Wikipedia:2017 ANI reform RfC

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Editors commonly complain about the incidents section of the administrators' noticeboard. It is popularly perceived as a drama center, as a pile-on kangaroo court, as chaotic and disorderly, and as a place where vicious personal attacks can be made with impunity. This can be easily confirmed by just looking at the plethora of redirects to ANI: it is derisively referred to with the terms "cesspit," "drama board," "pitchforks," "happyplace," "Great Dismal Swamp," "airing of grievances," "popcorn," "Slough of Despond," "wikihell," and "Complaints Department."

ANI does serve a necessary function, but the vast majority of editors would probably agree, in principle, that ANI has some issues. However, there has never been any substantive effort to address those issues. This RfC is intended to change that.

Be aware that this RfC is mostly conceptual. The proposals in this RfC are, by design, generally phrased; the object is to get an overall consensus for broad ideas, so as to avoid arguments over petty details. Except for Proposal C, the proposals cannot be implemented immediately, as such. If a proposal passes, further discussions will be held to hammer out the details. 18:00, 5 August 2017 (UTC)

Proposal A: ANI template

To bring some degree of order to ANI, create a template (loaded and used by default) to organize ANI discussions. The template would have separate sections for:

1. The statements of the parties. The filing editor would name the parties upon initiating the thread.
2. The observations of, and discussions among, uninvolved administrators. It is, after all, called the administrators' noticeboard.
3. The general comments of other, non-party editors.

There would also be a "Proposals" section, in which sanction proposals could be initiated and voted upon. However, this section would be commented out and invisible by default. If an editor proposed a sanction, they would simply remove the comment wikitext. Template:TOC limit could be used to avoid bloating the table of contents.

See an example template here.

Support

1. Support. The chaotic, sprawling threads at ANI are a very major part of the problem. If there is not any order, there will not be any civility. Biblio (talk) 19:20, 5 August 2017 (UTC)
2. Support #1, but oppose the other two, as ANI has grown more into a place that all editors can express their complaints, and not be confused by whether they are allowed to post in the administrators section to reply to something and stuff like that. RileyBugz^会話_投稿記録 20:51, 5 August 2017 (UTC)
3. Support but do not mandate use of template per concerns raised by oppose voters. Would make most discussions much more readable for all participants, encouraging more editors who aren't necessarily drama seekers to try to diffuse situations. This works for WP:ARC, why not ANI? feminist 16:06, 6 August 2017 (UTC)

Oppose

1. Oppose. ANI gets lots of different types of complaints, forcing a process on it will make the underlying problems addressed on that page worse, even if it makes ANI itself better. Power~enwiki (talk) 19:47, 5 August 2017 (UTC)
2. Oppose. A template will not help much, but it will make it difficult to adapt to the huge range of issues brought to ANI. I especially oppose separating admin from non-admin comments; all editors are on equal footing at ANI. ~ Rob₁₃^Talk 21:22, 5 August 2017 (UTC)
3. Oppose ANI gets all sort of matters that require administrator involvement and/or blocks, some are complex and some are simple. Forcing/defaulting a template on the quick issues is overkill. As for creating a template that can be used as an option for refactoring the longer discussions, I agree with BU Rob13 that separating admin/non-admin comments is not a good idea. – Train2104 (t • c) 01:18, 6 August 2017 (UTC)
4. Oppose Formatting of complaints on AN/I isn't really a big problem. Templates are most useful for very formal processes like AfD, where there's some machine processing of the votes. John Nagle (talk) 05:12, 6 August 2017 (UTC)
5. Weak oppose. I'm sympathetic to the notion underlying this idea, and templates are certainly the least invasive of the proposals in this RfC, but the span of issues makes it unlikely that a single template will allow for the flexibility necesary for a filing. Maybe if the template were incredibly simple, with very general fields for involved parties, summary of events, supplied evidence, and proposals, I could get behind one, but I'm skeptical of lending any support to the proposal in these broad terms. Snow ^{let's rap} 05:35, 6 August 2017 (UTC)
6. Oppose. As Winston Churchill once said about democracy, it's the worst possible political system, except for all the others. AN/I (and to a much lesser extent AN) are a bit like democracy: messy and sometimes chaotic, but in the end the job gets done. The "drama board" aspect of AN/I and AN is highly overstated, and the boards are surprisingly efficient at sanctioning those who need to be sanctioned. In fact, the biggest problem at AN/I is that not enough admins patrol it and implement community decisions, meaning that discussions go on a lot longer then they need to, with unnecessary pile-on !votes before a sanction is put into effect. I don't believe that any of these suggestions are workable "solutions", since I don't think that the problem is nearly as bad as is presented here. Beyond My Ken (talk) 06:00, 6 August 2017 (UTC)
7. Oppose This is structured in the belief that every ANI post is some sort of unsolvable dispute. No, the last two posts were me reporting on general issues which would not require this. ^{My name is}not^dave (talk/contribs) 08:17, 6 August 2017 (UTC)
8. Oppose. This presupposes that every ANI post is about some kind of inter-editor dispute. This is not the case. ‑ Iridescent 10:06, 6 August 2017 (UTC)
9. Oppose. Too many types of disputes for this straitjacket, and administrator views are not privileged in ANI discussions. The Big Bad Wolfowitz (aka Hullaballoo). Treated like dirt by many administrators since 2006. (talk) 12:30, 6 August 2017 (UTC)
10. Oppose I particularly dislike the idea of separating admins and non-admins. Lepricavark (talk) 21:05, 6 August 2017 (UTC)
11. Oppose – ANI does not need to be, and should not be, similar to formal mediation or arbitration. The free form format also permits to adapt to any kind of situation. Participation should be as easy as possible including for new editors. Hmm we also should nominate some of those redirects for discussion... —PaleoNeonate – 20:29, 7 August 2017 (UTC)
12. Oppose - Biblioworm doesn't get it. The problem at ANI is the editors. This tilt is a waste of time. Chris Troutman (talk) 23:33, 7 August 2017 (UTC)
13. Oppose I don't see this helping ANI at all. We are also separating admins and long-standing editors, making it seem non-admins are less important. Callmemirela 🍁 ^talk 00:02, 8 August 2017 (UTC)
14. Oppose this proposal seems to not understand ANI. Not every issue raised at ANI is an inter-editor dispute, and the main issue that makes ANI so toxic is that it is often a last resort (or bad first resort) for troublemakers or users at their wits end with a troublemaker. I don't agree with separating admins and non-admins this way. — InsertCleverPhraseHere 06:03, 8 August 2017 (UTC)
15. Oppose No idea what this seeks to implement, although it does seem to have a good intent. Inherent oppose here because ANI is meant for discrete diverse topics and templating will (or might) hinder that, for sure. --QEDK (愛) 18:57, 8 August 2017 (UTC)
16. Oppose – We already have formal boards such as AE and Mediation. The ANI "free-for-all" provides a level-playing field between newbies and experienced editors. The format of the complaints and subsequent discussions also helps assess the merits of the claims and the behaviour of participants. Let them vent! — JFG ^talk 20:48, 8 August 2017 (UTC)
17. Oppose - The informal discussion format at ANI is not and has never been a problem; ANI is a flexible catchall for issues that aren't solvable by existing formal processes such as AN3 or SPI. Swarm ♠ 19:04, 9 August 2017 (UTC)
18. Oppose join the pile-on: a lot of ANI threads are simple issues like legal threats that just need urgent attention but need nowhere near the trappings of SPI or an ArbCom case. This would ruin the only thing ANI is actually good at: the simple quick fixes for disruption. TonyBallioni (talk) 20:43, 9 August 2017 (UTC)
19. Oppose Segregating out discussion based off user right will not solve anything. One can find out if a commenter is an admin in a matter of seconds be enabling hoverlink. Much better to remove the "steer clear of ANI" form the Official New Admin Manual. L3X1 (distænt write) )evidence( 13:52, 10 August 2017 (UTC)

Proposal B1: Word limit for posts

Limit the number of words that any given editor could post, on a per-comment basis. There would be separate word limits for different categories of editors, with parties and uninvolved administrators having a higher word limit than general commenters. The object of this proposal is to discourage lengthy rants (i.e., "walls of text"), and force editors to exercise restraint.

If Proposal B2 fails, such failure will not affect the status of this proposal. If this proposal and Proposal B2 are each successful, both proposals will be eventually implemented, following further discussion of the details.

Support

1. Support. A word limit would prevent editors from posting massive, unhinged, "too long, didn't read" rants and walls of text. It would force commenters to limit and restrain themselves to some degree, and to think about what they're writing. Biblio (talk) 19:20, 5 August 2017 (UTC)
2. Strong support – The 500-word limit at WP:AE has proven effective in stamping out wild rants and fishing expeditions. If an editor thinks they need more space, it generally means their complaint is too vague, too broad or too aggressive. — JFG ^talk 20:48, 8 August 2017 (UTC)

Oppose

1. Oppose. This prevents us from dealing with complex issues that require explanations. Besides, who would enforce this? ANI clerks - more bureaucracy? Also, any per comment limit is very gameable. ~ Rob₁₃^Talk 21:25, 5 August 2017 (UTC)
2. Oppose. While I understand the motivation for this IMO ANI reports just can't be reduced to "One size fits all." For one thing (as BU Rob13 points out) evidence is required for any action to be taken on a given report. In complex cases that could require more than even a generous word count might allow. I would also note that WP:TLDR exists for those that get long winded and it has occasionally caused an editor to make their report more concise. MarnetteD|Talk 21:40, 5 August 2017 (UTC)
3. Oppose. Far too unwieldy a proposition. Issues at ANI may address a wide scope of potential issues, ranging from brief and petty battleground exchanges that can be adequately captured with a handful of diffs, to sprawling issues of vandalism, socking, and many other forms of abuse that (not altogether unheard of) can have gone unnoticed for years and may require masses of links and exposition to detail. And many types of behavioural issues require filings of lengths inbetween. No, I'm sorry, not only is this a WP:SNOW call for me, but (meaning no disrespect) I honestly wonder how much experience a person could have of the forum to even propose this as something workable. And frankly, when a filer posts too much content, relative to the disruption, that's actually useful to the community, as it can be red flag as to the OP's sense of perspective on the matter. Snow ^{let's rap} 05:17, 6 August 2017 (UTC)
4. Oppose. Posting a wall of text on AN/I seldom helps the party who does so. See [1] for a famous case. John Nagle (talk) 05:18, 6 August 2017 (UTC)

   Heck, you only need to go back a couple of days to find a contributor who nearly got themselves WP:BOOMERANG banned (despite their position not being entirely baseless) for failing to keep their posting tight, reasonably-sized and well-organized: [2]. Snow ^{let's rap} 06:23, 6 August 2017 (UTC)

5. Oppose - See my general comment in the first section. Beyond My Ken (talk) 06:00, 6 August 2017 (UTC)
6. Oppose Not practical. ^{My name is}not^dave (talk/contribs) 08:15, 6 August 2017 (UTC)
7. Absolutely not. This doesn't work at Arbcom, and wouldn't work here. Some situations need complex explanations. ‑ Iridescent 10:07, 6 August 2017 (UTC)

   Word limit bad, gameable. Telegraphese annoying. Beyond My Ken (talk) 13:45, 8 August 2017 (UTC)

8. Oppose. Presumes that ANI readers are unable to distinguish between worthwhile analysis of more complex situations and longwinded blather. If that's the case, the whole process should be scrapped. The Big Bad Wolfowitz (aka Hullaballoo). Treated like dirt by many administrators since 2006. (talk) 12:33, 6 August 2017 (UTC)
9. Oppose because troublemakers would be deterred by your word limit. Chris Troutman (talk) 23:34, 7 August 2017 (UTC)

   +"Not"? Beyond My Ken (talk) 13:47, 8 August 2017 (UTC)

10. Oppose If something has to be on the table, let it be on the table. TLDR situations are the risk that one accepts when posting walls of text. Samsara 05:07, 8 August 2017 (UTC)
11. Oppose Walls of text are almost always ignored and only assessed by the conflicting (and/or interested) parties + closer. But as stated, there is no point to limiting them with an arbitrary word count; it will either be gamed and might hinder constructive long assessments (if you're going to add exceptions, that also essentially breaks the purpose and making the basis futile). --QEDK (愛) 19:00, 8 August 2017 (UTC)
12. Oppose per BMK. Swarm ♠ 19:07, 9 August 2017 (UTC)
13. Oppose Boomerang solves the problem of tl;dr is the words aren't warranted and are just throwing things at the wall. TonyBallioni (talk) 20:43, 9 August 2017 (UTC)
14. Oppose If you want to keep ANI at reasonable text amount, go through and clear out all the "trivial" (not requiring a large amount of input from the community) threads withing a few (8-18) hours of completion, rather than wait 3 days for a bot to remove a 2-editor thread request for rev-del. L3X1 (distænt write) )evidence( 13:54, 10 August 2017 (UTC)
15. Oppose - Cases that go to AN/I tend to be complex and evidence, which would add to the word count, is usually needed in order to make a convincing argument. Simpler cases probably don't need to come to AN/I anyway. PCHS-NJROTC ^{(Messages)Have a blessed day.} 00:59, 11 August 2017 (UTC)

Proposal B2: Word limit for threads

Limit the number of words that any given editor could post, on an overall, per-thread basis. There would be separate word limits for different categories of editors, with parties and uninvolved administrators having a higher word limit than general commenters. The object of this proposal is to discourage lengthy rants (i.e., "walls of text"), and force editors to exercise restraint.

If Proposal B1 fails, such failure will not affect the status of this proposal. If this proposal and Proposal B1 are each successful, both proposals will be eventually implemented, following further discussion of the details.

Support

1. Support. See my support vote for Proposal B1. As for this particular proposal, it would greatly compliment B1 and make it even more robust. In addition to prohibiting walls of text, ANI threads themselves would no longer be sprawling, unrestrained, unreadable, and unmanageable. Biblio (talk) 19:20, 5 August 2017 (UTC)

Oppose

1. Oppose per my comment to B1. Adds bureaucracy and reduces adaptability. ~ Rob₁₃^Talk 21:33, 5 August 2017 (UTC)
2. Oppose for many of the same reasons as mentioned in my comment to B1. This also has the potential to affect editors where English is their second language. Who is going to enforce this - is someone going to have the temerity to remove posts that exceed the limit - deciding which ones to go is likely to cause more threads at ANI about their removal. MarnetteD|Talk 21:46, 5 August 2017 (UTC)
3. Oppose This would effectively prevent a response to raised issues or (one hopes rarely) personal attacks. Martin of Sheffield (talk) 23:03, 5 August 2017 (UTC)
4. Strongest possible oppose. Per my reasoning on the above proposal, but viewing this one as even more insanely infeasible. This proposal would not only deeply restrain the ability of the community to explore and remedy issues, it would give perverse incentives to not fully divulge/elucidate on issues that should be in the record, for fear of needing reserves to discuss later. It could even be gamed by people wishing to hide details they know the community currently expects them to disclose in good faith, by being able to later appeal to this rule that they needed to keep their comments short. I also do not like this bonus probelamtic notion of tiering the amount of contributions parties can make, according to their status in the community. That runs well afoul of numerous of our policies on how consensus is formed here and the general equality of voices. All told, frankly this looks like nothing that has ever been accepted by the community for any space, let alone one of the most important, and the one that lets us address our most problematic and persistent problems. To adopt this approach would frankly be nothing short of the single greatest act of shooting ourselves in the foot I have ever seen in a policy discussion. Snow ^{let's rap} 05:26, 6 August 2017 (UTC)
5. Oppose - See my general comment in the first section. Beyond My Ken (talk) 06:01, 6 August 2017 (UTC)
6. Oppose This will be deeply impractical. ^{My name is}not^dave (talk/contribs) 08:15, 6 August 2017 (UTC)
7. Oppose as totally impractical. Some things need explaining in detail, especially given that ANI when being used for dispute resolution has the explicit purpose of inviting comment from people who aren't familiar with the underlying issues. ‑ Iridescent 10:09, 6 August 2017 (UTC)
8. Oppose. Utterly misguided. The Big Bad Wolfowitz (aka Hullaballoo). Treated like dirt by many administrators since 2006. (talk) 12:35, 6 August 2017 (UTC)
9. Oppose and we're going to have editors police this? Chris Troutman (talk) 23:35, 7 August 2017 (UTC)
10. Oppose I am in no way going to enforce myself or other users to respect a word limit. Some things require more explanation. A word limit will certainly not solve any ANI complaints. Callmemirela 🍁 ^talk 23:59, 7 August 2017 (UTC)
11. Oppose as noted above, this is impractical. Lepricavark (talk) 02:04, 8 August 2017 (UTC)
12. Oppose – This would likely be cumbersome to enforce. The concern is addressed by limits per post per B1. — JFG ^talk 20:48, 8 August 2017 (UTC)
13. Oppose Snow and Iridescent. L3X1 (distænt write) )evidence( 13:56, 10 August 2017 (UTC)
14. Oppose - For the same reason I oppose word limits for posts. PCHS-NJROTC ^{(Messages)Have a blessed day.} 01:02, 11 August 2017 (UTC)

Proposal C: Waiting period

This proposal is intended to address the "pile-on" issue. When any type of sanction against any editor is proposed, voting on that proposal shall not commence for 24 hours, to allow the accused editor a fair opportunity to post a statement in their own defense.

If the editor is blocked and unable to post at ANI, that editor will be allowed to make their statement via an alternative means (such as on their talk page). They must do so within 24 hours. If they do so, another editor must then post the statement at ANI on their behalf. Provided that the blocked editor makes use of an alternative means within 24 hours, voting on the proposed sanction shall not commence until someone posts the statement at ANI on that editor's behalf.

Should it pass, this proposal can be implemented immediately.

Support

1. Support. It is time for some due process. By giving accused editors some time to actually defend themselves, this policy would be a major step toward addressing the "pile-on kangaroo court" perception of ANI. Currently, some users seem to be almost enthusiastic about slapping the "banned" template on someone's user page. I suppose it makes them feel powerful and important, at the expense of the unfortunate recipient. Biblio (talk) 19:20, 5 August 2017 (UTC)
2. Support - while I certainly don't know the situation there too well, the idea here is one which should generally apply: A user can't be penalized for his/her apparent wrongdoings without a decent chance to defend himself/herself against the accusations. עוד מישהו Od Mishehu 03:32, 6 August 2017 (UTC)
3. Support Apart from anything else, often the proposals contain false statements, and misrepresented diffs. As it is now, then a situation can arise where voters have often already cast their votes based on a misleading statement of the situation before the accused has a chance to respond, or even knows that the ANI is underway. They then come to the action, seeing a long discussion all based on things they know are false. I think more is needed than this measure, but it's a starting point. Better would be to apply this to discussion as well as voting. That nobody can comment on the accusation until the person accused has a chance to respond, so debate only starts after you have both in place - the accusation and the response. As it is now, the accuser has a head start on the process and can swing opinions with a wave of support before the accused gets to the discussion. The 24 hour period also would not work for anyone who is on a holiday from editing wikipedia, or has to travel or has a trip for work that is incompatible with taking part in wikipedia discussions. But it's a start. Robert Walker (talk) 17:50, 7 August 2017 (UTC)
4. Support – Elementary courtesy of due process. There are too many knee-jerk reactions in some of those threads. — JFG ^talk 20:48, 8 August 2017 (UTC)

Oppose

1. Oppose. I don't support a waiting period on !voting. I would support a waiting period on closes. ~ Rob₁₃^Talk 21:40, 5 August 2017 (UTC)
2. Oppose. I can't see any rationale for forestalling community proposals that may speedily remedy issues. Again, a large variety of issues wind up at ANI, some of which can be readily addressed and where the interests of consensus and forestalling further disruption would only be hurt by this proposal. Like Rob, however, I may be more amenable to a short waiting period on closes. Though frankly, getting closes implemented too fast is rather the opposite of the usual issue in the space. Snow ^{let's rap} 05:31, 6 August 2017 (UTC)

   @Snow Rise: You'd actually be surprised. It swings both ways. There is a rare but significant "lynch mob" mentality, especially when one of the unblockables feels wronged. See, for example, Arthur Rubin's absurdist community ban. ~ Rob₁₃^Talk 17:13, 6 August 2017 (UTC)

   Oh yes, I am aware; note I was actually the sole oppose !vote in the above mentioned discussion, and not because I was completely convinced that Arthur had acted appropriately, but very specifically because I thought the process was moving too fast and the proposals seemed less than pragmatic as preventative measures. So point taken. Even so, this is definitely the less common problem, when compared against the large number of threads that may have been resolved simply with just a little administrative attention, but go on to be entrenched nightmares because the parties and passersby are left to go around in circles for too long, due to short staffing. Snow ^{let's rap} 20:19, 6 August 2017 (UTC)

3. Oppose - See my general comment in the first section. Beyond My Ken (talk) 06:01, 6 August 2017 (UTC)
4. Oppose as written. I'd strongly support a ban on closing any thread early except in exceptional circumstances and a total ban on early archiving, but that's not what's being proposed here. ‑ Iridescent 10:11, 6 August 2017 (UTC)
5. Oppose as written. Sometimes rapid action is appropriate. The Big Bad Wolfowitz (aka Hullaballoo). Treated like dirt by many administrators since 2006. (talk) 13:33, 6 August 2017 (UTC)
6. Oppose - Would make resolving disputes much slower, which would make threads longer, which I think that we agree is something that is not needed. It would also make ANI much more formal, which is not really needed. In addition, some issues need to be resolved quickly. Finally, what do you define as a proposed sanction? A comment on how it could possible be resolved, and then comments that support that resolution? Too much bureaucracy. RileyBugz^会話_投稿記録 17:07, 6 August 2017 (UTC)
7. Oppose like some of the others above, I'd be more inclined to support a ban on fast closes, but I can't support a ban on !voting. Lepricavark (talk) 00:30, 7 August 2017 (UTC)
8. Oppose The pile-on this the best thing on Wikipedia. Chris Troutman (talk) 23:35, 7 August 2017 (UTC)
9. Strong oppose per my oppose above. We will never get anywhere with a limit of, say, 2000 words. Callmemirela 🍁 ^talk 00:00, 8 August 2017 (UTC)
10. Oppose - The usual consensus building process does not appear to be broken. Also, this seems unenforceable. People would still state their opinions during the waiting period, and those opinions are what form the basis of consensus. We don't hold "votes" and never have. Swarm ♠ 19:12, 9 August 2017 (UTC)
11. Oppose Even though I support some early closures.L3X1 (distænt write) )evidence( 14:27, 10 August 2017 (UTC)
12. Oppose I agree with the above opposers that a waiting period on closes makes more sense than a waiting period on voting. This sort of restriction will just result in more 6-page tennis matches in which the same points are repeated ad nauseam. ZettaComposer (talk) 15:08, 10 August 2017 (UTC)

Proposal D: Clerks

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Create the position of "ANI clerk." ANI clerks would have functions similar to ArbCom clerks. They could address formatting issues, hat off-topic discussions, redact personal attacks, etc.

Support

1. ...

Oppose

1. Oppose. Absolutely no. More bureaucracy and administrative overhead is never the solution. ~ Rob₁₃^Talk 21:36, 5 August 2017 (UTC)
2. Oppose Anybody can deal with formatting issues. Anyone can hat off topic discussions. Redacting personal attacks is controversial for ANYone. Sorry, but don't see a need. --Hammersoft (talk) 22:36, 5 August 2017 (UTC)
3. Oppose As Rob says. WP is prone to too much micro-management in the first place. Martin of Sheffield (talk) 23:07, 5 August 2017 (UTC)
4. Oppose Too much bureaucracy. – Train2104 (t • c) 01:20, 6 August 2017 (UTC)
5. Oppose too much paperwork. — xaosflux ^Talk 04:13, 6 August 2017 (UTC)
6. Weak Oppose I'm more sympathetic to this proposal than the strictly procedural ones above, but I still think this is unlikely to streamline ANI or address the issues which are supposedly to be addressed by this RfC (which are, in any event, "overstated" if I am to try to pick a generous term). I don't buy the "Buro" argument of some of my volunteer colleagues above, but I do think that create extra (and mostly needless) formatting work for our volunteers in that space, when we already struggle to resolve and close discussions in a timely manner as is, is no particularly strong solution to anything. Snow ^{let's rap} 05:40, 6 August 2017 (UTC)
7. Oppose - See my general comment in the first section. Beyond My Ken (talk) 06:01, 6 August 2017 (UTC)
8. Good god no. If anything we should be deprecating "clerk" positions in those places like Arbcom that do have them—I reject on principle the "we're too important to do the routine maintenance, let the servants do it" mentality. The last thing we need is yet another tier of self-appointed busybodies. ‑ Iridescent 10:13, 6 August 2017 (UTC)
9. Oppose. Given all this apparatus, is this set of proposals very different from "send all disputes to ArbComm"? The Big Bad Wolfowitz (aka Hullaballoo). Treated like dirt by many administrators since 2006. (talk) 14:42, 6 August 2017 (UTC)
10. Oppose - Too much bureaucracy for this to run well. It would just slow down ANI. RileyBugz^会話_投稿記録 17:01, 6 August 2017 (UTC)
11. Jesus fucking christ no. Per Iridescent: I already thought clerks in ArbCom were a stupid invention anyway. FACE WITH TEARS OF JOY ^[u+1F602] 19:39, 7 August 2017 (UTC)
12. Oppose – Per same rationale in my oppose post at #Proposal A: ANI template. I also think that the requirements for clerks (timewise) would be unrealistic, as unlike for arbitration, cases don't typically last weeks or months, they often require immediate ad-hoc attention. —PaleoNeonate – 20:32, 7 August 2017 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Proposal E: Edit filter

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Create an edit filter for ANI. The edit filter would (1) identify keywords possibly indicative of a personal attack, (2) at least warn the author of that comment before they post it, and (3) tag the comment if it is posted anyway.

Support

1. Support. We'll never fix ANI until we directly get to the true heart of the problem: the incivility and insults. This edit filter would ensure that those insults don't go unseen, and it would force editors to think twice before personally attacking someone. It's really just common sense. Biblio (talk) 19:20, 5 August 2017 (UTC)

Oppose

1. Oppose. I suggest from a practical point of view that such a filter is beyond design. It's been mooted before for project and talk pages without much success (eg filters 219, 1, and 834). Personal attacks and insults, especially those which could be tested by filter, do not go unnoticed at ANI. There's some work with edit filters going on at Wikipedia:Community health initiative - I'm still not hopeful, but the results of their work, when known, would be a minimum requirement. -- zzuuzz ^(talk) 20:17, 5 August 2017 (UTC)
2. Oppose as impractical. Based on my experience as an edit filter manager, such a filter would often get it wrong (both from false positives and false negatives), be relatively expensive on a per edit basis, and not solve the issue at hand. Blatant personal attacks are not an issue at ANI because they're quickly noticed and responded to. It's the minor incivility and rudeness that's an issue. ~ Rob₁₃^Talk 21:39, 5 August 2017 (UTC)
3. Oppose not technically viable, too many false positives and false negatives will occur. — xaosflux ^Talk 04:16, 6 August 2017 (UTC)
4. Oppose for purely pragmatic reasons. This is both technically infeasible and (even if it could be implemented as envisioned), it would be unlikely to remedy any problems. Determining the nature of combative comments requires context and nuance, which is not the sort of thing that you can get from the revision log; seeing those items flagged in the revision history and clicking on the diffs is hardly a simpler or more intuitive process than simply reading the comment in context of the thread. I will say that I agree with the OP's more general observation here: the root of the issue of most behavioural problems being brought to ANI of late is issues with civility. But that's a problem that goes beyond ANI and any technical fixes to version history logs. If anyone ever holds a broad ranging RfC on the issue of what has become of our regard for (and protection of) WP:C in recent years, please count me in; I have plenty to say. But this proposal feels like a non-starter, sorry Biblio. Snow ^{let's rap} 05:49, 6 August 2017 (UTC)
5. Oppose - See my general comment in the first section. Beyond My Ken (talk) 06:02, 6 August 2017 (UTC)
6. Oppose as technically impossible. Even multinational corporations struggle to come up with algorithmic means to flag potentially problematic comments; the idea that the Bot Approvals Group could succeed where Google, Facebook and Twitter have failed is so far beyond wishful thinking it's in leprechaun country. ‑ Iridescent 10:38, 6 August 2017 (UTC)
7. Oppose. Why privilege creative misspelling? A ¶huggink bad idea. The Big Bad Wolfowitz (aka Hullaballoo). Treated like dirt by many administrators since 2006. (talk) 13:37, 6 August 2017 (UTC)
8. Oppose - For two reasons: 1. As mentioned above, it would take a lot of time, and probably would not be worth it, as people are pretty good at recognizing personal attacks. 2. We should dedicate our research and such to things like quality control, not ANI. RileyBugz^会話_投稿記録 17:08, 6 August 2017 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Comments

• If this RfC gains has gained some traction after a week or so, I will add it the watchlist notice. I did decide to just include Proposal D, since clerking is commonly floated solution for everything. It may as well be discussed. However, I am actually leaning oppose on that proposal. Biblio (talk) 19:20, 5 August 2017 (UTC)
• Open hostility should be blockable on sight and should be blocked on sight. If we can't do even that without causing more commotion than we're preventing, self-governance is a failed proposition and any ANI reform initiatives are a waste of time. ―Mandruss ☎ 04:36, 6 August 2017 (UTC)

+1 Snow ^{let's rap} 06:03, 6 August 2017 (UTC)

• I'm going to break with my personal practice and make a purely impressionistic observation here: Yes, I have, on a few score occasions over the years, heard AN/ANI referred to as the "Dramahzboardz" or something similar. It's usually the user with a block log a mile long and a habit of treating other people like dirt, and who never responds well to others who call out their behaviour (or that of the other needlessly caustic editors they gravitate towards working with and who share their editorial perspectives), no matter what the context it is done in, or how civilly and reasonably it is approached. Assuming that there is a problem with ANI on the basis of this kind of commentary/"metric" is not particularly rational, but in fact rather susceptible to a number of different cognitive biases, not the least of which is confirmation bias. It's easy to notice when people complain, or especially when genuine issues of disruption break out in a space. It's even easier to fail to notice when the process (much more routinely) works well (or as well as might be expected for space designed specifically to handle the most contentious and intractable issues, including longterm abuse and personal disputes).

ANI is not perfect, but respectfully, proposals which mostly handicap open and productive discussion by the community (which describes most of the above) would only worsen the issues there, not remedy them. The single best thing this community can do to alleviate the issues at ANI is encourage more good, level-headed and experienced admin candidates to take up the bit and volunteer there. The major issue is that threads languish too long, with two few hands to take action (and many of the regular admins a little too hesitant to take a hard line on disruptive behaviour, especially incivility--but that's backseat mopping, I recognize. :) Snow ^{let's rap} 06:03, 6 August 2017 (UTC)

Per my comment in the first section, I agree that the lack of patrolling admins is one of the primary problem at AN/I. If someone really feels the need to improve AN/I, how about a short (1 year?, 6 months?) probationary period for newly minted admins where they are required to do nothing but patrol the areas of en.wiki which are notorious for being under-patrolled or getting backed-up, places such as AN/I, AIV and UAA. Beyond My Ken (talk) 06:11, 6 August 2017 (UTC)

I had thought about suggesting something similar, but I have to think the community is likely to view it as running afoul of WP:VOLUNTEER and the general principle that admins (like all contributors) volunteer their time when and where they choose and are best encouraged to work in the areas they already know best. There's also the issue of possibly flooding these spaces with a sudden surge of newer and less tested admins. I do think it would not be a particularly bad idea to make more effort at RfA to make sure that candidates have made themselves well aware of how these spaces work and how to best to resolve the kinds of issues that arise in them. That would at least help assure that more of our incoming admins might spend some time within them, and contribute productively to easing the backlog. Snow ^{let's rap} 06:20, 6 August 2017 (UTC)

Not going to work; ANI has that reputation whether justified or improvable or not, to the point that "too many contributions to ANI" is held up as a negative in RfA. Also, having passed RfA does not by default entail that one knows how to manage an user conduct dispute. Jo-Jo Eumerus (talk, contributions) 10:23, 6 August 2017 (UTC)

No, but maybe having that skillset is something we could stand to prioritize a little more at RfA, that's all I'm saying. As to "too many contributions at ANI" being cited as a negative--I've mostly only seen that in cases where the party has themselves been the subject of and/or started too many threads. I think most people who !vote in an RfA have better sense than to hold it against an editor for having volunteered a lot of community hours specifically in an administrative area. I would hope so, anyway. Snow ^{let's rap} 11:19, 6 August 2017 (UTC)

• I'm unable to support any proposal here, but I have a few extra comments. First, I think each section above could do with its own comments section. ANI has a wide range of topics - most of my participation involves responding to - "please move this page", pr "please block this vandal" and similar, which are vastly different from the problems being discussed in this RfC. These options clearly have context in the long rambling threads where people are trying to get others blocked, and it's not helping support that more focus isn't placed on the particular situations that cause the problems. I would certainly not like to see a return of the Community Sanction Noticeboard (or Request for Banning as it was known), but some separation of how these functions are performed might be an idea.

I can't see a word limit working, but I've long wanted some restriction on how many separate responses people can make in some types of threads. Sometimes you'll find and read a huge thread requesting a ban or something, only to find out that it only has three participants, and none of whom declare their prior interests. The idea of a non-involved section is also a good one. And another thing related to this - I find non-admin input is often useful all across the admin noticeboards, but on some occasions you'll find 10 non-admins offering their unnecessary input on a topic before encountering an admin who can quickly resolve an issue. I've even heard it said at RfA that participation at ANI is a good thing to practise (which I don't agree with). And some people are happy to just comment at ANI all day. I don't know the solution to that, but it hugely bloats the size and complexity of the noticeboard. -- zzuuzz ^(talk) 06:32, 6 August 2017 (UTC)

I don't have an easy solution to that issue either, other than to say that it's definitely not restricting who can comment there; this is a vital community space where all contributors need to feel free to speak their piece, provide input and make observations, regardless of privileges. I think the only reasonable solution to the problem is the one discussed immediately above: more admins reaching the issues more quickly, before involved parties (or those who comment regularly with axes to grind) can hijack discussion and heighten disruption/complicate otherwise easy solutions. But of course that's no real answer at all, because it begs the question: how do we get those extra admins there? Maybe free vouchers for temple massages for admins who respond to a certain number of ANI threads in a month? :) Snow ^{let's rap} 07:17, 6 August 2017 (UTC)

it's definitely not restricting who can comment there Sure about this? Especially since you follow up with involved parties (or those who comment regularly with axes to grind) which to me indicates that there is a group of users whose contributions is less than useful. Jo-Jo Eumerus (talk, contributions) 10:27, 6 August 2017 (UTC)

The problem is, you can't create blanket rules which play to the lowest common denominator if they will have a significant impact upon the open consensus process. That's throwing the baby out with the bathwater. Besides, people who enter the space for the explicit purpose of engaging with someone they've had a previous dispute with tend to be pretty obvious. Snow ^{let's rap} 11:19, 6 August 2017 (UTC)

That assumes the current process is just what we want—but in truth, the way most contentious discussions are decided are by a straw poll followed by people trying to persuade individual voters to change their minds on their views. This leads to a lot of redundant conversation, making it hard to follow all threads, which just exacerbates the problem by leading to more repetition. Maybe ANI should have the concept of a facilitating admin, a little bit like the volunteers at the dispute resolution noticeboard. Everyone could post an initial statement, but without any threaded responses until an admin responds as the one who will be providing some initial guidance on this incident. The admin would then pose questions or otherwise shape the discussion to try to make it as efficient and effective as possible. This would avoid the inflexibility of a fixed template, and save time for everyone by reducing repetition. For incidents with a clear resolution (e.g. a vandal report), any admin would be free to jump in at any time and implement appropriate action. However the same problem as always arises: I don't believe there would be enough people willing to staff this. (Hiring people simply to do facilitation would alleviate the staffing issue; so far I haven't seen much interest in this approach.) isaacl (talk) 14:19, 6 August 2017 (UTC)

• Good that you are thinking about reform of ANI. But there doesn't seem to be much here to deal with some of the main problems. Which I'd say include, first, misleading statements and diffs. Even often stale issues brought up as double jeopardy. Eventually someone may notice a dozen comments down a long confusing discussion, but the original statement with all its falsehoods stays. People do lie to achieve what they want, and that definitely happens on ANI and there is no due process to check against this at all. I've voted for the waiting period of 24 hours. But I think what we need as well is some process of checking the accuracy of the accusation too, before anyone votes. Just checking for obvious things. Double jeopardy as I said - referring to things in old closed cases without saying that they are closed. Also, diffs that are misleading - that don't mean what they are claimed to say when they are read in context. Those are the two most used tactics but there are many more. So, there are many ways we could deal with this, but for the action I voted for, in the 24 hour waiting period I'd suggest that members are encouraged to check the accusations for accuracy - not just wait for the accused to respond. The only discussion permitted would be over the accuracy of the accusations in that time period, in a separate section of the action. That would encourage accusers to pay more attention to accuracy too. It's sometimes deliberate deceit for sure. And sometimes the way anyone can get carried away with the heat of the moment and their feelings and exaggerate and mislead without even noticing they are doing it. I'd add that as a new proposal but I don't think we are permitted to add new proposals. So just a thought for anyone reading this for the future. Robert Walker (talk) 18:03, 7 August 2017 (UTC)

• tl;dr I personally don’t believe ANI in its current form is the most optimal way to report and resolve harassment occurring on Wikipedia. Professionally, my team and I are looking into ANI and want to involve the Wikipedia community in our decision-making process.

Hello! I’m Trevor, a product manager at the WMF working on the Anti-Harassment Tools team, a segment of the Community health initiative that has already been referenced on this RfC.

While ANI is not solely for reporting harassment, it is one of the primary conduct dispute locations. We’ve recently begun a project to look into ANI to gain deeper understanding into how it serves (or fails to serve) Wikipedia users. We want to use both qualitative and quantitative data to inform our work. Our goal for the end of September is to prepare this research methodology, with executing this research scheduled for the final months of 2017. Our first step in this journey is to attempt to parse ANI archives to get some high-level statistics. Unsurprisingly, the unstructured nature of wiki page makes this a bit unpredictable. You can view our findings on T171091.

We’re hoping that we can use this research to form some shared agreements about the problems worth solving — do cases drag on too long? does the lack of word limit help or hurt? what about the lack of structure? the lack of ownership or responsibility for participation? Would splitting harassment reports into a separate workflow help or hurt? We (all) have hypotheses but need them to be validated before we move into the deciding on solutions.

Most importantly, we want to include the English Wikipedia community in this entire process. This weekend at Wikimania we’re holding both a workshop and giving a lecture about our work. (Join us if you’re attending!) In the coming weeks we will move these discussions on-wiki. In the meantime, is there any data about ANI you’d like to see? We may be able to get it from our first parse or may add it to our list for future gathering.

Thank you! — Trevor Bolliger, WMF Product Manager 🗨 23:18, 7 August 2017 (UTC)

Please be careful to weed out true cases of harassment from the complaints of overly sensitive editors who see any prolonged disagreement with them as being "Wiki-hounding". It's important not to fall into the current societal trap of taking every complaint and complainant at face value, on the theory that "where there's smoke, there's fire." Sometimes, there's just smoke (and sometimes mirrors as well). Beyond My Ken (talk) 13:53, 8 August 2017 (UTC)

Absolutely. We're not trying to create the 'nice police' — we believe that healthy debate is vital to maintaining the high content standards of Wikipedia. Unfortunately harassment is nuanced and contextual, so what one person may think is a curt message may be perceived as harassment by another. We hope to determine the best possible reporting mechanism that deters junk reports, and evaluation tools that allow admins (or other community leaders) to quickly determine if a case has any merit. — Trevor Bolliger, WMF Product Manager 🗨 14:55, 8 August 2017 (UTC)

• Personally, as an editor and then administrator who's been watching & participating at ANI for several years, I think it's working just fine. Yeah, it often gets dramatic and uncivil, because that's the venue where we resolve complicated disputes with editors who (are supposed to) have tried and failed to resolve their disputes through discussion and/or in the more formal process-driven venues. Dumping more process and restrictions into this is only going to give upset editors more things to be upset about and hinder dispute resolution. We do give editors at ANI a little more leeway in terms of civility, it's rarely helpful to tell an angry person to use nicer language, but we already don't tolerate outright personal attacks, and at ANI I don't think we should go much further than that. If ANI is known as a venue of last resort (excepting ArbCom, which is incredibly formal) and a place to avoid when possible, all the better for resolving incidents which actually need wider community intervention. Ivanvector (^Talk/_Edits) 12:40, 11 August 2017 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.