This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

View current draft

View current draft

Resolved

 –   Roger Davies ^talk 05:31, 23 March 2014 (UTC)

• Two things - quick self-explanation first. I have been an editor for a few years now and I still find much of the deeper-level administrative process opaque. I appreciate the opportunity to comment, and the suggestions below are requests to provide a more clear and succinct explanation of what is at stake and what is going on for those who do not live in your administrative world.

1. In my comment on an earlier version, I had suggested that "sanctions" be more clearly defined (this is what definitions are for!). The definition of "sanctions" remains too vague, in my view. I am glad that there is a new section added ( Wikipedia:Arbitration_Committee/Discretionary_sanctions/2013_review#Imposing_sanctions that does some work toward defining it, but even this remains somewhat unclear. Jytdog ( talk) 16:23, 16 March 2014 (UTC)
2. Also the whole notion of "discretionary" remains unclear to me, as someone who is not familiar with your work. I don't know what other kinds of sanctions there are that are not "discretionary" nor from what kind of procedures or users they arise. What is the context for this separate "discretionary sanctions" policy - what are you distinguishing it from? If you don't want to handle this the definitions section, a short paragraph explaining that would be immensely helpful. Thanks!! Jytdog ( talk) 16:23, 16 March 2014 (UTC)

Re #1 What remains unclear? The kind of sanctions that can be imposed? If so, it's not a limitless list. For instance, admins can neither ban (as against block) nor desysop though they can use all the usual ones: topic bans, interaction bans, blocks, revert restrictions etc.

Re #2 The "discretionary" bit means that admins may use their judgment ("discretion") to apply whatever sanction they believe will fix the problem.   Roger Davies ^talk 16:33, 16 March 2014 (UTC)

Thanks for replying (and for your hard work on this!). That is clearly an obvious answer to you, it is right there in the world you live in... not so to me... and it would be great to include this information in the actual document so that everyday users understand what is going on if they are notified! Thanks again. (btw in my actual job, I deal with policy and law on a daily basis and have to explain it to people - I understand how hard it is to draft robust, clear policy and to provide explanations that make it usable to those to whom it applies - it is hard but important work) 16:43, 16 March 2014 (UTC)

• I don't think the definition of an 'enforcing administrator' should include the word uninvolved, as that seems to have the causality rather backward. As written, it appears that the enforcing administrator shall be considered uninvolved by definition. In the (hopefully hypothetical) case that an involved administrator applies enforcement, how would one refer to that person when the term 'enforcing administrator' has been defined to apply only to uninvolved administrators? If the intent of including that word in the definition was to forbid involved administrators from taking enforcement action, that should instead be done elsewhere (and it currently is.) -- Noren ( talk) 22:19, 16 March 2014 (UTC)

I agree. The requirement for the enforcing administrator to be uninvolved is covered in later sections and shouldn't be in the definitions. At the moment it makes vacuous the later statement "Any enforcing administrator who, in the opinion of the committee: 1. imposes a sanction when involved" since the present definition makes that combination impossible. Zero ^talk 11:34, 17 March 2014 (UTC)

Yes, that duplication is an oversight. I'll remove involvedness from Definitions.,   Roger Davies ^talk 05:31, 23 March 2014 (UTC)

I concur. It's like saying "All Thursdays are cloudy." The notion that an enforcing admin is uninvolved cannot be definitionally embedded, since it's an independent factual matter. And there's nothing hypothetical about this; it's already happened. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 18:12, 20 March 2014 (UTC)

This is my first contribution to this kind of discussion.

I was 'topic banned' following controversy in thermal physics (in general). The initial warning was a complaint about my contributions on talk pages here.

My question is, does the definition of 'an editor' WRT topic bans include 'talk page' contributions?-- Damorbel ( talk) 20:52, 29 April 2014 (UTC)

Resolved

 –   Roger Davies ^talk 06:20, 31 March 2014 (UTC)

View current draft

• Two thoughts

(A) Suggest limiting these requests to registered editors (no IP motions)

(B) Wordsmithing... When it becomes apparent Any editor who believes that discretionary sanctions for an area of conflict are no longer necessary, the committee may be asked may ask the committee to rescind the authorisation of them.

NewsAndEventsGuy ( talk) 14:31, 22 March 2014 (UTC)

Hmm, but that sort of implies that the committee cannot rescind of its own volition ...   Roger Davies ^talk 05:34, 23 March 2014 (UTC)

Incorporating various points, and de-duplicating a bit,   Roger Davies ^talk 07:41, 25 March 2014 (UTC)

View current draft

Resolved

 –   Roger Davies ^talk 08:10, 25 March 2014 (UTC)

• Where are the behavioural expectations defined? • • • Peter (Southwood) ^(talk): 16:08, 17 March 2014 (UTC)
  • In the section headed "Behavioural expectations" ;)   Roger Davies ^talk 05:37, 23 March 2014 (UTC)
    • They are listed there, but not defined. • • • Peter (Southwood) ^(talk): 09:31, 23 March 2014 (UTC)
      • I've linked but not defined.   Roger Davies ^talk 10:22, 23 March 2014 (UTC)
• Add, "or discussing discretionary sanctions for a subject area on user talk pages"? That would address many eds' comments in this discussion about bullying/threatening type of comments when user A notify/warn/alerts user B, but would still allow people to talk about the subject beyond the naked template. NewsAndEventsGuy ( talk) 08:28, 25 March 2014 (UTC)

View current draft

• For this, the more narrow the definition, the better. One way this could be handled is with a list of the involved people appended to any arbcom decision, with notification at their talk, clearly saying that they are presumed aware of the AE/ DS situation. Future individuals could be added to that list if they have been given an "alert" as outlined in the next section. Only after that should anyone be deemed vulnerable to any DS. I recently stumbled into one of these situations where I was "warned" about something that I had no idea even existed. (I also believe my actions were not within the scope of the AE in the first place, but that is a different issue). Previously unaware folks or people whose knowledge is uncertain should not be subject to DS out of the blue. Montanabw ^(talk) 02:16, 16 March 2014 (UTC)

• With reference to the parenthesis ending §1.ii, how does having successfully appealed a sanction make one unaware of it? Indeed, while I can imagine someone ‘brushing off’ a topic ban by editing elsewhere, or a fixed-term block by just waiting it out, without paying much attention to the rationale (especially if he or she believes it to have been motivated by personal animosity, or to be an instance of systemic “admin abuse”—or for whatever other reason feels arbitrarily victimized), but surely an appeal would have to address the relevant remedies, cogently and directly, in order to be successful! Or is there a presumption that winning an appeal will clothe the victor in a rosy glow of imagined immunity, or blot out all memory of the unpleasant episode?— Odysseus 147 9 02:41, 16 March 2014 (UTC)

• It doesn't make one aware of it but it reduces the subsequent period to twelve months. That is intended to help people move on.   Roger Davies ^talk 13:34, 16 March 2014 (UTC)
• Perhaps the parenthetical should be better worded because such appeals will fall under §2.i until 12 months have passed. ~ Ningauble ( talk) 15:40, 16 March 2014 (UTC)

• That makes sense to me, and thanks to Ningauble ( talk · contribs) for taking time to bring this up since I was about to do so. NewsAndEventsGuy ( talk) 14:40, 22 March 2014 (UTC)

• As previously mentioned, I think that having alerts expire makes little sense. If an editor has been alerted three times already, a fourth alert in their fourth year will not really tell them anything new, and will on the contrary be perceived as a punitive or confrontative act - precisely what this review seeks to prevent. There have not been cases, as far as I recall, where editors have genuinely argued that they forgot that DS applied, so this looks like a solution in search of a problem. Also, we expect editors to follow policies even if they are not aware of any special sanctions regime. Per Odysseus1479 above, it also makes no sense that winning an appeal against a sanction makes one unaware of discretionary sanctions. This, too, reinforces rather than removes the mistaken notion of warnings or alerts being punitive in nature.  Sandstein  10:59, 16 March 2014 (UTC)

• And I think it's utterly unrealistic to expect people to remember a simple notification for much more than a year. I'm not going to lose sleep if the odd person games it by making four dodgy edits in four years.  Roger Davies ^talk 13:34, 16 March 2014 (UTC)

• Someone who rarely edits in a DS-authorized topic area and gets alerted probably shouldn't be expected to remember it after a year; someone who edits every day in such an area could probably be expected to. The question, of course, is whether we can write something that covers the latter without making unrealistic expectations for the former. T. Canens ( talk) 14:34, 16 March 2014 (UTC)

• FWIW, in the real world, things like sentence enhancements DO expire for some "crimes" - after enough time, you get to go back to your first "offense" in punishment (Driving with no insurance on your car being an example I'm thinking of in my area) and as a rule, state of mind is irrelevant (you can be totally unrepentant and recalcitrant or just a doofus, the law doesn't care) Roger is on the right track; Timotheus' valid point can be addressed by the timeframe; six months probably too short, more than a year, too long. Montanabw ^(talk) 19:07, 19 March 2014 (UTC)

• The Awareness section is currently ambiguous. It implies that they must fall under both 1 and 2, whereas I believe the intent is 1 or 2. Guy ( Help!)12:18, 16 March 2014 (UTC)

• Yes, that is the intent and it's not perfect, I know. Tony1 has suggested some changes below,   Roger Davies ^talk 13:34, 16 March 2014 (UTC)

• If "and" is the intention, then the word "and" should be added to leave no doubt about it. Zero ^talk 15:56, 16 March 2014 (UTC)

• In item 2.ii the two section links to "#Issuing alerts" are broken. I presume the intended target is the section headed "Alerts". ~ Ningauble ( talk) 15:51, 16 March 2014 (UTC)

• Thanks!  Fixed   Roger Davies ^talk 16:08, 16 March 2014 (UTC)

• For the purposes of the first clause of item 2.ii, "alerted other editors", I think posting an explicit reference to sanctions under the arbitration case could be taken by an enforcing administrator as conclusive evidence of awareness, even if the post did not involve the formal notice described under "Alerts". (Cf. my remarks about informal discussion under #Alerts (comments) below.) ~ Ningauble ( talk) 16:51, 16 March 2014 (UTC)
• Under the clock (whether it runs for 12 months or a different period), could we add "A sanction for the area of conflict expired"? and have the server send out a Template:WelcomeBack/Re-Alert when the clock runs out in the associated log (assuming there is one)? NewsAndEventsGuy ( talk) 17:48, 22 March 2014 (UTC)

This is Tony1's suggestion. Does it do the job better than the current text?   Roger Davies ^talk 14:03, 16 March 2014 (UTC)

So if someone's been sanctioned 10 times and successfully appealed one of those ten sanctions, he doesn't fall within #2? T. Canens ( talk) 14:15, 16 March 2014 (UTC)

Doubly blessed. He falls within #4 and #2.   Roger Davies ^talk 16:05, 16 March 2014 (UTC)

"all such sanctions have not been successfully appealed" reads to me as "each of these sanctions has not been successfully appealed". Do you mean "at least one such sanction has not been successfully appealed"? T. Canens ( talk) 17:40, 16 March 2014 (UTC)

Good wording, Tim, which I've used. I've also added a bit explicitly explaining that the twelve-months run from the date of successful appeals (ie they don't become instantly unaware because an appeal has succeeded).   Roger Davies ^talk 05:41, 23 March 2014 (UTC)

Changed my mind on this having seen Alt Text #2 below, which moves this on considerably.   Roger Davies ^talk 05:47, 23 March 2014 (UTC)

I agree with T.C. that the wording is unclear. However I don't know why it is needed at all. As someone else mentioned, someone doesn't suddenly lose awareness on winning an appeal. It seems a red herring to me. Zero ^talk 08:26, 17 March 2014 (UTC)

• The original purpose of this review, as I understood it at the time, was in part due to excessive legalistic thinking about AE warnings. In that a user had to be warned before being sanctioned, and thus some wanted to be able to appeal their warnings. Which is/was ludicrous.

  Somewhere along the way this has been forgotten or only partially remembered. Tying a user's awareness of DS in a topic area to a 12 month window is equally legalistic and quite arbitrary. The above point by Odysseus1479 re: appealed sanctions should point this out.

  Once you are aware of something unless you have dementia, Alzheimer's or some other medical cognitive or memory based issue you remain aware. It would be better to build in AGF for lapses in memory/mistakes than add an absurd legalistic notion of awareness that ties everyone's hands. This is about discretionary sanctions not automated sanctions. Concretely my suggestion is remove the 12 months clause and leave it that if one takes part in a DS discussion one is aware of the sanctions. And in response to teh litany of "I just commented at AE on topic X" - AE discussions are not casual - you shouldn't take part if you don't want to deal with the consequences-- Cailil ^talk 15:14, 17 March 2014 (UTC)

"Somewhere along the way this has been forgotten or only partially remembered." Hmm, you mean "awareness" of it has faded? I think Roger Davies has made it clear that the 12-month window is, in his words, "to help people move on". This seems reasonable. It's not actually reasonable to assume that people will remember exactly what is and is not under discretionary sanctions, which is a moving target anyway, for years and years, if they are not devotees to getting into AE squabbles. To the average wikipedian all of this stuff is a bunch of legalistic FUD. We don't actually know who may have memory problems (and WP is not off-limits to those who do), but the supposition that "Once you are aware of something unless you have dementia, Alzheimer's or some other medical cognitive or memory based issue you remain aware" is actually false anyway; people clear trivia, especially bureaucratic trivia, out of their minds continuously and automatically, and often do so much faster than a 12-month time span. If anyone still has cognitive dissonance with the idea, then stop thinking of it as a presumption that memory has faded, and accept it as a simple "statute of limitations"; having one here is sensible, and doesn't require belief that anyone's definitively forgotten anything. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 18:27, 20 March 2014 (UTC)

• Re: point 3 – Having participated at AN/ANI would not at all necessarily inform someone of discretionary sanctions on a topic/dispute, if DS wasn't mentioned there, since DS is an AE/ArbCom matter, not an AN/ANI matter. This is part of the problem of treating AE as some special little fiefdom, and forking the "place to go to get administrative attention to resolve a problem". — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 21:11, 20 March 2014 (UTC)

• @ Roger Davies: Has there been a discussion about not using the alert template. For example it can be a lot more friendly (especially with new users) to leave a custom written message which still links to all of the applicable pages and includes advice on what they need to do differently. Would it be possible to include a point under the twelve month heading which includes this? Callanecc ( talk • contribs • logs) 05:33, 24 March 2014 (UTC)

(Butting in) Key thing is to focus on two distinct tasks: (1) letting someone know and (2) starting the official "awareness clock". In my view, we should encourage, or at least explicitly protect, original civil informal text. That is simple to do provided provided such text is divorced from the official "awareness clock" because the official template creates a clock log entry essential for the mechanics we're talking about. We can protect civil informal text by saying something about it in the "alerts" section, such as a portion of this suggested tweak. Maybe not all of it, but the boomerang portion certainly. NewsAndEventsGuy ( talk) 06:19, 24 March 2014 (UTC)

I'm referring to it starting the 12 month clock as well. For example if I'm warning someone for edit warring on a page covered by DS, I'd much rather tell them not to edit war and that discretionary sanctions can be used if they continue. Then slapping {{ Ds/alert}} on their talk page and writing something about edit warring. Callanecc ( talk • contribs • logs) 06:31, 24 March 2014 (UTC)

You were clear the first time, but thanks for reiterating. What I was trying to say is that there are advantages to "starting the clock" only with the template. (A) it allows tagging and thus an entry in the log so we can easily tell if someone was alerted within 12 months, (B) it eliminates complaints that the alert that starts the clock is a form of stealth bullying, (C) others? Anyone could still write informal text to help newbies. I think our only disagreement right here is that in my view, the net advantages of using the template to start the clock outweighs the net advantage of also starting the clock with original text. NewsAndEventsGuy ( talk) 10:07, 24 March 2014 (UTC)

I don't disagree, I just think that forcing people to template other editors could have unintended consequences given the dislike some experienced editors have of being templated. Callanecc ( talk • contribs • logs) 09:32, 25 March 2014 (UTC)

Personally, I prefer the original over Alt 1 above. But I think this would be even more clear.

NewsAndEventsGuy ( talk) 15:21, 22 March 2014 (UTC)

That's nice and simple, although I would change 1 to be "The editor has received and/or appealed a section in the area of conflict". This makes it clear that someone who has received a sanction but without participating in any process regarding it is aware. If you appeal a sanction you are aware of it, regardless of whether your appeal is successful or not. Thryduulf ( talk) 17:04, 22 March 2014 (UTC)

If I understood right, you describe an impossibility.... It makes no sense to provide for editors whose only awareness derives from a past sanction under rules requiring even earlier awareness as a precondition. NewsAndEventsGuy ( talk) 17:13, 22 March 2014 (UTC)

No you've misunderstood. This is providing for the following cases (1) a person has received a sanction in the relevant topic area in the past 12 months, whether or not the chose to participate in a process regarding its imposition. (2) a person has appealed (successfully or otherwise) a sanction in the relevant topic area in the past 12 months, regardless of when the sanction was placed and regardless of the process used to make the appeal.

The idea behind the first one is that refusing to engage with the process and/or not making any statements in your own defence does not make you unaware of a topic area you have been sanctioned in. For 2, the difference is that appealing a DS sanction makes you aware DS is active in the topic area, regardless of whether the appeal was successful or not.

While at first all these sanctions will have been placed under the previous (present) regime, but we are discussing a new regime that will be in place (hopefully) for several years at least, not just transitional arrangements. If you have been sanctioned within the past 12 months that means you are aware of DS, regardless of how or whether you were aware at the time the sanction was placed - i.e. the sanctioning starts a new period of awareness Thryduulf ( talk) 18:50, 22 March 2014 (UTC)

• What about someone who was under a sanction but it just expired? They are treated as unaware, but someone whose sanction is just about to expire is treated as aware. That seems to be a mistake. I suggest that "under an existing sanction" should be "was under sanction for the area of conflict at some time in the past 12 months" (which needs some rewording). Zero ^talk 20:04, 22 March 2014 (UTC)

I too would like to see the addition, at a conceptual level, as to how to handle expirations. See my prior comment under the original version of text. NewsAndEventsGuy ( talk) 20:57, 22 March 2014 (UTC)

I understand and agree with your intent, but your proposed wording introduces a new problem - namely someone who's sanction expired 1 day shy of two years ago is still regarded as aware, despite 12 months being regarded as the maximum elsewhere. The best I can think of is a general instruction to treat all time periods with common sense rather than rules lawyering about them (as I commented elsewhere). Thryduulf ( talk) 22:53, 22 March 2014 (UTC)

Thanks for all the input, folks. Here's an amended draft, based closely on N&EG's draft, and incorporating various comments. I've added an override for the twelve-month window.   Roger Davies ^talk 06:00, 23 March 2014 (UTC)

• Unless there's something horribly flawed with this, I'll substitute the current text with this in the next day or two,   Roger Davies ^talk 06:00, 23 March 2014 (UTC)
  • Two comments.
    1. The gaming the system text is good, but could be more generalised to "Any editor apparently gaming the system, for example by engaging in serious disruption just outside the twelve-month window, may be dealt with as if the alert was still in force. That removes any possibility of gaming the condition about not gaming the system and lawyering about whether disruption is "serious" or not.
    2. Reading this again, I'm not sure that point 1 is actually needed. If a sanction has been (successfully) appealed then the editor has participated in a process at one of the listed venues. If I'm wrong about this and it isn't redundant, I still do not see the need for the appeal to have been successful - if you appeal a sanction you are aware of the DS whether the appeal is successful or not. Thryduulf ( talk) 10:37, 23 March 2014 (UTC)

• Yeah, is #1 redundant? I notice that #1's text just says "appeal" (generally) while #2 says appeal at a specific venue. Seems like #2 could be broadened as to venue and we could do away with #1. NewsAndEventsGuy ( talk) 11:57, 23 March 2014 (UTC)

• Since WP:Gaming the system is always implied, I'm dubious that violating WP:BEANS by explicitly bringing it up returns enough benefit for the cost. If someone's 1-week sanction runs out, in nearly all cases they are still inside their original 12-month alert window. If someone's 1-year sanction runs out, and they make 1 single disruptive edit, then (A) that 1 edit shouldn't trigger the "serious disruption" clause, and (2) in an active conflict area, while we wait for more evidence of gaming, some editor is going to look at the alert log and give them a new one to start the new clock.... and when its time for an AE complaint, then part of the complain should be GAMING. So what do you think? Is the expiration business adequately covered by relying on existing mechanisms? NewsAndEventsGuy ( talk) 11:57, 23 March 2014 (UTC)

• If we have to have the 12 month window (which I think in practice will be difficult) this is a better version. I agree with most everything Thryduulf said - especially in relation to appeals.

  Fundamentally once you've been sanctioned (unless the sanction was out of process and/or a result of sysop misconduct) your name is on the final decision page and in my view one can never become unaware after that point. So even if an appeal is granted this will not make them unaware - in fact it'll make them more aware of the RFAR ruling.

  Thus I'd suggest changing is under an existing sanction for the area of conflict to has been sanctioned in relation to the area of conflict.-- Cailil ^talk 11:48, 23 March 2014 (UTC)

When I was 25 I would have agreed with the last point ("once sanctioned always aware"). But alas! the ravages of age! I've already forgotten more than I now know! I like the clean slate restart suggested by the 12-month clock, and I think it communicates something positive towards editor retention... for some eds who unwittingly put a foot down wrong, something in the notion that the stain will come out might - to them - promote a positive environment. Warm snugglie fuzzy blankets all around. NewsAndEventsGuy ( talk) 12:08, 23 March 2014 (UTC)

I think you're actually confusing issues NaEG. If a new editor "unwittingly" breaks the rules they wont be sanctioned (especially after this review). Now if an editor's reaches the point of being topic-banned this should stay on the record for that topic in terms of their awareness. The whole point of an indef topic ban being lifted is that the user now understands the rules of working in that area. They cannot then be simultaneously unaware of them. It's illogical & impractical to maintain such a legalistic fiction.

Once sanctioned (not admonished, alerted or warned - sanctioned) nobody should EVER be considered "unaware". That doesn't mean that they are "stained", in fact it should be communicated that reforming & being aware & abiding by the rules is better than claiming unawareness & repeatedly line-stepping. I'm not sure how this (rewarding reform) can be done because any time-based system for "awareness" will always mitigate against rewarding reform, & unfortunately (& unintentionally) facilitate rules-lawyering-- Cailil ^talk 14:13, 23 March 2014 (UTC)

• I meant "unwittingly" loosely, as in.... an editor who stumbles into conflict and by getting emotionally hooked freefalls faster than they learn the culture here. We should make it easy for that person to try again, without the legalistic folderol of begging for a sanctions lift. NewsAndEventsGuy ( talk) 15:47, 23 March 2014 (UTC)
• Also, as for repeat offenders intentionally waiting out the clock instead of reforming, the solution there is pretty simple. Much harsher/longer sanctions for gaming the system. If new procedures that encourage a relatively small number of SOBs to game the system in this way seems like a painful outcome, well, that's the price of de-stigmatizing the giving of notice for everyone. NewsAndEventsGuy ( talk) 15:52, 23 March 2014 (UTC)

The purpose of the 12 month window is for people who are not continually active in a topic area but who return after an extended absence. If someone receives a sanction, especially a time-limited one, then leaves the topic area, you cannot expect them to know whether the area is still (or now) under discretionary sanctions a couple of years later. For example in early 2005, shortly after I began editing, I was involved in the then-contentious area of Thomas the Tank Engine and Friends, but I haven't edited the subject for several years, can't even remember what the dispute was about and have no idea whether it is still contentious or not. It didn't raise to the level of topic bans, but if it had any sanction I received would have long expired. I would definitely need an alter to let me know if the topic was now under discretionary sanctions. Thryduulf ( talk) 16:20, 23 March 2014 (UTC)

Thryduulf don't get me wrong in the case of issues not rising to sanction (topic ban, page ban, long block) I fully understand (even if I hasve concerns about) the 12 month window. But TBH I've yet to see a case at AE where someone has actually forgot about an RFAR. It may have happened but in my experience at AE this is a solution looking for a problem.

I'm really only talking about the situation where someone has been banned/blocked that they should always be cognizant of an RFAR. Especially when the criteria for that sanction being lifted is cognizance of the RFAR ruling. At least in the case of indef sanctions (where they can be reimposed if misconduct recurs) we need to say that 12 months good behaviour does not prove ignornace (i.e unawareness)-- Cailil ^talk 19:29, 23 March 2014 (UTC)

What we have here I think is partly a terminology problem. In the way I use the words, and how I understand how they are used in this draft, topic bans, page bans, blocks, revert restrictions, etc. are sanctions, whether imposed by arbcom directly, at AE or by the community elsewhere (e.g. [1]). In no case are we saying that 12 months of good behaviour proves ignorance, but that after >12 months of not (a) interacting with DS processes related to the topic area, (b) being sanctioned under DS related to the topic area, and/or (c) being explicitly made aware that DS is active in the topic area, then awareness is not proven. In practice what this will mean that someone who is aware that DS sanctions are active can be sanctioned after 1 incidence of misconduct, whereas someone who isn't formally aware can be altered to DS after a first incidence of misconduct and sanctioned after a second. There is no loophole. Thryduulf ( talk) 23:48, 23 March 2014 (UTC)

• @Thryduulf, this part of what you said isn't quite right: "whereas someone who isn't formally aware can be altered to DS after a first incidence of misconduct and sanctioned after a second." As I understand the new system, there is no requirement for misconduct before the alert can be sent. For example, I read the draft text about (paraphrased) "anyone can alert anyone" as saying that I could hypothetically alert you to {{subst:alert|topic code – Tree shaping}} even if you have never visited any of those pages. The only text in the draft that frowns on that is whether my templating someone who never stopped by is "disruptive". But after your very first edit - even a wonderful one - then anyone can "alert" you. And in my opinion, if the goal is to de-stigmatize alerts, someone should. Ubiquity is the antidote to stigma. NewsAndEventsGuy ( talk) 00:16, 24 March 2014 (UTC)

• You are correct there is no requirement for misconduct (and that this is a good thing). My point was that if someone, who used to be aware but no longer is, makes a disruptive edit to the topic area it is not the disaster that @ Cailil: seems to fear. Thryduulf ( talk) 01:09, 24 March 2014 (UTC)

• Yeah, I got that gist the first time and I do agree with it. NewsAndEventsGuy ( talk) 06:07, 24 March 2014 (UTC)

Well I think we're still talking at cross purposes Thryduulf but if you're okay with this then so be it. It less that a repeat offender repeat offending will be a disaster and more that this will cause bureaucracy and put new sysops off AE IMHO (& I honestly still believe that it is at variance with the way that we administer sanctions and appeals in quite a significant & illogical way). We've ended up creating a new level of bureaucracy, and a new technical definition for awareness of DS in a process that was initiated because of excessive legalistic thinking at AE in general - that paradox is fundamentally concerning to me. But if consensus here is that a time based window has to happen then fine. But I reserve the right to say "I told you so" (yet I hope I wont get the chance)-- Cailil ^talk 14:53, 24 March 2014 (UTC)

@Cailil, (A) If it is you and me speaking at cross purps, I'd welcome the chance to clear things up at my talk page without further clutter here. (B) When you say "new layer of bureaucracy, are you referring to the "awareness" provision and associated alert template or something else, because while I do see changes in the mechanics I don't that new layer you spoke of, e.g., there is no new preliminary filter of "Spokespeople" nor "conflict reviewers" nor "clean hands referee" one must pass through on their way to AE. So are we talking about new mechanics within the existing organizational structure, or some "new layer of bureaucracy" I have not yet noticed? NewsAndEventsGuy ( talk) 15:01, 24 March 2014 (UTC)

I don't understand the new bureaucracy comment either. As it currently exists editors have to be aware of DS before they can be sanctioned under them. Under the new system nothing changes except that notifications that may or may not be warnings and may or may not imply misbehaviour are replaced by alerts that are explicitly neither warning nor aspersion. Thryduulf ( talk) 15:37, 24 March 2014 (UTC)

Oh, please get rid of the the "gaming the system" sentence. It should state a time period and stick to it. I don't get booked for speeding when I deliberately drive just under the speed limit. If 12 months doesn't mean 12 months, change it to what it means. Zero ^talk 10:35, 24 March 2014 (UTC)

The gaming the system comment could be better phrased, but what it means is "approximately 12 months, but use common sense". The intent here is that it be treated like the three revert rule - if your fourth revert is 24 hours and 1 minute after your first you are in breach as much as if it were 23 hours 59 minutes. Thryduulf ( talk) 11:11, 24 March 2014 (UTC)

It should be removed from the 3RR page too, where it damages a crystal clear bright-line rule by making it vague and fuzzy. If 4 reverts in 24 hours and 1 minute breaks the rule, change "24 hours" into "24 hours and 1 minute". Why write rules in such a way that argument over the boundaries is encouraged? Much better to define the boundary precisely and stick to it. Zero ^talk 13:30, 24 March 2014 (UTC)

Because that would encourage gaming the system and rules-lawyering. The intent isn't to stop people making 4 reverts in 24 hours, the intent is to stop people blindly reverting and discuss things, the 3 in 24 is a guideline as to the boundary between acceptable and unacceptable. In this case, the intent is to ensure that people are actually aware that discretionary sanctions are in force, to stop people being able to falsely claim they are not aware, and to stop excessive templating. To this end it is felt that awareness lasting approximately 12 months is reasonable. If you deliberately edit just outside the window then you are obviously aware of it. Thryduulf ( talk) 13:44, 24 March 2014 (UTC)

• @Zero000 and Thryduulf, the equally reasonable contrasting opinions have been stated, and there isn't an objectively correct answer. Let's not go around too many repeat reiterations that say the same thing of the multiple similar expressions because....... well shooooooot........that would be redundant and mean the same thing. NewsAndEventsGuy ( talk) 14:45, 24 March 2014 (UTC)

View current draft

• Sometimes the alerts come from people who editorialize on them in an ominous or threatening manner. See an example HERE. IMHO this should be strongly discouraged/dealt with/punished. Lou Sander ( talk) 01:30, 16 March 2014 (UTC)

• I agree it shoulds be discouraged. The template is the solution.   Roger Davies ^talk 13:50, 16 March 2014 (UTC)

• Alerts are far superior to "warnings" as a first alert that there is an issue; I stumbled into an area once that was under DS and I was unaware of it, made a comment on a peripherally related issue and was "warned" for simply being involved. It seemed over the top. Having an alert should be mandatory before anyone not clearly involved previously could be warned or previously sanctioned. Montanabw ^(talk) 02:08, 16 March 2014 (UTC)

• "Editors who abusively issue alerts may be sanctioned" - ok, but by whom? Via discretionary sanctions? Or only by the Committee?

  In general, I recommend cutting much of the rules creep about alerts, such as the one-per-year rule. The more you legislate here, the more you will have to face complaints or clarification requests about editors using these rules as tools in their feuds with one another. The focus should be on making people aware of the rules and enforcing them effectively, not creating an overly complicated procedural framework.  Sandstein  11:05, 16 March 2014 (UTC)

• I would like to get simpler myself but not at the expense of turning it into conveyor belt for sanctioning.   Roger Davies ^talk 13:50, 16 March 2014 (UTC)

• Query: What is the operative definition of "abusively issue alerts"? And would issuing the "second alert within a year" be subject to automatic review lest editors figure out that simply having multiple editors issue successive aalerts to a single editor would trigger one rule but not trigger the "abusive alerts" rule? Collect ( talk) 12:06, 16 March 2014 (UTC)

• There is no abusive alerts definition. So groups tag-teaming alerts to piss off the opposition could find themselves warned off.   Roger Davies ^talk 13:50, 16 March 2014 (UTC)
• I have a pretty firm belief that post-adoption, various eds will develop new tools to aide the giving of 1-but-only-1 notice per year, and probably renewing notices after the year is up or sanctions expire. So FWIW, at least in my case, I'm not too worried about any long running bruhahas over disruptively passing out alerts....except for the ed who makes a bot to alert everyone about all the DS cases on everything. NewsAndEventsGuy ( talk) 16:12, 23 March 2014 (UTC)

• Mandating the use of a specific template is a recipe for wikilawyering. It should be sufficient to require that the editor's attention is drawn to the discretionary sanctions. Guy ( Help!) 12:19, 16 March 2014 (UTC)

• The template is the lesser of two sets of wikilawyering choices, I'm afraid.   Roger Davies ^talk 13:50, 16 March 2014 (UTC)

• Roger is right, what constitutes "attention is drawn to?" That would be a can of worms. Make it clear. Those who know about the situation will also know about the template. Arguing if a set of diffs does or does not constitute awareness will become a bandwidth and time sink. Montanabw ^(talk) 19:11, 19 March 2014 (UTC)
• Agree NewsAndEventsGuy ( talk) 16:12, 23 March 2014 (UTC)

• I think that on the whole this is very good. I'd prefer to see some language explicitly addressing rules lawyering about the interval between alerts - two alerts in 364 days is no more or less objectionable than two alerts within 366 days for example. Thryduulf ( talk) 14:08, 16 March 2014 (UTC)
• The language in this draft, "Any editor may notify.... This may only be done by...", almost looks like one is strictly prohibited from informally telling someone that they are in dangerous territory without actually using the template. It might be better to indicate that to satisfy the Awareness requirement for an "alert" this may only be done by.... I.e., that informal discussion does not count as formal notice. (Personally, I would be inclined to post informal advice before dumping a load of boilerplate on someone, particularly a newcomer, and I have seen other users do this, including administrators.) ~ Ningauble ( talk) 15:46, 16 March 2014 (UTC)

• These are very good points.   Roger Davies ^talk 16:15, 16 March 2014 (UTC)

• Agreed. Montanabw ^(talk) 19:11, 19 March 2014 (UTC)
• Agreed & important enough to chime in to say just that. I often write hopefully friendly original prose about DS, and people should be encouraged to do so. If there are bullies, well that is what DR is for. NewsAndEventsGuy ( talk) 19:16, 20 March 2014 (UTC)

• In context of complaints that some notifications were done in a ... overly personalized, accusatory, or otherwise problematic manner, I think that "only be done by" serves to attempt to do away with that behavior. That would not prevent someone from separately doing an alert, and sending personal comments to the effect that you object to the behavior in a more detailed manner. But it would serve to separate that the official "alert" had no associated value judgement and that the comments were individual comments by an uninvolved reviewing admin. Even if left by the same admin. Sensitivity and perception of actions, etc. Georgewilliamherbert ( talk) 23:20, 19 March 2014 (UTC)

• Agreed, and I say that as a party aggrieved by such an "overly personalized, accusatory, or otherwise problematic" DS notice; the concern to me it not an idle one. It's crucial to separate the official, neutral notice that DS applies to a topic area (and any logging of such a notice), from any editor's (including an admin's) accusatory editorializing about why they feel some other editor needs to be put on such notice. In one case alone, the failure to distinguish them (including continued-to-this-very-day refusals by some Arbs to acknowledge any such difference) has resulted in three productive editors resigning from the project (other than to try to resolve the dispute), for a year and counting. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 18:38, 20 March 2014 (UTC)

• As I understand it, the phrase "only be done by" is to create an objective test regarding explicit notice at user talk. To the extent people were reacting to past bullying under the new system that is easily addressed by (1) letting the new culture of informational alerts take root and (2) simply using existing WP:DR remedies if any original prose about DS is out of line. It would be a shame to start punishing good faith original prose just because some folks understandably would like to legislate away bullying, because bullies are creative SOBs that will find workarounds. NewsAndEventsGuy ( talk) 19:18, 20 March 2014 (UTC)

I've added words about formal/informal to the new draft below,   Roger Davies ^talk 08:58, 23 March 2014 (UTC)

• Participation as an involved party. Debresser ( talk) 22:56, 16 March 2014 (UTC)
• It appears that in order to place the template without being considered abusive, a user is required under pain of sanction to exhaustively search through all edits that have been made to the other user's talk page during the previous year to see if a warning template was ever placed and subsequently deleted. This encourages (even more than before) each wikilawyering user to rapidly delete warning templates from their own user page, in the hopes that future editors will miss the previous warning and be sanctioned. It also makes it time-consuming and risky to ever choose to apply an alert template. -- Noren ( talk) 14:50, 17 March 2014 (UTC)

• As a sometimes AE administrator; if you re-alert someone with an alert visible for the same topic on a short talk page I would consider this a violation. If it was near the top of a 50-subsection talk page and you didn't notice it, I would not consider it a violation, but would ask them to look more carefully. If it had been removed from the talk page (normal user allowed to archive / remove content on their talk page) I would not consider it a violation. We can attempt to write up rules to cover all this, in case you're afraid of that, or hope that the AE admins will not be overly aggressive and vindictive and will adhere to the spirit of this (and the reasonableness provisions elsewhere...). Georgewilliamherbert ( talk) 23:20, 19 March 2014 (UTC)

• If I've learned anything about AE it's precisely how prone it is to wikilawyering, gaming and vindictiveness, so yes, it would probably be very helpful to be explicit here. The kinds of users liable to repeatedly earn this templating are precisely the sort to waste everyone's time with nitpicking complaints about how/when/where/by-whom the template was left. This will only serve to increase, not decrease the level of legalistic FUD surrounding AE, DS and ArbCom generally. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 18:36, 20 March 2014 (UTC)

• The point raised by Noren scares the crap out of me also. On the other hand, I don't want to be peppered with the same templates every few weeks either. I'm only just starting to pay attention to draft 3, but during draft 2 discussions there was some chitchat about tagging the alerts/loggings. QUESTION: Suppose we decide that more than 1x/year for any given subject area is too much, and I try to alert Roger for a second time during his year. Can't the system be taught to fish-slap me with a friendly notice that Roger was already alerted less than 12 months ago? If the alerts are friendly (and not for cause), then everyone should be encouraged to spread the word without quaking in their boots that a friendly giving of a friendly notice might result in being struck by Zeus' bolt of DS lightning. NewsAndEventsGuy ( talk) 19:25, 20 March 2014 (UTC)

• You scare too easily, N&EG ;) We can set a bot but this is getting extraordinarily over-engineered. In any case, giving someone (or even half a dozen someones) a second alert is scarcely abusive.   Roger Davies ^talk 19:54, 20 March 2014 (UTC)

Eeeeeek, thanks for the explanation/reminder addressed to Noren (below). NewsAndEventsGuy ( talk) 20:00, 20 March 2014 (UTC)

• No, Noren, it's easy. The new templates include a tag for the tag editor. To check if someone has had an alert, go to their contributions page and type "DS" or whatever into the "TAg filter" box just under the "Namespace" selector. Hit go and all their DS alerts should display.   Roger Davies ^talk 19:54, 20 March 2014 (UTC)

Come to think of it, should there be text about those mechanics in the alerts section, or at least a link to such text residing elsewhere? NewsAndEventsGuy ( talk) 20:01, 20 March 2014 (UTC)

Yes of course there should (and it's on my little list). I'm trying to get the structure sorted first,   Roger Davies ^talk 20:09, 20 March 2014 (UTC)

e/c

Regarding "abusive" repeat notices, I'd like to see a litmus test tied to a functioning log of tags. We could maybe address both points by adding something like

Before sending a user a DS-alert, check their WP:AlertLog. Sending an alert for any given subject area within 12 months of the user's last alert for that subject area (as shown in their alert-log) may be deemed abusive and could be grounds for sanctions against the sending party.

Thanks for the hard work on this NewsAndEventsGuy ( talk) 20:21, 20 March 2014 (UTC)

Sorry, I don't agree. I really don't want to see the creation of a specific new crime. On second thoughts, the way round this is to change abusive to disruptive, which is already well established and understood. I'll do that now,   Roger Davies ^talk 08:36, 23 March 2014 (UTC)

My main concern about DS is the alerts. I've seen them used as weapons to silence editors with a particular point of view. I'd prefer if only admins could post alerts but since that is unlikely to come to pass, I'll post a couple of other concerns:

1. A template should be used and, if it is posted by an editor (not admin), there shouldn't be any addition of statements like, "If you continue editing like this, you will be blocked." Editors should not threaten other users with blocks because it is not within their authority to follow through on these threats.
2. DS alerts should be given out for disruptive behavior, not for having a differing opinion. Alerts shouldn't be used to settle content disputes on articles by intimidating editors with unpopular points of view. They should be about conduct, not content.
3. Following #2, if there are two sides to a dispute on a topic subject to DS, both sides' conduct falls under DS. The specter of DS covers all editors working on problematic subject area or editing feature, not just some editors (like new users, IPs, inexperienced editors who are not savvy about AE). There shouldn't be two different standards of behavior depending on what side of an argument one is on.

Liz ^{Read! Talk!} 00:25, 23 March 2014 (UTC)

Yes, Liz, I agree that the common characteristic of disputes under DS is the win-at-any-cost-mentality that often prevails among the protagonists (or should that be antagonists?). We have gone to great trouble to blunt the use of alerts as weapons by adopting a no-fault vanilla for-your-information approach.   Roger Davies ^talk 06:09, 23 March 2014 (UTC)

Indeed, Roger. And within a short time of deployment eds (me for one) will pay attention that new names popping up in the dispute pages have been alerted, even if all they did was to make enormously beneficial contributions. Why would I do that? Because the alerts are FYI friendly did-you-know type things. If we're not blanketing all participants in the DS areas then the selective issuance of the alerts will continue to be perceived as accusatory. The best defense against perceptions of "accusatory" is ubiquity, i.e., treat everyone the same in terms of alert-issuance. NewsAndEventsGuy ( talk) 13:54, 23 March 2014 (UTC)

This incorporates the point about formal/informal alerts. Please read this for creep/drift,   Roger Davies ^talk 08:59, 23 March 2014 (UTC)

• Both disambig and repeated de-stigmatizing hammering neeeded

In this discussion, when an editor says "alert" they might be talking about

1. The template {{ Ds/alert}}
2. When Editor-1 uses the template to formally "alert" Editor-2 that DS are in play
3. When Editor-1 simply talks to Editor-2 informally about DS
4. Talking about starting the 12-month clock by way of an "alert" ignoring the implied connection to the template

To reduce confusion, I think we need to coin a term. At the expense of adding words, here are some suggested tweaks to hopefully make the draft more clear and to really hammer home the FYI friendly nature of the alerts, so that this new attitude can take root as quickly as possible.

A "DSAlert" is hereby defined as the giving or receipt of a standard template message approved by the committee, currently {{ Ds/alert}}. The text of the standard alert template forms part of the procedure and may be modified only with the committee's consent.

A DSAlert

• Neither implies nor expresses a finding of fault,
• Is purely informational,
• Automatically expires after twelve months, and
• Cannot be rescinded or appealed

The committee intends the DSAlert to only be an "FYI" that makes editors aware that discretionary sanctions are in effect and provides information about the procedure, nothing more or less.

The only way to give an editor a DSAlert is by posting the approved template on the talk page of the editor being notified. Any editor may give another a DSAlert for an area of conflict. Whether or not the template has been given, any other commentary about discretionary sanctions posted on another's talk page must conform with existing wikipedia principles and the final ruling in the case or the speaker may incur WP:BOOMERANG sanctions.

No editor should receive more than one alert per area of conflict per year. Editors who issue alerts disruptively may be sanctioned.

Thanks for reading. NewsAndEventsGuy ( talk) 16:12, 23 March 2014 (UTC)

View current draft

• Administrators already are terrified of complaints by the noisiest policy violators and refuse to sanction with even 24 hour blocks the worst kinds of violations of policy. I don't see any incentive for them to give any sanctions at all here. Bad editors are running riot and ticking off and/or driving out better ones. There needs to be a fix, like the foundation figuring out how to hire a few hundred part-time tough but fair ones who are not ruled by fear of action from policy violators. Carolmooredc ( Talkie-Talkie) 06:02, 16 March 2014 (UTC)

• Yep. This is about checks and balances though,   Roger Davies ^talk 13:52, 16 March 2014 (UTC)

• I recommend omitting "participation as a party in an AE request does [create involvement]". Otherwise, I as an administrator would no longer be allowed, without becoming involved, to submit a case of suspected misconduct to AE to seek the advice of my colleagues, but would instead would have to act unilaterally without discussion. In cases where admins are genuinely personally involved, this will already be apparent from the context.  Sandstein  11:08, 16 March 2014 (UTC)

• Only in theory. In practice, I think we all know that admins are generally very loath to apply WP:INVOLVED against other admins, or criticism other admins about much of anything, and that even once it has happened, more admins will often immediately leap to the defense of the involved one, because for some it is more important to stand firm in defense of DS and other status quo facts of WP administration and its culture than to deal with occasional over-stepping of administrative authority. Administrative transgressions can be obvious and protracted without any action being taken, even after Arbs have suggested that an admin is going too far. Something, in particular, must be done here about the propensity for some admins to engage in what appear to be highly personal witch-hunts, doggedly pursuing actions against editors who have criticized their treatment under the DS system, and/or who are involved in disputes that the admin in question is personally intolerant about. While "participation as a party in an AE request does [create involvement]" may not be the right wording, there's something important to address there. Resolving it would make some admin's own wiki-lives much easier, by giving them clearer bounds.  —  SMcCandlish ☺ ☏ ¢ ⚞(^Ʌⱷ҅̆_⚲͜ⱷ^{^})≼  18:59, 23 March 2014 (UTC)

• I agree with Sandstein that taking a matter to AE for discussion in an administrative capacity should not make the admin involved. Further, I'm not convinced about the "imposes a disproportionate sanction" part, as proportionality for sanctions is not an exact science. One person's "disproportionate" might be another's "OK, but on the high end", and so on. I'd prefer a stronger qualifier, like "grossly" or "significantly". T. Canens ( talk) 14:30, 16 March 2014 (UTC)

• Thanks, Tim Let's try some language and examples to get a sense of scale of proportionality. Say, consensus is that a reasonable block would 24-hours. Would you call a 48-hour block significantly disproporionate? And say a 7-day one grossly disproportionate? In other words, how would you frame these (it's probably easier with examples).   Roger Davies ^talk 16:12, 16 March 2014 (UTC)

• Usually the reasonable sanction is not a fixed point but a range. If I would have blocked for 24 hours, then to me 48 hours is probably "high end of what's reasonable", 72 hours would probably be "disproportionate", 1 week "significantly disproportionate", and anything longer than that would likely qualify as "grossly disproportionate". T. Canens ( talk) 14:42, 20 March 2014 (UTC)

• Thanks Tim,   Roger Davies ^talk 19:57, 20 March 2014 (UTC)

• This is an important observation. If an administrator faces sanctions for in good faith attempting to enforce the Committee's decisions, because the committee decides that the sanction was disproportionally strict (or lenient?), you may have substantially fewer admins working at AE, not more. In addition, by making (mere) disproportionality not only forbidden, but even sanctionable, the provision would remove the "discretionary" part from "discretionary sanctions", because it would in effect remove the degree of discretion and independent judgment the Committee has so far entrusted admins with in applying sanctions. You are free to do so, of course, but I for my part would seriously consider abandoning AE if, based on the wording of this provision, I must expect to be desysopped if a Committee majority decides that a one-year block instead of a warning (or vice versa) would have been preferable in any particular case.  Sandstein  15:52, 16 March 2014 (UTC)

• I would no more like to see one-year blocks instead of warnings than the folks at WP:ANI like to. While the admin would probably not be desysopped, it would not pass without, let's say, feedback,   Roger Davies ^talk 16:12, 16 March 2014 (UTC)

• In addition, I think that the provision saying that imposing a sanction "idiosyncratically" is sanctionable is so vague as to be similarly problematic. Idiosyncrasy means "an unusual feature of a person", "odd habit", "eccentricity or peculiarity", according to our article. So... only sanctions from the Committee-approved schedule are allowed, if we want to be sure? But on the other hand, the draft authorizes "any other measures which the imposing administrator reasonably believes are necessary to prevent disruption", which not only allows but specifically calls for sanctions that are tailored to the individual circumstances of a case. This is contradictory.  Sandstein  15:57, 16 March 2014 (UTC)

The word "idiosyncratically" is indeed vague and it would help to know an example of the sort of thing someone had in mind in inserting that word. Clearly it isn't supposed to refer to a sanction that is merely novel but otherwise within an admin's reasonable discretion. On the other hand, since this section is only listing conditions under which arbcom might act, and we can assume some common sense on their part, it doesn't actually matter much. Zero ^talk 10:54, 22 March 2014 (UTC)

• I'd like to see an explicit mention in the role of the administrators whether this gives them authority to supercede regular dispute resolution (consensus, 3RR, etc.... ) once they decide to involke discretionary sanctions. Basically, does it given them the right to override consensus (assuming normal exceptions for consensus aren't met ), 3rr (again, assuming usual exclusions aren't already met). Not to hamper any sysop, but as a further check and balance.  KoshVorlon. We are all Kosh   10:49, 17 March 2014 (UTC)

• Good point, which I think has already been addressed as some point in the past by a motion or procedure. It does need to be covered here. The short answer is that admins should be very careful indeed about overriding existing consensus. A crude example of the problems that might arise is if, say, the community decide not to act for 3RR when in fact there's actually a 1RR on the article. I'll see if I can find the motion. It'll be a good departure point,   Roger Davies ^talk 10:59, 17 March 2014 (UTC)

• Yes, KoshVoron's concern is a crucial one. The number one problem with AE from an editorial standpoint is that it seems like WikiThunderdome, some kind of apocalyptic arena with no predictable rules, and an already rigged game with in-house champions. Outside-AE dispute resolution processes, policies and other rules cannot be randomly superseded by AE admins, or the entire system is a sham. Roger Davies's concern, that the opposite will happen (e.g. someone will get a pass on a 3RR when there's actually a 1RR restriction) is a rare type of occurrence. I don't think anyone's arguing that AE couldn't decide to impose a 1RR in place of a 3RR. What's at issue here is "discretion" to impose sanctions as if 1RR were in place when it is not. Another example would be to "broadly construe" some topical RFARB decision subject to DS, to broadly as to topic-ban someone for seeking dispute resolution at AE, AN, ANI or some other dispute resolution forum, just because happened to arise from a content dispute subject to DS, when the dispute resolution case is a meta-discussion about editorial conduct, not itself a content discussion. That question alone is a crucial one and has implications for actual sancitions already imposed and their validity. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 20:49, 20 March 2014 (UTC)

• Overall I think this is good but points 2 & 3 (disproportional and idiosyncratic sanctions) need work. While I'm in agreement with the base idea the points are not worded in the best way. And while we do need to address this we also need to do so with clarity because by definition DS contain an amount of idiosyncracy given that they are based on an individual's discretion, but it's the kind of idiosyncracy that is at issue. Rather than placing vague interdictions might I suggest giving guidelines within which sysops can use their discretion (i.e persistent abuse results in a topic ban; recidivism after a ban is lifted results in a higher sanction; sanctions should/shouldn't escalate beyond). If we're going down this road you can't expect sysops to be mind readers, and besides that we need to encourage more sysops who have never used AE to work there and a guideline for sanctions would help everyone (especially those who are unfamiliar with the page)-- Cailil ^talk 12:51, 18 March 2014 (UTC)
• As a separate point this wording "participation as a party in an AE request" makes perfect sense to me BUT is that how we want to frame AE? Does introducing an adversarial aspect to filing an AE help anyone or does it compound the legalistic issues?-- Cailil ^talk 12:56, 18 March 2014 (UTC)

• AE is already grossly adversarial, so much so that even daring to participate and not doing it well can lead to sanctions. The question here is how to make it either less adversarial, or more consistently and arbitrarily adversarial so that it stops favoring experienced gamers of its internal system. The main problem with AE from an outsider perspective is that the "discretionary" nature of all of this, and the culture of mutual protectionism among admins, means that AE is effectively "ANI to the death" – it's an administrative remedy and dispute resolution forum in which admins just get to do whatever they want, no holds barred. Whether you feel that's accurate or not is irrelevant; you must know by now that it's a common perception, and that problems with how AE operates and individual admins have operated within it have directly led to multiple productive editors leaving the project. (Note that much of the discussion in this section is administrative handwringing about maybe not being able to play fast-and-loose and make up remedies on the fly instead of adhering to a predictable, standardized approach to administration. This does not inspire regular-editor confidence!) — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 20:51, 20 March 2014 (UTC)

• I'm very conflicted on this section. As I'm a resigned editor (at least until a long-standing dispute is resolved, and my commentary here relates to resolving it and seeing that it doesn't recur), I won't be participating in a protracted back-and-forth discussion here; I'll write out my thoughts on this in essay and hope that it provides others some grist for further consideration milling.

  Some admins here, e.g. Sandstein and Calil, have some valid points that there's a chain of reasoning problem in the wording, even some outright contradictions; Calil has some suggestions on how to move the draft forward. But we need to account for the fact that some admins are markedly more prone to proposing year-long blocks/topic bans than average, and generally treating AE as a "special" zone in which there are greatly "enhanced" (in the legalistic, negative sense) risks for bringing or even participating in dispute resolution in that forum, and in which they in particular are "needed" to enforce these enhanced risks, despite the fact that none of this treating of Wikipedia a police role-playing game or a moot court serves the interests of the community at all. Sandstein's argument here seems to be advocating for a clear entitlement for admins to pursue such blocks without question, based solely on individual administrative discretion. I'd have to oppose that in no uncertain terms. But I would like to see the wording clearer, as it really isn't fair to AE-participating admins to have uncertainty about what they can and cannot reasonably impose as sanctions, under what circumstances.

  I think it would be disastrous to explicitly permit, or worse yet (for disputation levels) to implicitly suggest the permissibility of excessive solutions, especially on the basis that the admin most interested in responding is somehow most likely to know the appropriate thing to do (in point of fact, hotness to act it's a red flag of an WP:INVOLVED admin, and I think we all know this; unfortunately it means that part of the basis for DS is arguably logically unsound). The old canard from the early 2000s, that we just trust admins to do what is right because that's why we gave them the mop in the first place <insert warm fuzzy, tiny-editing-community feelings here>, is clearly not working in the 2010s. WP has changed, the admin community has many participants who do little encyclopedic editing work and are mostly dedicated wiki-cops in their time on the system, with insufficient common ground and empathy with regular editors. Not unrelatedly, "lowly editor" faith in both ArbCom and administration generally is at an all-time low. The tools are being wielded more like sidearms than mops by quite a few admins who specialize is this sort of enforcement instead of more varied and diverse administrative functions, and broader editing generally. (Some quantitative limits could help resolve that issue, but that's out-of-band for this talk page.)

  Meanwhile, it's clear that many admins themselves feel more put-upon than ever, more reluctant to act, more liable to be charged with wrongdoing, and many of them have resigned, too; we're not just losing good editors. DS vaugness "setting up" admins to feel they have more leeway than the community is actually granting them, and thereby entrap themselves, is not helping.

  I honestly think this calls DS entirely into question in the long run. The problems with it that have cost this project about a year of my and several other productive editors' participation, and are chilling the participation of many others, while doing very little to rein in actual inveterate disruptors, are still not resolved. I applaud the putsch to reform this controversial approach to administration, but have to note that it's still controversial, for reasons that haven't changed. One possible solution is to have AE decide that an ArbCom enforcement must happen, and then remand this decision to AN/ANI for actual enforcement, so that the remedy is decided by the admin community, not by some "lone star" maverick admin with an axe to grind or simply a fondness for aggressive solutions. Or make AE another AN sub-board, subject to normal procedure there from top to bottom. But theses idea are out-of-band too. I'd like to see them raised somewhere seriously, where ever that might best be discussed.

  In the short term, this wording in this section of the draft is problematic, but in more and sometimes different ways than some admins suggest. It would be convenient but a bad idea to resolve these problems in a way that only addresses enforcer-admin-focused concerns while bypassing countervailing ones.

  Finally, the ability of a clearly over-involved admin to escape a finding of involvement by "letter of the law" bureaucratically gaming the system needs to be curtailed, and this is the section in which to do so. Pursuit of sanctions against an editor by one admin or tagteam thereof, in venue after venue, is a problem, no matter how much they protest that they're just doing normal admin work. A displayed anger and impatience with a particular discussion type or topic (a "how dare you argue about that trivia?!" attitude) is every bit as much of a "dog in the fight" as taking one side or the other in that dispute, and in a sense is bigger dog. A desire to censor and sanction both sides for failure to shut up about a topic you can't stand is not necessarily less involvement, but may be double the involvement, especially if the dispute is legitimate and part of the consensus-building process (versus, say, the ethnocentric hate-mongering at the root of a number of RFARB cases). This is especially true when the dispute is about internal WP issues, not article content. Abuse of AE and DS to silence editors – sometimes just for using dispute resolution not for engaging in actual content disputes – is an ongoing problem here. The "discretionary" part not seeming to have bounds, and AE being its own microcosm, as if other rules did not apply, together form a serious systemic weakness that can be exploited and long-term gamed to the direct detriment of good-faith editors by any POV-pushing admin or tagteam. They also lead to an insular AE subculture increasingly at odds with the rest of the project. Whether one believes all of this has already happened or not, the clear potential for it is enough reason to nip it in the bud. There is certainly already a common perception that it has all happened, and that's problematic by itself for ArbCom and WP.

— SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 20:49, 20 March 2014 (UTC)

There's something wrong when a system which should be used to protect what is important (mainspace) is used to "protect" what is not (AE). As a non-content, non-admin editor, I consider it much more high risk to open an AE thread than to file an arbcom request -- if I screw up the latter the worse that will probably happen is a clerk will revert it and leave me a note on my talk page. Screw an AE request and I'm likely to blocked / restricted or some other nonsense. NE Ent 19:29, 23 March 2014 (UTC)

NE Ent Do you have any specific examples of this happening please?   Roger Davies ^talk 20:05, 23 March 2014 (UTC)

If one has truly clean hands, truly wants to resolve conflict as opposed to "win", starts by working up the DR ladder in a meaningful way, and saves AE for a truly wit's-end-last-resort, then I can't fathom why such a person would see AE as high risk. If any of those are not the case, then of course I can see why they might wanna think twice. NewsAndEventsGuy ( talk) 20:39, 23 March 2014 (UTC)

Well, yeah, [2]. (As has been mentioned to the committee once or twice or a zillion times already.) Read Neotarf's statement. [3] -- that wouldn't raise an eyebrow at ANI. Contrast that to the response when I posted what some considered not be good faith in the evidence talk section of the Civility Enforcement case [4] -- I got a polite request from a now retired arbcom clerk: User_talk:NE_Ent/Archive/2012#Evidence. (As I recall, we eventually came up with compromise wording we were both okay with). NE Ent 21:25, 23 March 2014 (UTC)

As you know, the AE case was just the tip of the iceberg. This dispute spanned four months, with multiple visits to various fora by substantially the same participants. WP:DISENGAGE is policy too. If you wish to discuss this further, please do so on my talk page. This is not the place to relitigate this.   Roger Davies ^talk 21:45, 23 March 2014 (UTC)

(This is dispute resolution, not a court of law, so there is no "litigating" to be done.) You've probably figured out I'm not exactly hesitant to take action -- there was sufficient wiggle in the declines of Neotarf's Dec 2012 filing that I considered refiling but arguing a case vis-a-vis DS rules is like trying to grab a cloud: I don't think there exists words that I could write that would gain sufficient traction for a case to be opened.

No, I do not "know that." I just know what I read at Wikipedia:Arbitration/Requests/Enforcement/Archive129#Apteva: "I'm inclined to issue a warning also to The Devil's Advocate, Ohconfucius, Neotarf and SMcCandlish. Their comments here " (that is, AE) " serve no useful purpose with regard to deciding whether the reported edits are sanctionable, and are also mainly concerned with casting aspersions on others, further personalizing the underlying dispute(s)." The stated reason for the warning made no mention of four months or anything like that. The warnings were for disrupting AE.

Fortuitously, this exactly illustrates what I meant in my contribution below: DS makes everything topsy turvy. Had an editor posted something like Neotarf's statement and made reference to a chronic history, would a committee accept a case? Of course not -- the standard comment is " Wikipedia:Requests_for_comment/Neotarf is a redlink." But in DS land, (at least my perception of it), one misstep and you're screwed. Obviously not important since I don't edit anyway -- but consider (again) the quote below from Technology Review -- that's the external perception of Wikipedia. That is important because Wikipedia is a social phenomena: how do you get folks to work hard writing an encyclopedia for free? Do "Discretionary sanctions" contribute to, or ameliorate that, and how do you know that one way or the other? (Four comments from DS "survivors" are hardly convincing evidence.) All I'm seeing from is assertion confirmation bias: We know it makes things better because it makes thing better or measurement using bogus metrics -- there's less disruption. (Fully protect all the articles and block any admin editing through them -- that would bring "disruption" to a screeching halt, won't it?)

Incidentally, I'd try to find other examples but yet another thing about AE is the admins there insist on wrapping all closed cases up in hats, which makes scrutiny difficult. No other board I'm aware of -- certainly not arbcom or AN or ANI or BLPN -- does that. Which makes AE less transparent than everywhere else. NE Ent 23:11, 23 March 2014 (UTC)

Very brief responses. If you'd like to discuss specifics to a case in more detail, you're very welcome to do so on my talk page.

• You commented in the January 2013 discussions at WP:AN and WP:AE, so I assumed you did "know that" and were aware of all the background.
• When a warning is for "casting aspersions ... further personalizing the underlying dispute(s)", it isn't for "disrupting AE", it's for failing to focus on content.
• No, I don't think we'd take a case just on the basis of the AE page, but the overall history of personalising disputes is likely to be persuasive, given the extensive personalised commentary both before and after AE, especially recently at the ARCA page and various talk pages.
• Finally, by the time something gets to ArbCom and AE, about the only pracit's too late for dispute resolution. It's usually a matter of prizing people apart with a crowbar.

  Roger Davies ^talk 02:13, 24 March 2014 (UTC)

@ Roger Davies: No one is supposed to "focus on content" in AE; ArbCom does not make content decisions, so content disputes are off-topic, meanwhile the entire purpose of AE, like AN and ANI, is addressing alleged problems with specific editors' behaviors. Please just stop for a moment and internalize this. DS cannot rationally be applied to dispute resolution, including AE itself, or the entire notion of dispute resolution on Wikipedia starts to unravel. There is no way to raise concerns with and attempt to resolve disputes over editor behaviors without focusing on said editors. This isn't "personalizing" anything; the very nature of such dispute is that they are inherently personal. Since you all seem to love these legal metaphors, you've set up a new court rule where any bailiff gets to summarily shoot any plaintiff who files a case against anyone, if the bailiff just doesn't happen to like something about the case's allegations or the plaintiff's personality. PS: I for one would appreciate you not trying to redirect crucial discussions like this to ratholes in user-talk where no one will read them. This discussion belongs here. PPS: Your point #4 above seems to be missing text. PPPS: If you really think that AE, RFARB and ArbCom generally aren't part of WP's dispute resolution system, try removing mention of them from Wikipedia:Dispute resolution (hint: it's an entire section) and Template:Dispute_resolution and see how fast you're reverted.  —  SMcCandlish ☺ ☏ ¢ ⚞(^Ʌⱷ҅̆_⚲͜ⱷ^{^})≼  02:42, 29 March 2014 (UTC)

I got a question of general nature, related to this sub heading. It concerns the closing statement made by T. Canens in an old case from 2011, here, which I'll quote fully:

"Discretionary sanctions address editor conduct, not admin conduct. The phrasing of the remedy itself draws a clear distinction between editors and administrators, and so admin actions are never sanctionable misconduct under discretionary sanctions (cf. judicial immunity). Therefore, while AE may overturn individual admin actions for involvement on appeal, I agree that the discretionary sanctions do not give us the power to declare an admin involved. It would indeed be very surprising for one admin to have the power to unilaterally declare another admin involved (remember that the discretionary sanctions allows for action by "any uninvolved administrator"; what a consensus of admins can do under the discretionary sanctions a single admin may do as well), and we should assume that arbcom did not mean for such an unorthodox and irregular delegation of authority unless the remedy allows for no other reasonable interpretation. I voice no opinion on the merits of the block. T. Canens (talk) 18:14, 4 October 2011 (UTC)"

Of course this statement was made by a single individual so it may not represent actual policy or even practice. But T. Canens was and has been very active in AE enforcement, is involved in the discussion of this draft and is a member of the Arbitration Committee. Yet, I find this statement both disturbing in its implications and ... well, monumentally stupid.

It appears to say that admins are automatically exempt from discretionary sanctions. Hell. It doesn't "appear" to say so, it says so: "admin actions are never sanctionable misconduct under discretionary sanctions". And then, in some kind of "I play lawyer on Wikipedia" it invokes the ... principle of judicial immunity as if it was in any way relevant or applicable.

The rest of that paragraph is essentially irrelevant wikilawyering. But I do want this clarified. Are admins automatically exempt from being subject to discretionary sanctions by virtue of having admin status? Are admin actions automatically exempt from discretionary sanctions by virtue of having admin status? Is there a qualitative difference between "admin misconduct" and "editor misconduct"? If so, what exactly is it? Like, "admin misconduct" is abusively blocking someone or moving a move-protected page or protecting to "own version" and that can never be subject to discretionary sanctions?

In general the problem with this draft, IMO, is in what it leaves out, not in what it says. Volunteer Marek ( talk) 10:38, 31 March 2014 (UTC)

Tim's point is that an admin is not involved just because they took an admin action. However I agree with you VM, Tim's wording that sysop actions are "never sanctionable misconduct under discretionary sanctions" is confusing. Admin actions IF they use the tools while involved are VERY sanctionable - BUT this would usually be the basis for a new RFAR or a Committee motion but not an AE case. This kind of issue goes to the Committee because it's seen as grounds for desysoping - something AE (& therefore AC/DS) has no power to do. Perhaps a line saying "cases of serious admin misconduct should be taken to RFAR" should be added?-- Cailil ^talk 11:15, 31 March 2014 (UTC)

I tend to disagree with Tim's conclusion there. Speaking hypothetically I can imagine a case where an admin action is grounds for the imposition of a discretionary sanction, provided the admin in question was abusing his tools and the abuse is not serious enough to warrant an arbitration case. Then again, this provision would cover very few cases, so I would not have a problem with saying that admin actions can never be grounds for the imposition of a DS and that these cases must always be sent to ArbCom. That said, when an admin is acting in his capacity as an editor, he ought to be treated like every other editor and, so, he can be sanctioned when warranted. Admin status ought not to be a shield. Salvio ^{Let's talk about it!} 12:13, 31 March 2014 (UTC)

View current draft

• It occurs to me (unfortunately, at this late stage) that this section may be best merged with the preceding one. AGK [•] 00:03, 16 March 2014 (UTC)
• This section raises administrators to a position of authority, rather than trust. There is no reason why a sanction should not be imposed by community consensus. I don't see any reason why a long-term editor in good standing who has decided not to be a sysop, should be prevented from judging consensus of a community discussion and notifying an editor of a restriction. Obviously for technical reasons enforcement by block requires a sysop, but we are janitors not policemen. Guy ( Help!) 12:22, 16 March 2014 (UTC)

• The principle is that individual administrators make actions for which they are personally accountable. If the person making the decision is unaccountable, the system is weaker. Do you really want admins on tool breaks issuing topic bans? Or follow your tack, non-admin editors issuing topic bans of their own volition?   Roger Davies ^talk 13:55, 16 March 2014 (UTC)

• That would lead to a never-ending torrent of questionable bans and incessant appeals. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 21:15, 20 March 2014 (UTC)

• I remain unconvinced that it is necessary to require the empty ritual of requesting-tools-back-at-BN-then-relinquishing-the-tools-again-afterwards for an admin who is entitled to get their tools back by request to modify their own sanctions. T. Canens ( talk) 14:39, 16 March 2014 (UTC)

• Alternate theory - the "Appealing and Modifying Sanctions" section includes this:

No sanction may be modified without:

1. the explicit on-wiki prior consent of the enforcing administrator; or

We could interpret that and the former administrators section to the effect that a former administrator CAN still give permission for any other administrator to then modify the sanction, even if the former administrator is not allowed to do so themselves.

That places the hurdle not at "former admin changes mind, re-ups their admin permissions, makes change" but at "former admin changes mind, convinces any other admin it's a good idea, they make change". Which should be easier and more reasonable.

Georgewilliamherbert ( talk) 23:27, 19 March 2014 (UTC)

Sounds reasonable. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 21:15, 20 March 2014 (UTC)

View current draft

View current draft

I iterate my concerns about "broadly construed" being usable as a sledgehammer where "reasonably construed as directly related" would suffice. Collect ( talk) 12:08, 16 March 2014 (UTC)

Agreed. This "broadly construed" nonsense has been really, severely problematic, including leading to sanctions being issued for a particpating in dispute resolution at AE, AN[I], etc., not for content disputes, just because the underlying dispute that led to the attempt at resolution was subject to DS. The theory here is that anything that "personalizes" the dispute is bad, and an admin can "broadly construe" comments about an editor's behavior at AE or AN[I] to fall within the DS realm, leading to a punitive block or ban for disputing another editor's actions via AE or AN[I]. This result is obviously absurd, since DR processes are always about editorial conduct (personalized by definition). Abuse of DS in this manner has happened several times already. It also directly violates ArtCom's own instructions, that even the most restrictive restriction, such as an outright interaction ban, cannot be used to prevent editors from seeking good-faith dispute resolution as at AE and AN[I]. There is no way that a vague "broadly construed" discretionary sanction can legitimately do this, yet it's been happening and ArbCom has done nothing about it. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 21:24, 20 March 2014 (UTC)

View current draft

• What do we mean by Prior to imposing novel or innovative interpretations or sanctions, administrators must discuss the proposal with other administrators at the AE noticeboard? What is a "novel sanction"? One that has never been imposed before? And "novel interpretation" of what? Salvio ^{Let's talk about it!} 14:18, 16 March 2014 (UTC)

• I share these concerns, and also ask if this provision requires admins to search all past AE archives (or even all user talk pages) to verify whether there is precedent for any particular sanction. That would not be realistic. It is not even realistic to expect that admins new to AE (which I understand you would like to attract more of, and I agree - more eyes are always better) have any but the most superficial knowledge of previous practices and sanctions. Up until now, they did not need to, because the Committee entrusted them to act on their own discretion in the project's interest. Is this to change?  Sandstein  16:05, 16 March 2014 (UTC)

• As the sanctions routinely used are specified under Imposing sanctions, no, it is not to change. AGK [•] 23:40, 18 March 2014 (UTC)

• novel or innovative interpretations means interpretations of policy. A novel … sanction is one not listed in the examples at Imposing sanctions. Surely the meaning of 'novelty' needs no explanation? AGK [•] 23:40, 18 March 2014 (UTC)

• Yes an no, because it depends on how you define a "novel sanction". If it's one not listed in the examples, then it's ok, but, in that case, it's better to make it clear.

  For instance, if I were to impose, a 2-rr per fortnight on an editor, such a sanction could be both novel and non-novel depending on what the interpretation of the requirement is: if, as you say, a novel sanction is one that isn't mentioned in the examples, then this restriction would not be novel (it's a revert restriction); if, on the other hand, a novel sanction is one that has never been imposed before, then I think this restriction would qualify as novel, as I'm pretty sure it has never been imposed before.

  Anyway, I'm not sure this provision is necessary: every sanction can be appealed and there are various levels of a-posteriori review, so weird sanctions can be quickly lifted and the record purged. Also, discretionary sanctions are based upon the idea that we trust the judgement of our administrators; why, then, should we say "we trust your judgement, then again we occasionally don't". Salvio ^{Let's talk about it!} 11:20, 19 March 2014 (UTC)

First the background: the text is based on Best practice includes seeking additional input prior to applying a novel sanction or when a reasonable, uninvolved editor may question whether the sanction is within the scope of the relevant case which is already in one of the AE procedures. I'm trying to roll everything up into one procedure.

Second, I'm happy if it goes, but I'd like to see it discussed a bit first,   Roger Davies ^talk 12:00, 19 March 2014 (UTC)

Question - are we having a problem with admins imposing DS sanctions which are novel, and weren't discussed beforehand?

There are some admins who I would worry might consider doing something like that, but they're not participating at AE.

If we leave the wording as-is, and interpret that as "What has been done so far has been reasonable in the context of this new guidline/policy" then we have a baseline which is not changing behavior but reminding others what the assumptions are. Perhaps legalistic, but not counterproductive.

If I missed a novel sanction dispute or problem, my bad, just let me know what happened and when... Georgewilliamherbert ( talk) 23:33, 19 March 2014 (UTC)

There are several such novel and unacceptable sanctions cases. See my month-long topic ban (which should be vacated) at the enforcement log at WP:ARBATC; I was banned not for engaging in a proscribed topic ban, but for using AE as a dispute resolution forum and not bringing a case strong enough, and for using AN and actually succeeding there. The clearly involved admin who closed that (the end of a months-long campaign to sanction me at all costs, by him and by an admin who later resigned under a cloud over similar user-harassing behavior) got away with misapplying DS, which do apply to content discussions, to censure me for editorial behavior meta-discussion. It's not the only case like that. It's clearly, severely wrong. Even outright interaction bans cannot be used to prevent good-faith access to dispute resolution, as ArbCom has clarified more than once. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 21:31, 20 March 2014 (UTC)

@ Roger Davies: speaking personally, I like the previous wording much better. This one is both broader and more nebulous; we should exhort admins to seek additional input before imposing a sanction which may be reasonably construed as outside the scope of the relevant case – and, maybe, also before imposing a novel sanction –, but it should not be a requirement. Salvio ^{Let's talk about it!} 13:11, 21 March 2014 (UTC)

View current draft

View current draft

• In the logging section it says that failure to log sanctions or page restrictions do not invalidate them. This language is also present in the page restrictions section, but it is absent from the section about sanctions. This makes the consequences of failing to log a sanction appear contradictory between the two sections. Thryduulf ( talk) 14:26, 16 March 2014 (UTC)

• Good catch. I'll tidy it up, one way or another, probably to remove the duplication,   Roger Davies ^talk 14:31, 16 March 2014 (UTC)

• This needs further requirements that the logging be done in a neutral manner. You're already aware of the year-long, multi-party dispute that illustrates why. We should also clarify that actions that are successfully appealed are to be removed from the log. Part of why that case has persisted, and even attracted non-involved parties to re-open it, is that some DS log pages, e.g. WP:ARBATC, have clearly been used with the effect if not the intent of being highly selective "walls of shame", particularly targeting a handful of editors by particular admins, and leaving most recipients of warnings/notices never logged there, thus concentrating and exaggerating the appearance that the targeted editors are inveterate disruptors, when no proceedings at AN, ANI, AE, RFARB, etc., have made any such determination. It's pure character assassination, effectively legitimized by the idea that the log is just some neutral record, than entries on it have no consequence and never bias anyone. Time-limited actions (short blocks or bans) should expire off the log after some period of time (e.g. a year after their termination). They should also be removable from the log upon determination that they were never valid, even if they have expired or were rescinded. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 21:41, 20 March 2014 (UTC)

View current draft

• Note to self: As drafted, this means the committee cannot reverse an enforcement decision unless pt. 1 or 2 is satisfied. This obviously isn't what we intended to say. AGK [•] 00:31, 16 March 2014 (UTC)

• In my view, allowing appeals to proceed at AE has only a chance at producing useful outcomes if administrators, or designated individuals, impose an orderly structure based on individual, short statements without threaded discussion, and actively close requests after a certain time, as is the case at AE. If the Committee can't organize and maintain such a structure, it's better to leave everything at AE, perhaps with pointers from elsewhere. The most recent flare-up of the complaints about discretionary sanctions warnings at WP:ANI illustrates these problems. It seems that editors want to launch (another) appeal of sorts against AE actions there, and of course it's already completely chaotic. Please refer to my comments there.  Sandstein  10:46, 16 March 2014 (UTC)

• I'd be happy leaving stuff at AE if it drew on a much greater pool of admin. We have too few people handling too many cases, with predictable results.   Roger Davies ^talk 13:58, 16 March 2014 (UTC)

• I prefer the wider audience as an available step, because of the too-small-pond problems at AE. AN is more freeform, but we need to acknowledge that questions about bigger picture items, compatibility of specified rules with "the bigger picture" of project goals and needs / IAR etc., and other wider picture items do form parts of some of the problem cases we have faced. Between small pond and bigger picture type problems, we need another wider venue somewhere. Georgewilliamherbert ( talk) 23:40, 19 March 2014 (UTC)

• The current language prohibits appeals by email unless the editor is blocked. I think it would be better to allow an email appeal for non-blocked editors, e.g. if it involved confidential information. On-wiki appeal can (and probably should) be noted as preferred and the committee should be free to respond to an email appeal with a direction to appeal on wiki if they do not see a reason for it to be private, but I dislike ruling it out in all circumstances. Thryduulf ( talk) 14:38, 16 March 2014 (UTC)

• I was thinking the same thing myself. Thanks, AGK [•] 23:35, 18 March 2014 (UTC)

• Nothing is mentioned about how page restrictions may be appealed, nor who may make such an appeal. This is a critical oversight imo. Thryduulf ( talk) 14:38, 16 March 2014 (UTC)

• Fair point. It'd be easily fixed by including page restrictions as a "sanction" for the purposes of that section. AGK [•] 23:35, 18 March 2014 (UTC)

• Nothing is mentioned about how restrictions may be queried or clarification sought. Thryduulf ( talk) 14:38, 16 March 2014 (UTC)

• As above. AGK [•] 23:35, 18 March 2014 (UTC)

• AE actions should explicitly be appealable to the larger pool of admins at AN. Many of us already interpret policy as necessarily enabling such an appeal, and some Arbs have also suggested this themselves. The idea that AE actions should be appealed to AE is absurd on it's face. The fact that certain AE-regular admins vehemently oppose this idea, which subjects them to scrutiny that's inconvenient to them, is telling.

  The entire concept of appeal means that the decision is reviewed by a higher authority. AN, being the pool of all administrators, is necessarily a higher authority than AE, a tiny pool of administrators working on a particular sort of activity. AE is also obviously appealable directly to ArbCom (hopefully more often via ARCA than RFARB, which is way more bureaucratic). If ArbCom wants appeals to mandatorily go through AN first before ARCA, that's up to them. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 21:51, 20 March 2014 (UTC)

  • AE is a creature of Arbcom. Is there any reason to believe Arbcom will not simply rubber stamp anything from AE, no matter how onerous?
  • The appeal issue is non-existent, since "warnings" would be replaced by "notifications", which by definition are not appealable. The only appeals under this new system would be by editors who have already been sanctioned (blocked or banned), so would have to take place by email, in secret. — Neotarf ( talk) 09:07, 22 March 2014 (UTC)
  • SMcC, it is already possible to appeal a sanction to AN and a consensus of uninvolved editors can lift it. Alternatively, an editor may choose to appeal to AE. That's his prerogative. Or he may even choose to skip this step entirely and appeal per saltum directly to ArbCom. All this is possible. Generally, speaking for me, I'd like to see the sanctioned editor discuss the issue with the imposing admin beforehand, but, as I said, it's my personal preference and not a requirement. Salvio ^{Let's talk about it!} 10:17, 22 March 2014 (UTC)

View current draft

View current draft

Additional discussion Information

To get this discussion on track could people post their impressions of discretionary sanctions that worked well, and those that worked badly? If there were real examples to discuss, some conclusions might be possible. Railing against ArbCom with legalistic arguments is not productive. If it can be shown that DS don't work, then they won't be used; if they are shown to be useful in certain situations, we need to be able to identify them. Jehochman ^Talk 21:17, 27 February 2014 (UTC)

In v2 I presented evidence that these alerts are a disaster waiting to happen. For whatever reason many admins have no freaking clue how intimidating the alert is to someone who has no knowledge of discretionary sanctions. That combined with the warfare mentalities that are commonplace in DS topics, the act of one perceived "side" alerting another perceived "side" is likely to cause needless friction and disruption (see the v2 link for a real-life example of this). I therefore seriously suggest something like the following, containing text that should be perhaps even sillier than what is there now.

		The Barnstar of Bravery
		Congratulations for bravely editing an article that falls under discretionary sanctions. What the heck is "discretionary sanctions"? Well to find out, you need to read the following hugely intimidating notice, though it won't be intimidating for you because of your proven bravery.

Please carefully read the following notice:

This message is to inform you that the Arbitration Committee has authorised discretionary sanctions for topics relating to , which you may have edited. The Committee's decision can be read here.

Discretionary sanctions are intended to prevent further disruption to a topic which has already been significantly disrupted. In practical terms, this means that uninvolved administrators may impose sanctions for any conduct, within or relating to the topic, which fails to adhere to the purpose of Wikipedia, expected standards of behavior and applicable policies. The sanctions may include editing restrictions, topic bans, or blocks. Before making any more edits to this topic area, please familiarise yourself with the discretionary sanctions system as sanctions can be imposed without further warning. Please do not hesitate to contact me or any other editor if you have any questions.

{{ Z33}}

vzaak 04:16, 28 February 2014 (UTC)

How this message is perceived:

Please carefully read the following notice:

This message is to scare the shit out of you, for the Arbitration Committee has authorised discretionary sanctions for topics relating to something which you may have edited, thereby placing yourself in a heap of trouble. The Committee's ominous decision can be read here.

Discretionary sanctions are intended to prevent further productive editing to a topic which has already been significantly disrupted. In practical terms, this means that uninvolved administrators will impose sanctions willy nilly for any conduct, within or relating to the topic, which they dislike. The sanctions may include editing restrictions, topic bans, or blocks, thumb screws, and will make Theon Grayjoy's torture look like a spa treatment. Before making any more edits to this topic area, please get your head checked, as sanctions can be imposed without further warning. Please do not hesitate to contact me or any other editor if you are crazy enough to want to proceed.

{{ Z33}}

Somehow this template needs to be reworked to assume good faith. How about reassuring editors that they can edit normally, that if they are acting in good faith, they won't be sanctioned, that the special conditions are designed to make editing in the area more pleasant, by making it easier to remove any disruptive editors. Be on the lookout for disruptive editors. Report them here. Don't worry, if you are contributing productively, you are not going to be the target of any enforcements. Please ask for help here or there or there if you have lingering concerns. Welcome, we love you, edit onward... Jehochman ^Talk 03:40, 28 February 2014 (UTC)

Because, Jehochman, no one is going to get one of these templates unless someone is trying to get rid of them. — Neotarf ( talk) 08:55, 22 March 2014 (UTC)

Oh please. Could we drop the conspiracy theories? Tony (talk) 09:23, 22 March 2014 (UTC)

• Agree assuming possible good faith would improve alert text

How about changing the beginning of the template text as follows

Just so you know....

You may have recently edited something related to the topic of (whatever). This notice is not about whatever you said in your edit. In fact, this notice may be sent to everyone working in that area, even editors making awesome improvements in good faith, which you might be. The purpose of this notice is just to let you know that in the past, related pages have been the location of disruption. As a result, the Arbitration Committee has..... (continue with existing template text) NewsAndEventsGuy ( talk) 12:30, 28 February 2014 (UTC)

That probably goes a bit too far in the opposite direction but if you can bump this up in a week or so (I've got my hands full at the moment), we can look at the alert text once DSR v3 has been posted.   Roger Davies ^talk 03:05, 8 March 2014 (UTC)

I strongly support movements in this direction. It is absolutely undeniable that the extant (and former, even more accusatory) version has often had precisely the effect that Jehochman says it does in his parody. The abuse of DS notices as menacing, stigmatizing warnings is twofold: both the wording of the notices and the ability of admins to add personal, stigmatizing remarks of the official case page (e.g. WP:ARBATC logs of the notices, are both individually polarizing and harmful. The entire nature of DS and the amount of fighting caused by it would be greatly reduced if the notices read like Roger's, and the logs were required to be neutral, as well as pruned of invalid and moot entries. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 22:00, 20 March 2014 (UTC)

Recently, I further edited the {{ Alert}} template. Comments welcome on the latest version. AGK [•] 16:28, 21 March 2014 (UTC)

Commented at Template_talk:Ds#trimmed NE Ent 11:33, 22 March 2014 (UTC)

[Placeholder for BLPSE discussion, per recent motion]   Roger Davies ^talk 02:54, 8 March 2014 (UTC)

At the end of the Housekeeping section, there's a light yellow bar that says This indicates the end of the draft v2 updated Discretionary Sanction remedy. Shouldn't that be "draft v3"? BlueMoonset ( talk) 00:22, 16 March 2014 (UTC)

Well-spotted. That's mistakenly been carried over from the previous draft.  Fixed, thanks! AGK [•] 00:26, 16 March 2014 (UTC)

I had a look at one section. This bit could be simplified, even though it would involve some duplication. I've made some trivial edits (in most cases underlined to mark them out for your consideration. There are two substantive changes: singular to plural in the second point, and plural to singular in the last point.

EXISTING

No sanction may be imposed on an editor unless they are aware that discretionary sanctions are in force for the area of conflict. An editor is aware if:

1. They:

   i.  were mentioned by name in the Final Decision of the case in which the applicable discretionary sanctions were authorised; or

   ii.  have at any time been sanctioned for their conduct for the area of conflict (and that sanction has not been successfully appealed).

2. They have, in the past twelve months:

   i.  participated in any process at arbitration requests, arbitration enforcement, or appeal at the administrators' noticeboard about the area of conflict; or

   ii.   alerted other editors, or have been alerted about discretionary sanctions for the area of conflict.

SUGGESTED:

No sanction may be imposed on an editor unless they are aware that discretionary sanctions are in force for the area of conflict. An editor is "aware" if any of the following pertain:

1. They were mentioned by name in the final decision of the case in which the applicable discretionary sanctions were authorised;
2. They have ever been sanctioned for their conduct in the area of conflict (and all such sanctions have not been successfully appealed);
3. In the past 12 months they have participated in any process concerning the area of conflict, at arbitration requests, arbitration enforcement, or appeal at the administrators' noticeboard; or
4. In the past 12 months they have alerted other editors or have been alerted of a discretionary sanction concerning the area of conflict.

Tony (talk) 02:19, 16 March 2014 (UTC)

I think this is a big improvement. The existing language is unclear regarding whether the options are AND or OR. It might be worth some language recommending giving an alert if there is reasonable doubt about whether someone is aware or not. I really want to discourage both alerting someone every few weeks and rules-lawyering about whether an edit made 364 days ago. Thryduulf ( talk) 14:00, 16 March 2014 (UTC)

A few tiddlies that may or may not be useful ... strike-out for suggested removal, underlines for suggested insertion:

For the purposes of arbitration enforcement Definitions:

• AE ("arbitration enforcement" noticeboard) is the forum designated by the Arbitration Committee for requesting, applying, discussing and appealing most enforcement requests. It is currently at Wikipedia:Arbitration/Requests/Enforcement.
• An area of conflict is the topic or group of topics specified by the Arbitration Committee when authorising discretionary sanctions.
• The committee is the Arbitration Committee.
• An editor is anyone and everyone who may edit and has edited the encyclopedia. (See Wikipedia:Editor.)
• An The enforcing administrator is the uninvolved administrator who places the sanctions authorised in this procedure.
• A sanction includes any sanction, restriction, or other remedy placed under this procedure.
• An appeal includes any request for the reconsideration, reduction, or removal of a sanction.

I wondered about "enforcing"—under the definition, such an admin would be placing the DS notification(s), not necessarily enforcing anything; and it's a high-level word that might be better as a more neutral term.

Postscript: An 'editor is anyone who may edit and has edited ..." – please check that this is not gameable. So an "editor" escapes the definition if they can't currently edit (i.e. they're blocked or banned). I may be splitting hairs, though.

Tony (talk) 07:56, 20 March 2014 (UTC)

More suggestions—

(A) Under "Behavioural expectations":

Editors editing within the area of conflict are expected to: Within the area of conflict, editors are expected to:

1. adhere to the purposes of Wikipedia;
2. comply with all applicable policies;
3. follow editorial and behavioural best practice; and
4. comply with any page restrictions in force within the area of conflict.

Failure to meet any behavioural expectation is grounds for sanctioning.

Comment: why not remove "editorial and" from the third point? Does it mean if you don't follow the MOS, or use faulty grammar, you might be sanctioned? Perhaps if it's retained it could be clarified?

(B) "may be subject to any remedy, including desysopping, that the committee consider appropriate." -> "may be subject to any remedy the committee consider appropriate, including desysopping."

(C) "or any other measures which the imposing administrator reasonably believes are necessary to prevent disruption" – "reasonably" is what went on in the admin's head, and is thus not contestable by the committee or anyone else. If you mean "or any other reasonable measures", could the word be relocated? Otherwise removed?

(D) "No sanction may be modified without:

1. the explicit on-wiki prior consent of the enforcing administrator; or
2. the clear and substantial consensus of either (a) uninvolved participating administrators at the AE Noticeboard or (b) uninvolved editors at the administrators' noticeboard. If consensus is unclear, the status quo prevails.

Thereafter, the sanctioned editor may:

3. appeal to the Arbitration Committee in a request for amendment. If the editor is blocked, appeal may be made by email through Special:EmailUser/Arbitration Committee (or, if email access is revoked, to arbcom-llists.wikimedia.org).
"

Problem in "Thereafter", which presumably refers back to a sequence that hasn't been stated as such. Do you mean "If a sactioned editor has been unsuccessful in attempts to modify a sanction via 1. and 2., the sanctioned editor they may:" ...?

"Noticeboard" is lower case elsewhere on the page.

(E) "A" is standard in legal text in a few places here where "the" is used; please consider changing. For example: "Only the an editor under sanction may appeal the sanction." Withdrawn ... I misunderstood. Tony (talk) 14:16, 22 March 2014 (UTC)

(F) "Nothing in this provision prohibits an administrator from superseding an existing sanction with a new sanction if fresh misconduct has taken place after the original sanction was applied." – I don't think "supersede" can be transitive. Please consider: "Nothing in this provision prohibits an administrator from superseding replacing an existing sanction with a new sanction if fresh misconduct has taken place after the original sanction was applied." – or "supplanting", but "replacing" seems plainer.

(G) "Nothing in this current version of the Discretionary Sanctions ...". Could you downcase as in the page title and elsewhere in the main text?

(H) "on-going" – I think "ongoing" is pretty standard nowadays.

(I) Don't like the plural "committee", like a cricket team; but <sigh> it's no big deal: the committee consider, And it's inconsistent: the Nutshell has "the Arbitration Committee has", not "have", twice. Further down after No. 3 in "Appealing and modifying sanctions", there's a "the committee has".

Sorry to bloat: first time I've really been through the text properly. Nice work by those who've shaped it: should be a simple reform with strong community backing, frankly, and I see lots of emotion spilling from past perceived (some actual, probably) injustices. Please, let's pull together and get this reform on the road. This is no time for personal pride, for keeping alive the hurt that we've all felt as wiki editors at one time or another. Tony (talk) 12:17, 21 March 2014 (UTC)

"Thereafter" was already spotted and is on our list of things to fix after this round. Thanks for the rest; these are good suggestions. AGK [•] 16:27, 21 March 2014 (UTC)

Very quick comment about (E). That introduces an inadvertent change of meaning. I'll work through these methodically later,   Roger Davies ^talk 12:27, 22 March 2014 (UTC)

Vision

The Vision "Imagine a world in which every single human being can freely share in the sum of all knowledge. That's our commitment."

Goal

1. To get as many high quality contributions as possible, or
2. To get as many high quality contributors as possible or
3. To get as many contributors as possible.

Same idea, different formulations; while #1 might be the ultimate goal, we can't know now where the contributions will come from in the future, so our immediate goal should be #3.

The perception

"The loose collective running the site today, estimated to be 90 percent male, operates a crushing bureaucracy with an often abrasive atmosphere that deters newcomers who might increase participation in Wikipedia and broaden its coverage."
— Tim Simonite, Technology Review [5]

Ouch. That's embarrassing. Aren't you embarrassed? I'm embarrassed -- I certainly don't go around sharing my wiki-addiction with folks in real life. If that's true we need to address it, if that's not true we need to address the perception that it is.

Connection

What's that got to do with Discretionary sanctions?

When we try to pick out anything by itself, we find it hitched to everything else in the universe.
—  John Muir

I've previously opined with some length about the strange duality of the committee's important / unimportant duality at The Committee, so I'll just briefly reiterate that while the number of editors by ArbCom may be relatively low, the symbolic importance of your actions is high before it affects perception of Wikipedia as a whole.

Your task

Wikipedia:Arbitration Committee: "It has the authority to impose binding solutions to disputes between editors, primarily for serious conduct disputes the community has been unable to resolve."

Wikipedia Adminstrator, Yin and Yang

Simplifying for brevity, administrators can be classified as authoritarian -- often acting quickly and decisively with regards to the many many wp-this and wp-that's Wikipedia is filled with, and holistic -- taking a more nuanced approach appropriate to a place where one of the five big rules is not to be overly fixated on the rules. While I personally believe strongly the holistic types are better for Wikipedia, all-in-all, I suspect both are necessary. (I'm reminded of the two Kirks in The Enemy Within; unfortunately, our article leaves out the writer Richard Matheson's insight that both parts of Kirk's personality were necessary for him to function effectively.)

A close observer of past arbcom's will note we tend to elect committees with both types of editors. Although the committee has some sort of theoretical "absolute power" -- certainly folks in the community foolishly look to arbcom to solve all problems -- this mix of authoritarian and holistic, a mind numbing bureaucracy, the fact that arbcom takes few cases, and is slow to act results in an extremely judicious in its use of power.

— Preceding unsigned comment added by NE Ent ( talk • contribs) 13:52, 16 March 2014‎ (UTC)

In really bad disputes, the administrator corps becomes paralysed. The gap you describe between the two types of admin becomes a void. This becomes occupied by warring factions. This is the core problem in most really big disputes. ArbCom's role is often to reclaim the topic for the community and community processes, by reducing the influence of the people trying to own and control it   Roger Davies ^talk 15:47, 16 March 2014 (UTC)

The arbcom should support those who are supporting the WP principles, and the arbcom decisions; not try to weaken all the participants indiscriminately. And many editing groups do not even have an admin interested enough in the subject to get familiar with the issues and participants. — Neotarf ( talk) 14:42, 22 March 2014 (UTC)

In plain English, what is your point? AGK [•] 23:34, 18 March 2014 (UTC)

If you don't understand the point, you are probably in the first group. These are more or less classic leadership style descriptions, although some would add libertarian and teleological to the mix. Why this matters, and the answer to your question is in the next section, as well as the reason this isn't a problem at Arbcom or WP, but becomes a huge problem at AE. — Neotarf ( talk) 14:42, 22 March 2014 (UTC)

This page in a nutshell: Arbcom can delegate its power. It can't delegate its wisdom

In normal space, these two tendencies -- authoritarian and holistic -- counterbalance each other. The particular rules of admining -- the so called "second mover advantage" -- mean that the worst excesses of authoritarian admins can be quickly opposed by holistic types, and pushback from holistic admins and experienced editors acts as prior restraint of overly aggressive admin actions.

Discretionary sanctions destroy that balance. Overly strict sanctions are locked in by arbcom authority, and the primary appeal venue (AE) is a harsh, authoritarian place where merely screwing up a request can get you sanctioned.

Changing the rules for swaths of the pedia is beyond the stated remit of the committee ("disputes between editors"). I understand the community has at least passively accepted DS, so I'm not making a legalistic argument; I'm saying the committee itself should seriously reconsider whether DS is appropriate given its founding mandates.

— Preceding unsigned comment added by NE Ent ( talk • contribs) 13:52, 16 March 2014‎ (UTC)

You say "overly strict sanctions are locked in by ArbCom authority". That's not right because lenient ones are locked in to. (The present draft emphasises the range of options open.) Plenty of cases end with no action being taken. On the other point, the committee's founding mandates (from 2004) explicitly envisage enforcement by administrators. Who else is there to do it?   Roger Davies ^talk 15:41, 16 March 2014 (UTC)

What is the workload? How many admins participate in the course of a year? Has that changed over time? I have a rough idea of what the committee did last year, both in terms of cases and requests (see User talk:Neotarf/ArbCom 2013), but I have never seen any numbers for AE. — Neotarf ( talk) 17:39, 16 March 2014 (UTC)

Then please compile the "numbers for AE", and publish them in your userspace. AGK [•] 23:30, 18 March 2014 (UTC)

Thank you for that, Anthony; coming from you, it means a lot. Unfortunately, I do not have the information readily at hand. The list in my user space was compiled in preparation for the Signpost's year-end Arbitration Report, so it was just a matter of a few hours effort to paste all the AR reports for the year into one document, remove the duplicates, and make an informal count. I no longer have the time required for sustained volunteer effort. You could try the editor who has succeeded me in writing the Arbitration Report feature, but that might be a long shot, since I noted that his first report did not include clarification requests.

It would be easy enough for someone to compile a list like this for each year and tuck the data into an Excel document for line graph comparisons and such. With more time, you might be able to find a way to add the data from the earlier community-based sanctions as well. You could track other data, such as how many AE admins are active on each AE request, how many cases have been closed by each admin, how many non-admins participate in AE discussions, and how many times a particular ArbCom case is invoked at AE, as well as the type of sanctions typically imposed, and whether they are imposed on individuals named in the original case decision or on participants in the discussion, and whether these trends are remaining stable from year to year. As I understand you were "present at the creation" of discretionary sanctions, and I think also the community-based sanctions they gave rise to; you would probably be able to point to other fruitful areas of inquiry, and other individuals who might be able to give input.

This is just stuff that can be done with ordinary Microsoft Office-based programs. There have been huge strides in the processing of meta-data since I took my first undergraduate stats class over 20 years ago. This strikes me as a good project for a grad student in a summer internship program, but I don't know who you could ask about it on en.wp. Someone like Erik Zachte might be able to give advice, or say if the Foundation has anything that can be used by the volunteer side of the project. (See Meet the Stats Master: Making Sense of Wikipedia's Massive Data Trove.) It also occurs to me that Oliver Keyes is based in the U.S. and has written about WP data analysis topics on his personal blog, he might be able to point to an appropriate contact person.

Statistics of course are no substitute for human intuition and judgment, but I find it sometimes helps to immerse yourself in the first before indulging in the second. — Neotarf ( talk) 04:40, 22 March 2014 (UTC)

Discretionary sanctions are authorised by the arbitration policy, which allows us to act as a final binding decision-maker primarily for serious conduct disputes the community has been unable to resolve. Although DS involves our delegating responsibility to administrators, under The Committee may … designate individuals for particular tasks or roles, it nonetheless lets us discharge our responsibility for resolving these "serious conduct disputes". Also important is that the community treats discretionary sanctions as the paragon for calming bad topic areas; cf. community sanctions on The Troubles. AGK [•] 23:30, 18 March 2014 (UTC)

This is the crux of the biscuit right here. The Arbcom is authorized to designate individuals, but who has been designated and how? As far as I can see, some people self-selected themselves, without being vetted or nominated, and certainly without community input. There is no transparent process for appointing and none for removing, or even rotating. Once in, there are no limitations, and an AE admin can become the equivalent of a J Edgar Hoover.

The Arbitration Enforcement group is now small, at this point down to perhaps 2 or 3 admins, and has a reputation for groupthink and draconian action. The group as a whole has a deep-seated antipathy towards non-admins, and is deeply prejudiced towards having the evidence and process reviewed by outside editors. While the community has a good deal of confidence in the Arbcom, and rightly so, it has none at all in the Arbitration Enforcement process. — Neotarf ( talk) 10:58, 22 March 2014 (UTC)

Could you please stop dominating this review? That comment was addressed to NE Ent (and was indented separately from my comment to you). AGK [•] 22:13, 23 March 2014 (UTC)

Quote: I'm saying the committee itself should seriously reconsider whether DS is appropriate given its founding mandates. (NE Ent)

Discretionary sanctions are in conflict with policy in the two areas of warnings and bans. It is very clear there is both a community mandate and proven track record for DS, when used for articles experiencing rapid edits, such as situations of fast-breaking public events. The Manning case immediately comes to mind. It certainly would have been easy enough to lock the article for the length of the controversy, but the community has shown that it wants to edit through such events, and the admin corps will not be able to keep up with such rapid editing if they have to stop and warn every obvious vandal.

At the opposite end of the spectrum is a case like WP:ARBATC, which I participated in as a new user on my 30th edit ever to the Wikipedia. The single most important outcome of the case was the remedy that "In particular, a rapid cycle of editing these pages to reflect one's viewpoint, then discussing the changes is disruptive and should be avoided. Instead, parties are encouraged to establish consensus on the talk page first, and then make the changes., in effect, establishing 0RR reverts for a policy page used by a variety of users, scripts, and bots, where the stability and sedateness of the page was valuable to their smooth functioning. I see no reason for placing discretionary sanctions on pages like this, and bypassing the usual system of informational warnings for users with problematic edits. — Neotarf ( talk) 10:04, 22 March 2014 (UTC)

It's been apparent for months the committee, or at least the subportion actively participating here, is not interested in an actual review of the DS process. The focus remains on "efficiency," "less disruption" and less work for the committee, not the important questions, which should be:

• Are the mainspace articles under DS of better quality than the rest of the pedia?
• Does warning / logging / sanction process increase or decrease the quality of edits contributed by the affected editors, especially as compared to the normal (non DS) intervention processes?

To the extent these questions have been addressed, it has been with intellectually shallow glib responses, (e.g. 'It's obvious'). The fact of the matter is the span of Wikpedia -- about five million articles (6,818,163) -- makes such self-centric observations of little value. (Of course, this means I don't know the answer any more than anyone else -- DS could be the greatest thing since sliced bread). What the committee should be doing is formulating actually meaningful questions about the process, determining what data is necessary to address them, and then soliciting support from the wikinerds (e.g. WP:VPT) to gather the data, and then seriously analyze the data to determine what direction to proceed. — Preceding unsigned comment added by NE Ent ( talk • contribs) 13:52, 16 March 2014‎ (UTC)

The key question here, I think, is not whether articles are better quality under DS but whether they are editable at all, in any meaningful sense, without them. When the big disputes - Scientology, Climate change, Palestine/Israel, The Troubles, India/Pakistan etc - were at their peak, editors spent little time improving articles and a great deal of time feuding.   Roger Davies ^talk 15:10, 16 March 2014 (UTC)

As someone who has been editing in the Palestine/Israel area for over 11 years, I can say that Discretionary Sanctions are the best thing that ever happened to it. Even the 1RR ruling alone is a big blessing. That's not to say everything is perfect, but very few of the good-faith editors there would choose to go back to the bad old days. Zero ^talk 11:49, 17 March 2014 (UTC)

Same goes for my main area, climate change. NewsAndEventsGuy ( talk) 15:20, 17 March 2014 (UTC)

Roger, that may be one valid question to ask about DS, but another is whether articles (and non-article processes) become abandoned, either out of fear of editing, or by sanctioning away the only editors willing to work on them. We definitely know this is happening. As just one example, almost all work on MOS:ORGANISMS has stopped. It was very near completion when DS abuse chased me off the system. The cost in articles not created, and articles not improved, by DS and abuses of it so upsetting numerous editors that they just leave, or sharply curtail their editing activities, is impossible to calculate, and simply counting them up if we could wouldn't give us a sense of the qualitative loss to the project.

There's also the community cost of the appearance that ArbCom has run off the rails, that DS has turned AE into a horrid chimera of a "Wild West" shoot-'em'-up kill zone and a Kafka-esque nightmare of harsh legalism, and that the general inescapability of whatever weird sanctions some self-important wikicop slaps on you means that Wikipedia administration is becoming increasingly arbitrary (in the negative sense) and dictatorial, like a world overrun with clones of Napoleon. It's directly attracting the wrong kinds of people to seek admin status. Yes, DS does help to reign in certain things, like race-baiting hatefests on a few perennial hot-spot topics, but it's been applied in ways that, e.g., thwart consensus building on WP guidelines, and even censure people for using WP's dispute resolution forums. I'm still being pilloried for the latter in the log at WP:ARBATC. The very idea that failing to bring a successful case at AE is going to lead to a "boomerang" sanction against someone is insane, juvenile and stupid. It's like being sent to prison if the burglar you reported gets acquitted.

Treating AE like some kind of magical kingdom with special rules and uniquely empowered paladins is having terrible effects on the entire editing community, who have taken a long time to absorb WP's already over-complicated rules and norms, only to be told these can now be thrown out the window on a whim, in favor of wildly inconsistent, punitive actions by people who in some cases focus on doing almost nothing but telling other people what they can't do here. It's also worsening the public perception of Wikipedia, and those who spend most of their time in AE simply don't see any of it, as they're lost in the fantasy. This is not a game.

I told someone the other day that had been one of the two most active Wikipedia editors on pool and billiards articles. His response was "Are you fuckin' crazy? Why you wanna waste time with that buncha psycho control freaks?". This did not come from some techno-libertarian geek who may have come here and gotten into online community politics squabbles, but from a blue-collar pool league player who probably spends less than an hour a week online. Not a good sign. WP's editor-hatefulness is already legendary even among people who have probably never considered editing [yet, but now probably not ever]. You don't need a business or communications degree to know that "widely distrusted and scorned by potential as well as actual customers" is the absolute worst public relations situation to ever be in. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 22:42, 20 March 2014 (UTC)

I get the same thing from online contacts in academia, where I have been mocked repeatedly for taking part in Wikipedia. Word of mouth travels fast in cyberspace. — Neotarf ( talk) 07:00, 22 March 2014 (UTC)

Hello. I got notice that this page is at a new stage in development. I have not been following the history of this work and do not understand the nuance of controversy around this, but upon looking lightly at it, it seems apparent to me that a lot of thought has gone into this and everything appears to be orderly and right with this. I am sorry that I am unable to give any more thoughtful comment about this but thanks to those who have worked on this. It seems useful and encouraging. Blue Rasberry (talk) 18:43, 17 March 2014 (UTC)

Thank you for sharing your first impressions. Don't worry about not having more to say :-). AGK [•] 23:31, 18 March 2014 (UTC)

• Agreed that a lot of thought and work has gone into this. I raise concerns about whether DS is the right path for WP, and have had disagreements with AGK and others working on this about extant enforcement, but commend the resolve to try to make DS work and work better. It's nice to see the level of openness here, though I think this should perhaps periodically be "advertised" at Village Pump, even WP:CENT, to broaden it further. — SMcCandlish   Talk⇒ ɖ^⊝כ^⊙þ Contrib. 22:45, 20 March 2014 (UTC)
  • I second Bluerasberry's comment. SMcCandlish, although we're wikifriends, we don't seem to be on the same songsheet in relation to what I see as a good reform. I can only ask that the committee get on with it, removing any personal, punitive site to the templates: it should always have been about informing editors, so let's belatedly get it right. Thanks. Tony (talk) 07:50, 21 March 2014 (UTC)
    • What you are losing sight of, Tony, is that the purpose of these notifications is to get rid of editors. ^{citation needed} The current criteria for getting rid of editors says you choose editors who have some problem with their editing and inform them of the problem first. The new criteria for notifications is non-existent. Someone can get notified and vaporized for the wrong reason, or for no reason at all. — Neotarf ( talk) 08:47, 22 March 2014 (UTC)
• @SMcCandlish, I've never heard of those noticeboards before, why don't you post something? — Neotarf ( talk) 07:45, 22 March 2014 (UTC)

Arbitration Enforcement is conducted at a designated noticeboard, currently Wikipedia:Arbitration/Requests/Enforcement

At that noticeboard, discussions concerning sanctions against any editors concerning any area listed as being under "Discretionary Sanctions" by the Arbitration Committee are held. That Committee has the power to amend or remove any area described as subject to sanctions through that noticeboard.

Any misbehavior by an editor working within an area discussed at the noticeboard may be dealt with by sanction after discussion, and who has been made reasonably aware that the area is so listed. To that end, any editor may notify such an editor that the area is under discretionary sanctions.

Administrators reasonably considered as either involved with the area of dispute or with the editor being discussed may not participate as administrators in determining sanctions. In addition, administrators may not impose undue or disproportionate sanctions, but shall only impose such sanctions as would be reasonably anticipated by the Arbitration Committee. If a sanction is appealed, that administrator imposing the sanction may not also act as an administrator in any such appeal.

All sanctions shall be logged on the appropriate page or pages designated by the Arbitration Committee in any decision regarding the area under discretionary sanctions. No sanction shall be altered without the consent of the administrator originally imposing the sanction. If such an administrator is no longer an administrator or is inactive, such appeals shall be made to the Arbitration Committee and not to the Arbitration Enforcement noticeboard.

Prior sanctions made under discretionary sanctions are not affected by this revision.

Which I think is shorter, clearer, and pretty simple except for the required buzzwords. I do remove the "only the editor under sanction" bit from the appeals process because I fail to see why it is either useful or needed here. It is still unreadable, but far less unreadable than the first proposal. Cheers. Collect ( talk) 13:04, 22 March 2014 (UTC)

It might take less time to read, but it would consume more time in practice as people argue about various things that were edited out. In addition it doesn't destigmatize the current ambiguous warnings. NewsAndEventsGuy ( talk) 14:16, 22 March 2014 (UTC)

As it eliminates the "warnings" usage, I suggest that it does address that issue pretty clearly -- it refers to notifying people, which I suggest is the best practice here. And in any case, I suggest that those who argue will not be dissuaded by the longer version at all <g>. Collect ( talk) 14:27, 22 March 2014 (UTC)

If Draft 3 is implemented, along with its FYI-alert system, at least one and probably more than one editor in each area of active conflict will make sure that everyone who stops by gets the friendly "did you know" alert, and so the stigma will simply be washed away by ubiquity. NewsAndEventsGuy ( talk) 15:37, 22 March 2014 (UTC)

The "misbehavior" language pretty much sums up the dismissive attitude of way too many admins towards non-admins, that non-adins are naughty children who are not to be taken seriously. If they were talking about admins or arbs, it would be "conduct". — Neotarf ( talk) 14:58, 22 March 2014 (UTC)

I reacted negatively to that word choice also. NewsAndEventsGuy ( talk) 15:45, 22 March 2014 (UTC)

On this rare occasion, I agree that "misbehaviour" really really rankles. Every attempt should be made to minimise the belittling effect that editors all too easily experience WRT sanctions and other measures taken in relation to their activities. Tony (talk) 23:26, 23 March 2014 (UTC)

Perhaps "acts specifically contrary to Wikipedia policies and guidelines, or the sanctions placed by the arbitration Committee" would make "misbehavior" a far less utile word here? Collect ( talk) 22:22, 22 March 2014 (UTC)

That makes the fatal assumption that negative conduct isn't misbehaviour. We should not pretend that disruptive conduct isn't disruptive. AGK [•] 23:16, 27 March 2014 (UTC)

IOW you would find that being "negative" without violating any policies or guidelines should still be covered by sanctions? Example please -- I suspect that many would demur on such a broad extension of AE. And tautological arguments do not count -- show me an example of "disruptive conduct" where no policies or guidelines have been violated, and where the community would seek sanctions. Collect ( talk) 23:32, 27 March 2014 (UTC)

Wikipedia:Disruptive editing is itself a guideline and "This page in a nutshell" says "Disruptive editors may be blocked or banned indefinitely". It discusses disruptive editing at some length, with numerous examples. There's nothing new or strange in admins dealing with it,   Roger Davies ^talk 00:17, 28 March 2014 (UTC)

Which would be covered under the wording suggested. The point I demur on is the concept of "negative conduct" which is not found in any policy or guideline I have found, and for which I asked for an example.. That you affirm that the guideline exists for "disruptive editing" says not thing whatsoever about "negative conduct" which I fear is an amorphous and meaningless term to bandy. So I iterate that request -- as "disruptive editing" is barred by guideline, what precisely is "negative conduct" supposed to mean? Cheers. Collect ( talk) 13:18, 28 March 2014 (UTC)

As it's not in the DSR draft does it really matter?   Roger Davies ^talk 13:37, 28 March 2014 (UTC)

Yes it does. This section is not about the "DSR draft". This section is about concise version not broken up into single sentence "sections" which is what is being discussed here in this section. One editor stated that "negative conduct" is "disruptive editing" and I quite cordially asked for an example of "negative conduct" which did not already fall into "disruptive editing" but which should, in his opinion, be a bannable or blockable offense. Absent such an expository comment, I fear that the concept of "negative conduct" is an amorphous catch-all which, being undefined, could be used to cover anything at all including simply posting, say, alternative wording for a policy or guideline on a talk page, or the like. Cheers. Collect ( talk) 15:34, 28 March 2014 (UTC)

Is there any appetite among AE admins to try to compile a short document detailing the best practices currently followed at AE? I think that such a page, explaining to unfamiliar users how things actually work at arbitration enforcement, might be useful both for users (being reported or reporting others) and for previously uninvolved admins who wanted to give AE a shot. Another thing I think might be useful would be compiling some sort of sentencing guidelines; these would have to be merely advisory in nature (so as not to excessively limit admin discretion), but, in my opinion, would be useful both to users and to enforcing admins... Salvio ^{Let's talk about it!} 15:29, 23 March 2014 (UTC)

This non-admin thinks that's a great idea, but how about having some fun also with a hit list of common (comedic) flub ups? NewsAndEventsGuy ( talk) 15:42, 23 March 2014 (UTC)

Would it be useful to compile a list of prior cases by result. So that rather than writing a treatise on how to implement DS anyone can see what has resulted in topic bans, 1 year blocks, warnings etc, successful appeals. This has a dual purpose in giving the committee actual stats on AE and giving new sysops a context for decisions. This might also be faster/easier to compile-- Cailil ^talk 16:16, 25 March 2014 (UTC)

It would be relatively easy to extract the results of say the last three-years worth of cases by running scripts on Collapsed headers in the archive pages. This could give us something like:

Date	Archive/Case	Issue	Outcome
2012-09	[link to archive#Case name]	Broke 1RR (one-off)	Warned.
2012-10	[link to archive#Case name]	Edit-warring (3RR)	24-hour block
2013-01	[link to archive#Case name]	Personal attacks	Topic-banned

It would be even better if it were sortable. But with the raw data, it would be easy to produce ranges of sanctions tied to specific issues.   Roger Davies ^talk 20:11, 25 March 2014 (UTC)

It'd be good to capture if outcomes were appealed too & whether the appeals were successful, but I'm guessing that'd have to be done manually?-- Cailil ^talk 15:55, 26 March 2014 (UTC)

It might be possible to get most of them (if they were at AE), though another dsta route is via the sanction logs on the actual case pages, which would give you appeal entries too. One thing that would need doing is standardising the "Concerns" into categories for consistency but that's easy once the data are assembled.   Roger Davies ^talk 16:32, 26 March 2014 (UTC)

Yeah - that'd be good. It'd also be useful to capture the rare occurrences when the appeals go beyond AE to AN or ArbCom - just to show if the Committee & community have upheld the appeal or not. This would give the reader a good sense of how well founded such actions are or not. I've been involved in one or two and if everyone who was involved added them manually that'd reduce the labour involved in search through archives etc-- Cailil ^talk 12:31, 27 March 2014 (UTC)

Salvio: sounds like a really good idea. Tony (talk) 13:14, 27 March 2014 (UTC)

A proposal

An excellent idea, Salvio. Ironically, this was the subject of an AE request to ArbCom last year about this time.

While I deeply appreciate the effort and expertise that AGK and Roger Davies have brought to this review, I would like to see more arbs take a more hands-on approach to this question, and volunteer to spend a little time in the trenches at AE before a final version is presented. Governance is too important a topic to leave for a few individuals--these questions deserve broad attention and many eyes. I would also suggest that the Committee make a short list of experienced admins who are skilled in communications to assist them in this evaluation.

As a starting point, I see that in April 2013 there were 12 AE requests. Fourteen arbs, working in groups of 4-5 individuals per case, could do 3 requests per month each, or even less if they had a core group of admins to assist with evaluation and with the formulation of guidelines for users and new admins. This could begin within the week, so as to be finished before the summer vacation begins. While I am usually a fan of transparency, in this case offline communications would enable individuals to speak more freely and iron out any details more quickly.

This review has gone on for a good six months; most RFC's see little participation after the initial two weeks. The community has given what attention and advice it is capable of giving; it is now up to the ArbCom to solve these issues that the community has been unable to solve. It's what we elected you for. Good luck, to all of us. — Neotarf ( talk) 12:31, 1 April 2014 (UTC)

This has not been a whole block of discussion, because discussion has been repeatedly re-opened about points refined from before. Thus the RFC analogy does not work. Additionally, as you must realise, this is not about governance but about preventing disruption to the encyclopedia. As has been repeatedly pointed out, you perhaps do not realise (or refuse to recognise) this, because you view discretionary sanctions as a person who has been the subject of enforcement proceedings. (After all, you have retired from editing and only contribute in order to change or dismantle DS.) How are readers of your comments about DS, on this page and in many other venues, supposed to remember that? AGK [•] 21:48, 1 April 2014 (UTC)

Well this non-admin reg editor quit reading Neotarf a long time ago. NewsAndEventsGuy ( talk) 22:44, 1 April 2014 (UTC)

If you don't believe me, Anthony, check the page view statistics. I gave this review several mentions in the Signpost's Arbitration Report, and as a result, there were additional editors drawn into the discussion, but there have been very few editors interested in sustained participation. And I have been "the subject of enforcement proceedings"? Have you got me mixed up with someone else? Perhaps you are thinking of the AE case filed against Noetica by a vindictive and relentless user who has since been indeffed. I have no problem with being remembered for having tried to defend an editor of Noetica's caliber, or for the very small part I played in trying to stop this disruption. Wikipedia needs to retain quality editors, not alienate them by enabling their accusers. And I find the accusation that I have a some motive to "change or dismantle DS" bizarre and offensive. Please do not make up false motivations for me, or attribute to me views I do not hold. I find your continued expressions of hostility inexplicable and distressing. If you want to accuse me of bad faith, RFCU/Neotarf is still a redlink. Get some diffs and start filling it in. — Neotarf ( talk) 08:15, 6 May 2014 (UTC)