor 2 was perfectly capable of carrying out the supression. If there was a genuine belief that an arbitrator had been outed and time is so critical, why not act to protect oneself directly? Surely acting unilaterally to protect oneself from outing would withstand scrutiny even if it was technically out-of-process, a clear example of IAR in action... unless the claim of "attempted outing" itself would not withstand scrutiny? Also, why the secrecy over the identity of the "other administrator"? It leaves open the question of whether the administrator might be expected to be aware of the critique.
5. Other issues that remain unaddressed:

• is it true that editor 2's request for oversight was denied by en.wiki arbitrators and/or oversighters before the IRC request was made?
• were the blocking threats made by another member of ArbCom against editor 1 acting "within policy"?
• as a follow up, why has that same member of ArbCom not recused on the present WP:RfAr in relation to AUSC?
• why did it take so long for the oversight to be reversed... surely the first person on Oversight-L who saw the issue and was aware of the critique should have immediately reversed it as an obvious non-outing? Why was a discussion needed once someone with the requisite knowledge had arrived?

These are my initial reactions to a report that seems to me to raise more questions than it answers. I won't even ask about how it took so long to produce this report. AUSC members, I am deeply disappointed. EdChem (talk) 12:02, 17 December 2009 (UTC)

• Naturally these are my own views, of course:

1&2. Would you believe that I, who joined just a few months after "editor 2" and who have occasionally interacted with Giano, had also never heard of Randy from Boise before this mess? Wikipedia is a big place.

3. When an editor, who was less careful about the use of his real name when new to Wikipedia, has second thoughts and wants to pull back from that use and stick to his or her pseudonym, we ought to respect that. While the use of the real name, whether accidental or on purpose, is not suppressible for this reason, it can be redacted and the if the person who is using the real name is doing it as a form of harassment, that person can be blocked.

4. Oversighters are encouraged not to act in their own interest, but to find someone else to review the situation. This is especially important where the oversighter is involved in an ongoing dispute with the editor who made the edit at issue. When oversighters do act in their own interest, they are expected to immediately post the situation to the oversight mailing list for review. Seriously, if "editor 2"^[1] had done the suppression himself, there would have been deafening cries of "abuse of authority" from certain quarters. Asking for help is best practice. As far as "editor 3", who contacted DerHexer on IRC, being aware of the critique, my answer to 1 & 2 still applies. If you are going to hold every admin responsible for knowing the contents of every page, you aren't going to have very many admins.

5. No, the requests were made simultaneously, and "editor 3", who was in communication with the steward, was not included in emails between "editor 2" and the oversight mailing list discussing the situation. He had no way of knowing that the oversight list did not consider the edit a problem.

6 & 7. This has nothing to do with use of oversight or audit, which is probably why it was not addressed by the audit subcommittee.

8. Because it takes time to get consensus to do anything on committee operated by mailing list. Within the first hour of the request, only two members of the oversight list had commented, and only one of those two was strongly in favor of reversing the suppression (the other had made a tangential comment but did not directly address the issue of reversal). It took a couple hours to get a few other people to weigh in with concurring opinions. There was also a question of whether a steward action should be given greater deference than the action of a local oversighter. Ultimately it was not, of course, and this is one of the issues that informs the first recommendation.

My 2¢. Thatcher 14:43, 17 December 2009 (UTC)

^[1]This "editor 1", "editor 2" business is silly, in my opinion, but I respect the decision of the auditors even though I think it is silly.

(edit conflict) The first thing a clueful person does when confronted with a novel term is to run a Google search, such as google:Randy in Boise (to search Wikipedia only), or Randy in Boise (to search the Internet). Either of these searches surfaces the Wikipedia essay Wikipedia:Randy in Boise, written in August 2009 by User:Short Brigade Harvester Boris. Now fully clued up, the Oversighter should see that the request is entirely frivolous, and reject it. I know, and I'm sure every other Arbitrator knows, that the editor purportedly outed does not live anywhere near Boise. In my experience the Oversight mailing list gets a response very quickly. There was no need to go looking for somebody with zero clue about how the English Wikipedia works. The report, in my view, minimizes errors in judgment by several people. It would be better for those to own their mistakes, and for us to forgive them. Please, resignations are not desired. Jehochman ^Talk 14:45, 17 December 2009 (UTC)

When a vandal user creates the account User:Jehochman lives at 121 Nowhere Drive, and someone (either Jehochman or a vandal patroller) sends that to the oversight mailing list, the responding oversighter is not required to do either an exhaustive google search to see if it is true, or an exhaustive wikipedia search to see if Jehochman has ever revealed that information about himself. If there is a prima facie case that the edit is probably suppressible, often the best thing to do is to suppress it and then review it later. What aspersions will you cast on the editor who mailed the oversight list to report someone "outing" Newyorkbrad as Ira Brad Matesky? I think "editor 2" overreacted, but that does not mean that editor 3 or the steward did not act in good faith. Thatcher 15:00, 17 December 2009 (UTC)

I'm not privy to the instructions given to Oversighters. If our processes really are so naive, then we should improve them. It takes less than 5 seconds to type in the Oversight-proposed phrase, plus the word Wikipedia, and see what turns up. If this is not standard procedure, it should be. We all know that Randy doesn't live anywhere near Boise. How did this Oversight request pass the giggle test? Jehochman ^Talk 15:17, 17 December 2009 (UTC)

It is not standard procedure to do a google search before suppressing an edit. In fact, I'd probably trout an oversighter who did a google search before suppressing a potential privacy breach on AN/I. That page moves so quickly that another 20 or 30 edits might have to be suppressed during that interim. Suppression actions are reversible; in the case of possible privacy breaches, it is better to remove and then review rather than research painstakingly and then remove. Risker (talk) 15:24, 17 December 2009 (UTC)

How is it painstaking to check that a reference to a notable proper name is not a privacy violation? Are we just doing every request and checking later, or are we doing at least minimal sanity checking first. If, for instance, I call you Abe Lincoln because you've been honest ("Nice work, Abe Lincoln"), will a Steward from Bangladesh Oversight that because they've never heard of Abraham Lincoln? Shouldn't they at least try to fathom that it's a cultural reference? Jehochman ^Talk 15:29, 17 December 2009 (UTC)

Thatcher, thanks for your response. If I might pick up on a few points...

(1a) I have no reason to disbelieve you when you say you were unaware of the "Randy from Boise" reference, but I am very surprised. However, editor 2 (an Arbitrator) is certain to have seen Giano's name come up many times both on case pages and as a vocal critic of ArbCom. Editor 2 turned to a friend and colleague (and ArbCom clerk) who we are also asked to accept had never come across this critique regularly mentioned on wiki (ANI and AE, just for a start). Neither of these very long-term editors had apparently seen Giano's talk page. This strains credibility, to put it mildly. Bear in mind we are talking about not just regular admins; we are talking about an Arbitrator and two ArbCom clerks all not knowing about a critique which a prominent critic of ArbCom has on his talk page. It is certainly a well known meme amongst the science editors who have touced onto fringe topics and who have appeared in ArbCom cases.

(1b) You did not address my point that the description of the critique in the AUSC report implies it is obscure when this is not the case.

Re: (3) The use of Arbitrator / editor 2's name on ArbCom case pages (for which I provided links) both occurred in 2009. This is not an editor "who was less careful about the use of his real name when new to Wikipedia" having "second thoughts". Editors who are new to wikipedia are not serving arbitrators. I ask again: Editor 2 (aka the Arbitrator) must have known his or her first name was known on wiki and has been used, so why did she or he think this use of his or her name was any form of outing?

Re: (6) In a case of genuine outing it would be common for the out-er to receive a sanction such as a ban. An Arbitrator was in fact threatening editor 1 with blocks / bans in relation to pushing for this oversight issue to be addressed. AUSC has declared that all parties acted within policy, but you are now saying that this behaviour and the question of a block / ban, which would normally be part of outing incident, are outside the role of the AUSC. Sounds like there is something very odd in the AUSC mandate... or a reluctance to examine the behaviour of yet another Arbitrator. Amazing. By the way, I have stuck with "editor 1", "editor 2", etc formulation for fear of an accusation of outing and threats from this "protector" Arbitrator of editor 2.

Re: (8) What can I say but beaureaucracy gone mad... in the first hour, there was an oversighter who recognised that there was no outing and was strongly of the view that the oversighting be reversed. However, that person felt the need to wait for others to agree before taking action. It took several more hours for an (unfortunately, now former) Arbitrator to step in and act in the way that should have been obvious to all. Not only should the oversight have been rejected as ridiculous, but also the oversighters, having realised this, were concerned about deferring to a steward or scared by the involvement of an Arbitrator and so apparently felt the need for majority agreement to undo the incorrect oversight. AUSC asks the WP editing community to accept that this incident shows everything is fine with the system. Sorry, but I just find that isn't possible. This incident shows the system is badly broken, and it is tragic that the supervisory body either won't say so or can't see the faults. EdChem (talk) 16:33, 17 December 2009 (UTC)
Points to remember:

1. For what it is worth, I have never heard of the "randy from Boise" meme either, and I live in the US.
2. Suppression is reversible.
3. We should err on the side of protecting people; it is much easier to let cats out of the bag at a later point than to re-collect them once out.
4. Suppression is reversible.
5. Editors and maintenance personnel are not robots but people, and if it takes some time to discuss this amongst themselves, a little patience goes a long way.
6. Did I mention suppression is reversible?
7. Suppression is reversible and, if unnecessarily performed, can be corrected.
8. Since suppression is reversible and since we should err on the side of protecting people (since it is much easier to protect pre-release than post release) it makes sense to suppress quickly, then discuss if there may be any issues, and then correct the suppression--if unnecessary, after due diligence is followed.
9. Everyone, including maintenance personnel, can make mistakes, and acceptance of that is necessary in any small or large-scale social setting.
10. Oh yes, suppression is reversible.

-- Avi (talk) 16:35, 17 December 2009 (UTC)

The problem, Avi, is the Streisand effect. Clumsy use of Oversight attracts much more attention than merely doing nothing. Rather than impersonating Sargent Schultz, the Audit Committee ought to call out the weaknesses and try to fix them. EdChem's characterization of bureaucracy run amok is very accurate. (No, I have not outed anybody named Streisand or Schultz for those who've never heard of them.) Jehochman ^Talk 16:51, 17 December 2009 (UTC)

When done quickly, the effect is minimized, and even so, a lot of attention on an empty box still gives less information about what was in the box than a little attention on a full one, at least in my opinion and experience. -- Avi (talk) 16:55, 17 December 2009 (UTC)

Ed, I can't really speak to why someone else didn't know something that was obvious to you. My wife and I call it the "groundhog" effect, after the fact that I always seem to notice groundhogs, deer, rabbits, and other small furry creatures on the side of the road that she would say "awwww" over, except that she usually misses them. Personally, if you found the 10 people whose talk pages I use the most, I doubt I could tell you how any of them are decorated. As far as editor 2 is concerned, he stopped voluntarily using his real name after a series of harassment incidents (last year or early this year perhaps) but he signs his real name on private emails and emails to the functionaries (and presumably Arbcom) closed mailing lists. So if an arbitrator accidentally refers to that name in a discussion, it is not really proof that such references are welcomed, or that deliberate taunting (which it might appear to be to one who did not know the meme) would be permissible. The mandate of the audit subcommittee is to review the use of checkuser and oversight permissions and does not include reviewing comments by editors or arbitrators who are discussing a disputed oversight matter. In other words their mandate is the disputed matter itself, and not any conduct that arises from discussion of disputes where such conduct does not involve the use of the permissions. I don't see how it could be any other way. If I were to do something as foolish as block you because you object to the audit committee report, that block does not become an audit committee matter. And, so far as "bureaucracy gone mad" is concerned, Arbcom recently did something rather hastily, and then took what they thought was corrective action also in a hasty manner, and it did not work out well. Having suppressed the edits, it was sensible to be really really sure that reversing the suppression was the correct thing to do. In other words, measure twice and cut once. Thatcher 16:59, 17 December 2009 (UTC)

The report is a joke, I have seen this "Randy from Boise" thing at least 10 times. Seems that like usual the powerful ones get away with everything.  Dr. Loosmark  18:39, 17 December 2009 (UTC)

I'm sorry, but get away with what, exactly? Thatcher 18:48, 17 December 2009 (UTC)

I have to agree - I'm finding it basically impossible to see how anyone has "gotten away" with anything, or abused anyone, or harmed the encyclopedia or any living thing in this whole incident. On a practical level, this case has been useful since it's good to have clear guidance for oversighters, but I cannot for the life of me understand the political dimension that people seem so worked up about.

I, like many contributors here, value some level of privacy. If I were in the shoes of Editor 2 and thought, however mistakenly, that someone had "outed" me, then I would greatly appreciate a prompt response from oversight - which they provide with remarkable consistency. If, in hindsight and with the benefit of time, the suppression looks unnecessary or even a bit goofy, as in this case, then it can be undone.

This is actually exactly how I would want oversighters to behave when making decisions in real time (although a prompter resolution after the fact might have been desirable, but that drum has already been beaten). If there's an actual abuse here, as opposed to a real-time decision that can justifiably be second-guessed after the fact, then it hasn't been clearly explained to me. Right now it looks kind of like the usual Wikipedian issues with authority of any stripe - which I get - but there doesn't seem to be a lot of substance. MastCell ^Talk 19:04, 17 December 2009 (UTC)

As a general note, I have been vocal about maintaining the AUSC policy on not mentioning names as a default state. The reasons for this are various, but the general rule is that unless they effected editors request otherwise, there is a public request, or someone is being fingered as having screwed up, we stay silent as to (pseudonymous) identity. In this case, our conclusions devolve to "A series of inevitable and reasonable mistakes resulting from a drama laden situation, that ended without real harm to anyone" - and as a result there are exactly zero people to blame.

I don't think it would be shocking for me to suggest that there are elements of a finger pointing culture - as well as anti-authority and distrust tendencies among hooked in (pretty anyone reading this) Wikipedians. If we populated the above report with names - despite our conclusions being "no one did anything wrong" I have no doubt that the slightest defects in wording would be magnified a thousand times over to make some sort of point about the fitness, or illicit access, or whatever of a particular actor.

Making it clear that I am only speaking for myself, and there are other AUSC members who disagreed with my stance in this case, some of the responses I see here are exactly the kind of responses that compel me to adopt name silence as a default. The creation of any sort of structure like AUSC requires that at some point, you decide to trust us - or that you don't. That decision will effect what you see, or don't see in the report, more than the other way around.--Tznkai (talk) 19:19, 17 December 2009 (UTC)

• Let's be quite clear about this, there is no secret because it is openly discussed elsewhere. RLevse was the Arb who thought he was being outed - he has plastered his full name all over Wikipedia. he is not a newbie - he is (albeit surprisingly) an Arb. Having received no satisfaction from our own checkusers he persuaded MBisanz ( a 'Crat) to find a naive oversighter from another project on IRC and believe me RLevse persuaded very hard indeed! Does no one wonder why they have heard nothing from MBisanz? - beyond me who had an email of appology - which I accepted because MBisanz was duped by RLevse as ultimately was the oversighter. When I complained Coren threatened to block me. When I became even more concerned - eventually this pile of deceit shit - there is no other word for it was produced. However, this report is very telling and worrying - it is a report that is produced by their fellow Arbs - if this is accepted - what is there next encouraged deceit going to be. The Arbs who produced this heap of drivel for your delectation are people to watch in future. They deceive for their own ends and the ends of their friends. God help Wikipedia.  Giano  21:37, 17 December 2009 (UTC)
  • Wow. Let's just speedy delete WP:AGF, based on the above statement. I see no reason to pretend that Giano has any goals in maintaining it... --Jayron32 21:46, 17 December 2009 (UTC)

Mistakes happen. What is this silly drama? --Apoc2400 (talk) 21:57, 17 December 2009 (UTC)

I think this explanation overlooks the important effects of the Thetans, Area 51, and 9/11. Editing those in would help. AGF also requires at least one mention of Nazis in all Wikipedia conspiracies.

Alternatively we could all wake up and decide to sanity-check our facts one day. FT2 ^{(Talk | email)} 22:08, 17 December 2009 (UTC)

Black helicopters. Don't forget the black helicopters. Or, we could apply Hanlon's razor and chalk it up to confusion, mis-perception, and thinking Giano had bad motives when he didn't. A more critical report that said what we all know already--there was a cock up--and a statement to Giano that he was wronged, would probably put this situation right. The digging in heels and refusal to recognize wrongs isn't helpful. Jehochman ^Talk 22:31, 17 December 2009 (UTC)

How, exactly, was Giano wronged? I see how he was offended and irritated, but I see exactly no harms done to him, or to anyone else. This isn't a hand wringing kumbaya we have to look forward thing either. I mean seriously, there are no harms done to anyone. All we have here is a lot of noise and bluster. AUSC is not a tool to be used to continue interpersonal conflict. We are not a body interested in reputational damage. We are concerned only in the control of the misuse of the checkuser and oversight permissions.

If you prefer this in the venacular: there was a fuck up, because people are not omnicient and infalliable, the mistake has been fixed. It might happen again, but we've made recommendations so it doesn't. No one was hurt. End report.--Tznkai (talk) 22:50, 17 December 2009 (UTC)

Indeed, a comment might be appropriate from the affected Editors regarding exactly what the percieved lasting harm to Wikipedia was. The edits in question were restored after a brief interim. One would expect that such restoration (with admittance of minor errors) would suffice to mollify the affected Editors. —Beneficence 22:14, 17 December 2009 (UTC)

Well, I've posted to Giano's talk page often enough, and "Randy in Boise" had never registered with me whatsoever. We've had the findings of the committee now, so unless anyone has concrete evidence which undermines them, let that be an end of it. Giano, either put up or shut up. Either you've got something, or you haven't.--Scott Mac (Doc) 22:18, 17 December 2009 (UTC)

Ultimatums aren't especially helpful. I know Giano, and he's not going to out anybody. If anybody was concerned about Giano outing, they were misperceiving the situation and needed to double check their facts. Rlevse is a good and honest person. I very much doubt he engaged in purposeful deception; he was probably just confused. I think both sides need to recognize that this was nothing more than a run-of-the-mill cock up, something that will happen from time to time. We should understand what went wrong and see if we can strengthen our processes so that it doesn't happen again. Jehochman ^Talk 22:36, 17 December 2009 (UTC)

Oh, I agree. But I'm reading (or misreading) Giano as implying that there's more to this. If there is, he needs to be specific. If he can't, well then where are me.--Scott Mac (Doc) 22:39, 17 December 2009 (UTC)

• If you cannot see the wrong in Arbs going above the heads of our own oversighters and sending their personal 'Crats to IRC to find naive oversighters form other projects then there is little hope. If you cannot see the dangers of other Arbs threatening to block those concerned by this and if you cannot see the harm of other Arbs then producing deceitful reports to conceal such actions - then we have reached a very sad state - we may just as well allow the Arbs to become dictators. They are now exercising self given authority made legal by self-endorsement.  Giano  22:43, 17 December 2009 (UTC)

• Oh, I can see the wrong and danger in what you describe. However, I just can't see the evidence that that particular interpretation should be preferred to the one offered here. I always prefer cock-up as a rationale rather than conspiracy. Willing, as always, to be convinced otherwise by compelling evidence.--Scott Mac (Doc) 22:48, 17 December 2009 (UTC)

Oh there was no conspiracy - a conspiracy requires secrecy - all that was secret was Rlevse pursuading MBizans to pursuade the naive oversighter. RLevse did not want to sully his own hands or to put it more frankly, have the dirt come back to his own doorstep.  Giano  22:52, 17 December 2009 (UTC)

It seems more likely that Rlevse was busy, or distracted by something else, and not thinking clearly. I very seriously doubt he could be so diabolical. That hypothesis does not accord with what I know about his character. Jehochman ^Talk 22:57, 17 December 2009 (UTC)

• (edit conflict)Indeed, what they did was wrong. But it was naive, not malicious, in my humble opinion. They need to explain that their actions were wrong, and show that they understand things now. You say you forgave MBisanz. Perhaps you could forgive the others if they made corrective statements? Jehochman ^Talk 22:50, 17 December 2009 (UTC)

• Giano, what you're saying is unambiguously counter factual.--Tznkai (talk) 22:49, 17 December 2009 (UTC)

No. it is not counter factual - it happened!  Giano  22:53, 17 December 2009 (UTC)

No. It didn't. No one went over anyone's head. No one has personal servants that they directed. No one threatens to block you for any reason on than your own hostile behavior. No one was motivated by bad faith. No one created a decietful report.--Tznkai (talk) 22:56, 17 December 2009 (UTC)

Hostile behaviour? Moi? What on earth could make me hostile? An Arb plastering his full name all over the site and then suddenly becoming a little precious and delicate when the word "Randy" is mentioned in his presence - then hois fellow arns saying it was "reasonable" of him to think that he is "Randy from Boise" - No, surely not.  Giano  22:59, 17 December 2009 (UTC)

This would be a silly mistake to laugh at, except that no-fun Coren threatened you with a block. They should own up to their silliness, and you should let them off the hook. Jehochman ^Talk 23:03, 17 December 2009 (UTC)

Giano won't be able to reply. Coren has indeffed him. Jehochman ^Talk 23:36, 17 December 2009 (UTC)

Oh good, that will solve everything. Mackensen (talk) 23:49, 17 December 2009 (UTC)

I've been here a long time and I've never heard of "Randy from Boise." I still haven't, actually. If I encountered without context of any kind I'd assume we're talking about an obnoxious Boise State fan, although that may be redundant. Mackensen (talk) 23:05, 17 December 2009 (UTC)

On a more different note

Thank you for releasing the report. It is difficult for any comittee to write a document. I apologize for being such a bother through the process, and merely hope that no one holds it against anyone. Hipocrite (talk) 14:03, 17 December 2009 (UTC)

Eh?

As someone who has just read the report knowing nothing about the underlying matter, I find myself none the wiser. As to the bit of the report that (I think) I do understand, I must say that I am surprised by the number of "suppressions" stewards perform on enwiki (see here): 266 by DerHexer between June and August 2009 and 886 by Pathoschild in October 2009 alone. Were all these "suppressions" in response to emergency situations? Those are a lot of actions for those subscribed to Oversight-L to review. WJBscribe (talk) 23:01, 17 December 2009 (UTC)

Oh, you will just find that one of the Arbs thinks he's "Joe the Plumber" that's all that is. Risker probably thinks she's Jeanne d'Arc - and we have to all mentions of her removed. Brad will probably imagine himself to be Napoleon next week, and the other Arbs will all claim to have never heard of those people either. That's how worrying this is. Giano  23:05, 17 December 2009 (UTC)

Almost all of those high numbers are caused by crosswiki username abuse, where someone creates an account with libel or personal information in the name on another project where their IP is not blocked, and it gets automatically created here because of SUL. In those cases stewards hide and lock the username globally, and have also been encouraged to feel free to suppress the username on en.wp at the same time. Generally, these could be reviewed pretty easily, all at once when they occur. However, this practice is also something that we as oversighters should consider, since I have noticed in the past a disparity between some of the steward standards for suppression and the en.wp ones. Dominic·t 23:11, 17 December 2009 (UTC)

You know, if I were Napoleon I could go live on a small island for awhile, and no one would take the time and trouble to call me various names, and I would not have to worry about quite a number of things. Newyorkbrad (talk) 23:15, 17 December 2009 (UTC)

Napoleon was poisoned on his small island!  Giano  23:16, 17 December 2009 (UTC)

You don't see the poison on this page? I do. I suspect we disagree on its nature, though. Nathan ^T 23:23, 17 December 2009 (UTC)

I need a fast-forward button where this whole thing just quiets down to the non-event that it is, and becomes another bullet point on various people's "User:MountainsFromMolehills/List of Reasons Why Wikipedia is Corrupt" user pages. Does anyone have such a button? Wknight94 ^talk 23:45, 17 December 2009 (UTC)

IMHO, people shouldn't be doing things on both Wikipedia & IRC. GoodDay (talk) 00:11, 18 December 2009 (UTC)

The facts

This statement by Giano is full of untruths and blatant gross personal attacks:

Let's be quite clear about this, there is no secret because it is openly discussed elsewhere. RLevse was the Arb who thought he was being outed - he has plastered his full name all over Wikipedia. he is not a newbie - he is (albeit surprisingly) an Arb. Having received no satisfaction from our own checkusers he persuaded MBisanz ( a 'Crat) to find a naive oversighter from another project on IRC and believe me RLevse persuaded very hard indeed! Does no one wonder why they have heard nothing from MBisanz? - beyond me who had an email of appology - which I accepted because MBisanz was duped by RLevse as ultimately was the oversighter. When I complained Coren threatened to block me. When I became even more concerned - eventually this pile of deceit shit - there is no other word for it was produced. However, this report is very telling and worrying - it is a report that is produced by their fellow Arbs - if this is accepted - what is there next encouraged deceit going to be. The Arbs who produced this heap of drivel for your delectation are people to watch in future. They deceive for their own ends and the ends of their friends. God help Wikipedia. Giano 21:37, 17 December 2009 (UTC)

1. 'surprisingly an arb', a personal attack
2. ' Having received no satisfaction from our own checkusers he persuaded MBisanz ( a 'Crat) to find a naive oversighter from another project on IRC and believe me RLevse persuaded very hard indeed!', several falsehoods here: I did not contact a checkuser, I did not ask MBisanz to contact anyone nor do anything on my behalf,
3. 'duped by Rlevse', I did not dupe anyone, all I did was vent to him
4. 'pile of deceit', hardly, though more personal attacks by Giano
5. 'there next encouraged deceit', more personal attacks and insults
6. I could go on but I won't, I've made my points
7. Just for the record, I'd never heard of "Randy from Boise" before all this started

...in short, more insults (he previously called me a coward, etc when all this started), bad faith, and personal attacks; which would almost certainly not be tolerated from any other user, but ahem. I demand an apology from Giano to me personally, Arbcom, and AUSC. — Rlevse • Talk • 00:19, 18 December 2009 (UTC)

For heavens' sakes. "surprisingly an arb" is a personal attack? "pile of deceit shit" written to describe a report (which I might describe in the same terms if I weren't such a gentleperson) is a personal attack? There are some substantive points that you could have responded to, if you were interested in a dialogue with people who don't like how this went down, and a simple "I f**ed up" would have gone a long ways, but this response is not helpful. Whether or not you'd ever heard of "Randy from Boise," I'm still at a loss, if you don't live in Boise, how you could reasonably consider the phrase an attempt to out you. I'm not a fan of Giano's (not an enemy either; I just don't encounter him that much) but when he's right, he's right, and why you people are falling over each other proving his point is just beyond me. I guess I can overlook your getting to be an arb without encountering the phrase "Randy from Boise," but how in the world did you get to be an arb without learning that demanding an apology is almost always counterproductive? Woonpton (talk) 00:38, 18 December 2009 (UTC)

Because one is due and the environment that perpuates the need therefor does more harm than good. — Rlevse • Talk • 00:44, 18 December 2009 (UTC)

Rather than demanding an apology, which deminishes its value, could you ask for a retraction or correction? Could you state for the record that you now understand Giano was not trying to out you, and therefore he should not have been threatened with a block accused of doing that? Jehochman ^Talk 01:51, 18 December 2009 (UTC)

Was he, in fact, threatened with a block for trying to out? There a lot of accusations here, can we check the facts?--Scott Mac (Doc) 02:03, 18 December 2009 (UTC)

kinda Privatemusings (talk) 02:08, 18 December 2009 (UTC)

Not even close, really. The only directive in that message was that "Now for the less pleasant part of this note: you are to stay scrupulously civil and respectful to everyone if you insist on discussing the matter further.", the reference to outing was earlier and, quite the opposite, stated that "nobody seriously things you outed anyone", which is hardly a threat. — Coren ^(talk) 02:19, 18 December 2009 (UTC)

(ec)No, Not "Kinda". This is how urban legends start. Read the diff. Jehochman's statement is in fact erroneous. It is quite clear from that diff that Coren did threaten to block Giano, but NOT for "outing" - indeed the diff implies that Coren realised Giano had not been outing, only that the impression that he had been was an honest mistake. Indeed, in that diff, Coren refers to a "bad suppression". No, it is clear from the evidence, that Giano was threatened with a block does to his civility and intemperate remarks. "you are to stay scrupulously civil and respectful to everyone if you insist on discussing the matter further. This means no backhand sniping, no condescending monikers, no barely veiled caustic sarcasm, no accusations of lies or other sins of choice. If you do not feel you can discuss the subject with dignity, then avoid the subject entirely. I will enforce this with blocks as required." This is important, because an urban legend is starting here. Giano was threatened that he'd be blocked for "incivility and intemperance" - which are things he certainly is often guilty of. Five days earlier, Coren had already told Giano that the outing allegation was a mistake [7]. So the notion that it needs to be admitted now that it was a mistake, in order to pacify Giano, is silly, it has never, in fact, been denied/ --Scott Mac (Doc) 02:27, 18 December 2009 (UTC)

re: the 'kinda' Giano's statement 'When I complained Coren threatened to block me' - was presumably based on coren's statement '..I will enforce this with blocks as required...' - I'm not saying coren's message was inappropriate, and I agree that to say Giano was threatened with a block for outing is inaccurate, but giano's interpretation (the threat of a block) was reasonable in my view. Privatemusings (talk) 02:30, 18 December 2009 (UTC)

Yes, but it is important to note that he was not warned of an impending block for complaining, but for the manner in which he did so. I did, in fact, encourage him to pursue further resolution if he was unsatisfied. — Coren ^(talk) 02:44, 18 December 2009 (UTC)

Empiricly, "calm down or I will block you" never works. Giano was upset because he was falsely accused of outing somebody. He got upset as a result of a false accusation. Had that not happened, neither would all that followed. Jehochman ^Talk 03:31, 18 December 2009 (UTC)

< heh... for a section headed 'the facts' you're in danger of being slapdash, I reckon (the behaviour of a gigantic boob) - giano's simple statement 'When I complained Coren threatened to block me' is accurate in my view - interpretations, perceived consequences, and important notes can all follow. Jhoch's post runs the risk of confuddling matters hence the 'kinda' answer to scott's question; 'Was he, in fact, threatened with a block (yes) for trying to out (no)?' == kinda. Privatemusings (talk) 03:00, 18 December 2009 (UTC)

I corrected my post. It was a question, by the way. Rather than arguing, people could have answered directly, "No he can't say that because it's not exactly right. This is what really happened, and that's what he could say." Jehochman ^Talk 03:31, 18 December 2009 (UTC)

One of the best ways to obtain an apology is to make one yourself. Rlevse, do you think your conduct has been faultless in this matter? Where you fair in your "venting" about the matter on IRC (where Giano cannot respond and has a long history of being verbally abused behind his back)? It appears to me that Giano feels that, because the oversightings took place and it has taken so long for a report, his reputation has been tarnished in the meantime as someone who "outs" other users (something he clearly feels strongly is wrong). If he has expressed himself too forthrightly in the aftermath of this whole business, can you at least see why? WJBscribe (talk) 10:46, 18 December 2009 (UTC)

Giano has insulted me (not to mention countless others) more than once, including before the "Randy in Boise" episode, and never apologized to me, so he should apologize first. — Rlevse • Talk • 11:05, 18 December 2009 (UTC)

This gives the appearance that you knew he wasn't trying to out you (which I think some of the less naive amongst us already suspect), and were just pissed off because of being poked at with the "Randy in Boise" comment. Also, the whole "But he started it!" silliness does nothing to elevate your perceived maturity level, either. You are looking worse and worse the more you post regarding this situation, both with regards to the underlying "dispute" and your general disposition. UA 15:49, 18 December 2009 (UTC)

• Now we have it at last, it was a perceived insult that was annoying you and you wanted oversighting - nothing to do with outing at all.  Giano  11:31, 18 December 2009 (UTC)

Bullshit. — Dog The Teddy Bear • ^Bully! • 17:28, 19 December 2009 (UTC)

For those who have never clicked on DTTB's userpage, this is Rlevse's alternate account. Just an FYI. UA 05:47, 20 December 2009 (UTC)

• This childish in-fighting is getting so, so boring. AGK 11:38, 18 December 2009 (UTC)
  • To say the least. I stopped reading quite a while ago. --Deskana (talk) 12:42, 18 December 2009 (UTC)

View from the outside

I don't know who half these people are (but I'm sure I will), but I have handled complaints for a living for years. There are a couple of things here that I have noted.

Clearly there are histories of personal interactions here that I am unaware of. It is always important when investigating a complaint, that this kind of history is taken into account in order to remove it from the equation - ie to do their job properly, the investigator has to understand that X has done Y in the past, and Z will always react thus because of it, or whatever it is that has happened. The challenge for the investigator is to then make a finding of fact without a priori assuming that X is doing Y again.

An apology takes the wind out of the complainant's sails. It would be helpful if AUSC had issued a general apology that although the process was in accordance with policy from the point someone made the complaint, everything had happened too slowly, given that the steward's action featured a tacit charge of outing, and Giano had not been kept properly informed of what was going on. This could have been done by email, and should have been done by the Steward in his role as the person who carried out the action. Nobody knew what was going on until DerHexer got dragged over to ANI to explain. People have to sleep, so the correct course where something is oversighted in this kind of circumstance is for the person taking the action to also issue a notice. Since Giano wasn't also blocked (which of itself acts as a notice), a note to say "there has been a complaint raised that this edit may have outed another Wikipedian and the edit has been suppressed."Elen of the Roads (talk) 12:50, 18 December 2009 (UTC)

I would also recommend that Rlevse stop digging. Intentionally or accidentally, he's making it appear that what happened is not what the report says happened. So my third note would be to make sure everyone is going to sing from the same hymnsheet. This is most easily done by saying clearly anything that your first instinct is to fudge, because if you fudge it, you will end up with not only the complainant and those outside, but the parties involved coming up with different versions, as is shown here with Rlevse denying that he involved MBisanz, when Giano has reported that MBisanz sent him an apology. It would have been better to have made reference to that in the report.Elen of the Roads (talk) 13:00, 18 December 2009 (UTC)

Let's not let misinterpretations stick to this story unnecessarily. Rlevse wrote that he did not ask MBisanz to intervene, he didn't deny speaking to him (thus "involving" him). There's no contradiction between that and any apology MBisanz might have sent for contacting DerHexer. So I'm not sure I see the off-message versions that you allude to. Nathan ^T 15:01, 18 December 2009 (UTC)

"Rlevse wrote that he did not ask MBisanz to intervene, he didn't deny speaking to him (thus "involving" him)" Yeah but the point what exactly did he tell him that MBisanz felt the need to intervene.  Dr. Loosmark  15:15, 18 December 2009 (UTC)

You two have just made my point. It would have been better to have said more about this part in the response, to avoid precisely the kind of speculation that its absence is generating. Elen of the Roads (talk) 17:21, 18 December 2009 (UTC)

I thought your point was that the report fudged some things, which allowed competing and contradictory versions of events to unnecessarily muddy the water. Your basis for this conclusion was that Rlevse had given one of these contradictory versions of events, but it appears you drew this conclusion from an inaccurate reading of what he wrote. I'm not sure how pointing that out confirms your initial point. A lot of people seem to be issuing broad conclusions along with advice, but to be useful any attempts at post mortem analysis need to be based on facts.

On your other recommendations, since I'm here:

1. No one is on call for oversight, AUSC, arbitration or anything else. Things happen at the speed they happen because of the availability of various volunteers to do the needed work. A number of people have addressed the inevitable nature of the delay, and no apology seems to be necessary either for the pace of events at the time or for any perceived lateness in the arrival of the report. The suppression itself necessarily happens at speed; overturning the suppression takes longer because it requires the agreement of a body and because there is considerably less real risk in a delay.
2. Notice of oversight is typically not announced on AN/I. There's a well founded reason for this that probably does not require an explanation.

One area where I do agree with your comments is on communication; folks lodging a complaint would like to know, and ought to be told, that it's being handled and where in the process it might be. These things are best done by direct communication, or by both direct and public communication when there is public concern. In the future the subcommittee should consider updates on the progress of reviews at regular intervals, even if little real information is communicated other than continued attention. Nathan ^T 17:47, 18 December 2009 (UTC)

I'm obviously not being clear enough here. At no point anywhere did I mention ANI. What I said was that the steward never told Giano what he had done or why - given that DerHexer was probably the only one awake at the time, he was the one who should have done so. Normally, outing is dealt with by block, and the blocking notice or block log should show what the reason for the block was. DerHexer I presume does not have the tools to have blocked Giano, and didn't think it was his job to tell him that he was oversighting him. Rlevse spoke to MBisanz, who is the person who asked for a Steward to oversight the alleged outing, but MBisanz did not also block Giano or tell Giano there was a problem. Ergo no communication at the time, and a whole bundle of suspicion both then and now.

I don't actually give a monkeys about this business, no-one died, no-one was left out of pocket, and the main page wasn't deleted. All sides are clearly going to continue their simmering resentment of each other whatever happens. However, if one of my complaints investigators had handed in a report like AUSC did, I'd have sent them away to rewrite it. After years in the business, I can confirm that the consequences are no worse, and frequently better, if you let the facts out, than they are if you try to fudge it up.Elen of the Roads (talk) 23:56, 18 December 2009 (UTC)

Indeed, I think it's fascinating that even after this lengthly and high profile "audit" of the oversighting, at least one edit having nothing to do with the topic on which Giano was posting, but which came shortly after his in time, remains oversighted. I know because it's one of my edits; I further know that it was nothing that was thought to be controversial, because I have received no warning or questions regarding what I wrote. While the loss of my comment is probably not a huge deal in the grand scheme of things, I would have thought that a thorough audit would have resulted in the reversal of any improper or overbroad deletions. Steveozone (talk) 00:43, 19 December 2009 (UTC)

I am the oversighter who unsuppressed the Randy from Boise business, and I am pretty sure I unsuppressed everything.

here is your suppressed edit. Ah yes, I suppressed your diff! ;-) That was due to a different problem; I left a comment about it at Wikipedia_talk:COIN#Yet_more_COI_vs_OUTING.

ANI does regularly have problems that need to be suppressed, which is why there is a tendency to "suppress first, ask questions later" to problems on ANI and other high trafficked noticeboards.

As I understand it, mw:Extension:LiquidThreads will solve this problem, as each comment is given its own history. No doubt this will present new challenges for the oversight staff, but at least it will mean that unrelated threads wont be suppressed due to being on the same "page" as something else that needs to be suppressed. user:Werdna recently announced that Liquid Threads will be available soon. I don't know whether the English Wikipedia community has discussed implementing it yet. John Vandenberg ^(chat) 01:42, 19 December 2009 (UTC)

LiquidThreads discussion at Wikipedia:VPT#LiquidThreads_almost_ready_to_deploy. John Vandenberg ^(chat) 02:51, 19 December 2009 (UTC)