Schadenfreude as an Enforcement Tactic


(Shawn Holmes) #1

One of the fundamental leadership tactics that I continue to hear from mentors is “Praise in public, scold in private”. However, I continue to see examples of community moderators that break this rule, yet continue to remain effective and popular. A perfect example of this is one of the communities that the Discourse team looked at during its early R&D, the Elitist Jerks forum.

Like many other online communities, they have a clearly documented set of guidelines to follow. If a user chooses to ignore these rules, they are issued a temporary ban and issued a warning, which is sent to the user via private message. Permabans act the same way: the rule that was violated was included in the message, to provide the user with context, re-assuring them that what they did was inappropriate. All of which makes perfectly logical sense.

Where things get interesting is the Banhammer forum. Here, every warning issued, and every ban deployed, is listed in plain view for the EJ community to consume – a veritable cornucopia of offenses, nicely organized in a single list. What I learned today was even more interesting; EJ uses their twitter account, tied to a bot, which automatically copies these Banhammer entries to Twitter, for even more folks to see (convenient link back to the offense included!)

I’m very curious about this, because it seems to go against the rules we’ve been taught to build and foster strong communities (and EJ can’t be the only ones doing this). What makes this certain type of Schadenfreude effective without ostracizing the community itself?

In short, when is it ok to publicly humiliate your users in order to enforce your guidelines?*

(*humiliate is such an awful word – let’s just all say ‘point and laugh’)


(Simon) #2

That’s an interesting idea, never heard of that before. Though I think it is not necessarily meant to humiliate or ‘point and laugh’ at people. That entirely depends on the people who read those messages and on the people who are mentioned in them. At least to me, the examples in the screenshot do not read as if they were intended to ‘point and laugh’ at the menitioned users. Maybe I am just not that kind of person who points and laughs at others that much…

So, I think using some publicly available list of warnings/bans issued may be perfectly fine for your community, as long as you don’t cultivate it as a ‘point and laugh’ list. I would see such a list as a continuously updated repository of examples of what I should not do in this community. It takes the often abstract and general rules of the forum and gives concrete and real world examples of “violations” of those rules. If done nicely and the community does not ‘point and laugh’ at the users listed there, I can imagine this would help to “enforce” the rules and guidelines of your community.

I think this would make a great plugin and I could imagine using something like that (if I was running some forum).


(Jeff Atwood) #3

Well there is a real world equivalent, our local police blotter, which is frequently hilarious:

https://local.nixle.com/el-cerrito-police-department/

Examples:

Shoplifting – Safeway 11450 San Pablo Avenue
An officer accepted the citizen’s arrest of a 56-year-old man from Oakland after he stole groceries from the store. Due to the man’s extensive criminal history for theft, he was booked at the El Cerrito Police Department and transported to the Martinez Detention Facility.

Residential Burglary – 7100 block of Fairmount Avenue
Officers responded to a report of a residential burglary, in progress, at approximately 11:13am. While officers were en route to the address, the suspects fled in a silver vehicle. The home had been ransacked however, no loss was reported.

Public Intoxication – BART path at Blake Street
At approximately 1pm, uniformed bicycle officers contacted a 54 year old intoxicated male stumbling along the BART path. The subject was booked at the El Cerrito Police Department and later released to his wife with a citation.


(somnium) #4

Please allow me to (1) jump into the discussion without introducing myself, and (2) offer the analogy of driving on public roads.

We are all tempted to break the speed limit. Suppose that you pass by the side of the road, a powerful car that’s been pulled over by the police. Suppose that you see the same thing two or three times in the same journey. Will you be more or less tempted to speed yourself?

Suppose again that you move to a city where other drivers often break the rules (lane changing, speeding, overtaking on the inside, jumping lights, and generally behaving as it they were on a go-kart track rather than a public road). You never see anyone stopped by the police. How will you be tempted to behave?

I think these public notices are the equivalent of seeing other drivers pulled up by the police. It’s not so much about being afraid of being caught oneself, as it is about reinforcing the perception that civil behaviour is the norm, and antisocial behaviour an aberration.


(Kris) #5

I think an interesting place to look for moderation ideas would be B.F. Skinner’s work in psychology and operant conditioning… basically you have:

  • Positive reinforcement: A reward for positive behavior. A mouse
    presses a button and gets food.

  • Negative reinforcement: Removing something negative from the
    environment in response to positive behavior. A mouse is being
    shocked, and the shocking stops when a button is pressed.

  • Punishment: Punishing negative behavior. Shocking a mouse for
    not pressing a button when you tell it to.

  • Extinction: If no one pays attention to the behavior (good or bad) the behavior will stop. Nothing happens if the mouse presses the button, so it stops pressing the button.

Skinner’s research basically shows us that punishment for negative behavior is really only effective on the short-term and doesn’t actually change long-term behavior patterns. A punishment causes someone to avoid being punished rather than actually changing their behavior for the better. Prisons always come up as the obvious example - put someone in the prison system and when they come out they’re often still going to commit crimes, but will go to greater lengths to avoid prison rather than avoiding crime to begin with.

A person who has been punished is not thereby simply less inclined to behave in a given way; at best, he learns how to avoid punishment. — B.F. Skinner, Beyond Freedom & Dignity

The above applied to online discussion:

  • Positive reinforcement: Users get recognition in the form of awards or points for positive contribution

  • Negative reinforcement: Users with a certain number of awards/points/good posts have fewer restrictions on posting links, photos, etc

  • Punishment: Public humiliation, or ultimately the ban-hammer.

  • Extinction: Don’t feed the troll (or conversely, don’t recognize loyal and positive behavior)

Aside from cases that result in perma-bans for the worst of offenders (in forum terms these people cease to exist, which typically works out), I don’t think temporary punishment in the form of public humiliation is a great way of molding a community.

I think some sort of reward system (reddit’s karma system for example) is a pretty solid way of encouraging positive behavior while simultaneously discouraging negative behavior. Though honestly, I don’t think reddit should have a downvote system because it’s a form of punishment that seems too often cause hivemind scenarios and a community of “prison guards”.

A reward system also results in the building of a “reputation” that people feel a desire to protect, and in turn creates an investment in the community that often seems to result in a fair amount of community self-moderation.


(Simon) #6

I think this is not really a discussion whether punishment of unwanted behavior works or not. It is about the effects of publishing a list of “punishments” in a community which has already decided using punishments as a valid means for moderation (though I wouldn’t necessarily see something like a warning as “punishment”).

As I said before, I don’t see such a publicly available list as a necessarily negative thing or a further punishment by making it public. If presented with the right tone and in a friendly environment, this would not have to result in “Schadenfreude” (note that I am a native German speaker and know the meaning of the word quite well; I am not a “schadenfrohe” person). I really think this could work in some communities, while it could be useless or even counter productive in others.


(Jeff Lunt) #7

I’m definitely interested in fully open moderation experiments. I think there is a time and place to moderate privately, but also feel that moderation should become fully public after some time has passed, perhaps 30 days.

I would love to see a plugin that enables a view of this.


(Kris) #8

Right, I’m just saying that unless it’s a perma-ban — I don’t think punishments are all that effective at creating worthwhile community members because it doesn’t actually encourage positive behavior. Well-defined rules, maybe a warning or two, and perma-bans coupled with a system that rewards positive behavior is all that’s necessary in my mind.

I agree that this probably varies by community, so I can agree there — “Elitist Jerks” probably makes this work well because well, they’re a community that calls themselves “Elitist Jerks”. Their entire concept hinges on being incredibly tough on rules. I suspect such a structure wouldn’t work nearly as well in other communities that aren’t telling you they’re jerks up-front.

The name of these forums is not intended ironically; we have high
standards for the discussion that occurs herein, and we’re quite
unapologetic about it. If you feel our rules are stupid or arbitrary,
we don’t really care. If you don’t wish to follow them, you’re welcome
to return to the official Blizzard forums.

As a side note, I have in the past been a fan of the age-old disenvowelment - yes, it’s a punishment but it’s somewhat lighthearted/innocuous and covers multiple bases of being transparent, cleaning up a discussion, and serving as a warning.


(Jeff Atwood) #9

Well one big difference with the police blotter is it did not say this:

Public Intoxication – BART path at Blake Street
At approximately 1pm, uniformed bicycle officers contacted 54 year old intoxicated male Eugene Q. Moskovicz stumbling along the BART path. Eugene was booked at the El Cerrito Police Department and later released to his wife Marion with a citation.

However, it is an open question whether a forum pseudonym like “fluffybunny99” is really the same as publishing someone’s name or identity next to a black mark on their record. It is one degree of abstraction away, which might be OK… ish.


(Shawn Holmes) #10

So, perhaps the idiom “Praise in Public, Scold in Private” doesn’t apply when the public isn’t truly public; that is, when the offender is masked behind a layer of anonymity. This, coupled with a community which prides itself on strictness and clarity of rules, can warrant such an enforcement tactic.


(Jackdoh) #11

Does discourse currently track users’ “criminal record” (posts that have been flagged, etc) ?


(Jeff Lunt) #12

Actually I suppose what I’m in favor of is fully open moderation in the sense that the actions of moderators are audit-able, not necessarily attaching them to the identities of those moderated. What I wouldn’t want open moderation to turn into is a game of motivation toward desired behavior through shame.

The goal is to make it more transparent what exactly moderators do, and where they draw the line on specific behaviors on a community-by-community basis, in order to make accessible the moderation philosophy of a given community.


(Cliff_Figallo) #13

A lot of these considerations depend on the nature of the community, the size, the level of activity, etc. No one policy fits all just as community managers have different relationships with their communities. Originally, it would take a truly egregious pattern of offenses to persuade me to ban someone. I’d even call them on the phone to try to work things out. But eventually I learned that some people don’t respond to reason. We didn’t have to call them out but the community would understand what had happened because the rascal was suddenly absent.


(Chris Hanel) #14

I’d to offer up another variation of public enforcement - the voluntary petition.

This was used to great success on the Guild Wars 2 subreddit by Arenanet, and worked thusly: If you were banned or had action taken against your account, and you wanted to know why exactly you’d been targeted for punishment, you could post on a specific thread with your username and a request to know more. The catch was that by asking, you were permitting Arenanet’s staff to publicly display whatever idiocy it was that was the cause of being banned.

The result was that bystanders could see in plain view what would be viewed as actionable behavior, but the only people being made to pay in public for their actions were those voluntarily agreeing that this was an acceptable price for understanding what exactly caused their punishment. Between the combination of people ignorant of the rules, truly unsure of their infraction, and those fully knowing their crime but wanting to make a fool of themselves in public, there was more than enough to go around.

EDIT: Aha, found it! The thread is kind of hard to follow and most of the inquiries/responses have slid down due to reasons completely mysterious to me (god I hate reddit), but here’s the link:

Typical interaction in the thread (censoring added by me):

PLAYER: "Hey, can you tell me why I was banned?"
SUPPORT: “You were reported for shouting ‘GET THE F—ING CAVERN SCUTTER ENERGIES YOU F—ASSES.’ Not OK.”