I’ve wanted to write about this for a long time. In fact, I did write about this a long time ago. This essay has been sitting in the drafts folder for over a year. I keep meaning to publish it, but then I chicken out. This has been in my “rainy day content” queue for a while nowI’m not suddenly braver today, I’m just out of regular content..
See, the point I want to make has the stench of moral panic about it, and I hate, HATE moral panics. I hate when people start whining about how some awesome new thing is killing us, or corrupting our children, or problematic, or whatever. I’ve lived through periods where middle-aged dunces got themselves worked up over Dungeons and Dragons being Satanic. Rock and Roll is turning teens into hedonists. Videogames are “Murder Simulators”. Smartphones are giving us ear cancer and making us antisocial. I know how obnoxious it is when old people get worked up about New Things.
And yet here I am, 46 years old and wringing my hands over a New Thing.
I don’t want to start a moral panic over this. Instead let’s call it a Moral… “Unease”? “Mild worry”? “It might be good to keep an eye on this”? Whatever. I want to worry about a new thing but I want to do it in a chill and non-confrontational way.
The point is, there’s a game everyone is playing today and it really is bad for you. Having played this game for eight years I can tell you first-hand that it really does impact the way people behave and perceive each other. And I’m not talking about in-game behavior, here. I’m talking about real, lasting consequences in the real world. I’m talking about a game that can actually change the way you see other human beings, and how you treat them. It’s a game that’s genuinely harmful and continues to impact your thoughts and behavior, even after you stop playing it.
The game is Twitter.
“But Shamus, Twitter isn’t a game!”
Right, right. I mean obviously it’s not an actual game. But hear me out.
The Mechanics of Twitter
Twitter isn’t a game, but it does have some game-like mechanics. It has a system of positive and negative feedback, and it has an implicit goal. It rewards certain behaviors and punishes others, and this feedback drives behavior. Just like the score in Asteroids tells you your goal is to kill ships and asteroids – and not simply to fly around in circles dodging stuff forever – the mechanics of Twitter push you into behaviors you might never otherwise engage in. They also intensify the existing social rewards for negative behaviors. While not part of the deliberate design of the system, these behaviors are actually kind of insidious.
On Twitter, followers and re-tweets are your reward currency. Just as Pac-Man encourages you to eat stuff to accrue points, Twitter encourages you to say things that others will want to repeat, with the ultimate goal of growing your followers.
On paper, this sounds like a brilliant system for encouraging people to say healthy, intelligent, witty, or clever things. If I’d never used Twitter myself, I would incorrectly predict that this would create a system to reward positive behavior and discourage negative behavior. In theory, it sounds like a system that would create harmony and encourage understanding.
And to limited extent, it does. If you say things that are negative, obvious, stupid, or obnoxious, people aren’t likely to share your thoughts. The system is a lot of fun and works basically as intended…
Until you introduce politics.
And just to be clear: When I say “politics”, I’m talking about partisan politics. Red v. Blue. Democrats vs. Republicans. (Or whatever the teams are called in your area.) Right vs. Left. Commies vs. Capitalists. Us. vs THEM. Good guys vs. bad guys.
Let’s say there are two factions on Twitter who are embroiled in a long-standing cultural debate: the Purple Team and the Yellow Team. They’re not fighting about which Pokémon is best, or the pronunciation of gifFor technical reasons, it’s actually pronounced “Jiggity-jeff”.. They’re fighting about real-world political stuff that’s deeply personal.
I’m sure you’re already aware of the pull in Twitter towards groupthink. That’s where you’ve got two warring factions. Let’s say we (everyone on this site) is a member of the Purple Party. One day I say something snarky about Yellows and some of them unfollow me, while Purples retweet the joke and I get more Purple followers. If I play to one side like this then I’ll get more followers. More retweets. More engagement. In the game of Twitter, my “score” goes up.
Over time, I become a Purple talking mostly to other Purples about how stupid Yellows are. You know this one, right? Lots of people have noticed this already.
So the mechanics of Twitter pull us towards groupthink. That’s bad, but lots of sites do that. At this point a lot of people are going to link to this video by C.G.P. Grey about anger germs:
That’s a really good video, but Grey is talking about anger-inducing ideas and I’m talking about the system that delivers those ideas and what it does to the participants. People aren’t just passive consumers in this system. They are changed by it, and unlike Grey’s germ metaphor, being infected doesn’t make you immune, it makes you more susceptible to increasingly virulent anger-ideas.
Automated Confirmation Bias
Let’s say I notice a Yellow moron who says something shockingly, stupefyingly ignorant. Something so hilariously clumsy and uninformed that they accidentally make a point in my favor. I can share that. My fellow Purples will love it. We can laugh and enjoy a sensation of smug superiority. Tweets by idiot Yellow become a kind of currency. Whoever can find the most stupidest Yellow things becomes king of the Purples.
Over time, the image of the incoherent, uninformed Yellow becomes our default image of all Yellows. It’s how we see them. Every day we’re treated to hilarious images, memes, and screenshots of dumb Yellow comments. We don’t see the smart things they say. We don’t see their witty, calm, and engaging members. For us, Yellow is an army of enraged, uninformed, tone-deaf imbeciles.
And on the other side of the divide, Yellows are doing the same thing to us, building up a false image of what Purples are like and how Purples think, based on the behavior of our most obnoxious and stupid members.
I hear you saying, “Yes, Shamus, everyone knows what confirmation bias is.”
Except, that’s not what this is. Confirmation bias is when I see a news story on television about someone from the Purple Team who got caught in a scandal, I think of it as an anomaly. Then I see an Yellow caught in a scandal and I think “Man, that’s just how they are, always doing scandals and being corrupt. Dang Yellows.” Confirmation bias happens inside your head. It’s something you have to do to yourself.
Twitter is like owning a television that only shows me scandals by the Yellows, and acts of virtue by Purple. I no longer have to go to the trouble of fooling myself. We’ve invented software to automate and industrialize the process, and then added a scoring system so that the masses will constantly bring it fresh fuel. It’s a system of rules with the emergent property of creating a continuous flow of crowd-sourced propaganda. We’ve gameified tribal bigotry.
Shamus, you’re engaging in false equivalency here! One side is CLEARLY worse than the other!
How can you be sure of that? Do you hang out on the Yellow message boards, go to the Yellow news sites, follow the Yellow celebrities, and read books by Yellow thinkers? Or do you hang out with your fellow Purples and allow the group to curate your Yellow News for you?
And even if you are the sort of person who carefully balances their information intake to avoid social-media distortions, you have to remember that other people don’t. They are acting on the best information available to them, and from their point of view Yellows are great people with a few bad eggs and Purples are an army of vile jerks. In truth, it doesn’t matter which side is objectively worse because both sides are looking at conclusive evidence that their side is in the right and the other side is dangerous. It’s nice that your enlightened perspective enables you to see the truth, but it does nothing to stop the rage-war. Do you intend to convince the Yellows that they’re really the worst despite the evidence staring them in the face? Good luck with that.
There is indeed terrible injustice. I’ve seen people I love and admire as creative individuals endure horrible, vile abuse over the most innocuous things they said. I’ve seen people I respect say obnoxious and uninformed things about other people I respect. Yes, it’s nasty out there. But if both sides exaggerate the faults of the other while remaining blind to their own faults, then your virtuous behavior can do nothing to balance things. The feedback loop will continue to enrage them both.
One Hundred Punch Man
If I say something snarky about Yellows, then a hundred Yellows will lash out in response. To them they’re trading barbs with someone on the internet. I insulted them, and now they’re insulting me back. But from my perspective it feels like I’m being mobbed. Twitter is a world where throwing a punch means you get a hundred punches in reply.
Help! All I did was express my opinion and now I’m being flooded with hate!
It makes me feel like the oppressed underdog fighting the masses. This means I can feel comfortable hitting harder, fighting dirtier, and calling for like-minded people to join me. Because I’m outnumbered, right? I’m the victim. The underdog. Heck, if you look at my enemies you might even conclude I’m the hero.
The next day I see some sass from an Yellow and shoot back with some snark of my own, oblivious to the fact that I just joined a massive dogpile on one person, and they’re now going through what I went through yesterday.
The Cycle of Outrage
Bob is in a bad mood. Some Yellow-party wonk just said something completely insulting and unfair about Purples. So in an attempt at catharsis, he lashes out on Twitter. He says something equally cutting about Yellows. He gets retweeted, some Yellows see it, and they respond in kind. Suddenly Bob has a hundred insults from strangers on his timeline and he’s feeling worse than ever.
However, one or two of the responses go beyond mere insults. They reveal a horrible, ugly arrogance and hatred towards Purples. Out of his hundred angry responses, Bob screenshots these three or four and shares them with his followers. “This is what we’re fighting against!” he tells people. Purples rally to his side. They encourage him, they retweet him, they follow him. His score goes up and he finally gets the affirmation he was looking for.
And on the other side, a Yellow sees all of this happen. Annoyed, they seek catharsis. They lash out on Twitter…
Twitter Should Do Something!
People keep complaining that Twitter needs to “Do Something” about abuse. Well, they have. They tried. Sort of. They have some sort of quasi-curation designed to filter out “low quality Tweets”, which in theory should help you avoid being hounded by anonymousAs in: No established persona. A young account with no followers or profile pic, which suggests this is possibly someone who is starting over with a new account after being banned. trolls. Some people claim it’s helpful. Other people claim it doesn’t do a damn thing. It’s impossible to say who is right because everyone has a different experience.
The problem isn’t that Twitter doesn’t care, the problem is that the basic mechanics of Twitter create a no-win scenario. A social media Kobayashi Maru.
Twitter is a world where people speak in abbreviations, memes, movie quotes, and where sarcasm is more common than punctuation. It’s a world where people use outrageous hyperbole for comedic effect, and where people communicate in images that mean different things to different groups. It’s a world where people speak countless languages and come from different cultures and all have different ideas about where you should draw the line between abusive behavior and “they were just speaking figuratively”.
There are over 330 million of these people using the service. That’s 39 times the population of New York city. Do you have any idea how many moderators it would take to really police that many people? And then how many managers and HR reps it would take to make sure they’re all enforcing policy correctly? Twitter has been losing money since the service was switched on 12 years ago. That’s a long time to lose money, and it means they probably don’t have the funds to hire literally thousands of moderators.
People see vile hate in their Twitter feed and don’t understand why Twitter can’t just get rid of it. What they don’t understand is the sheer terrifying scale of the problem. People post over 500 million Tweets a day.
But even if they did somehow come up with enough money to hire the army of moderators required to personally investigate every dispute, it still wouldn’t solve the problem. Instead of saying vile things to people they want to harass, trolls will goad their targets into saying vile stuff and then (ab)use the moderation system to get them punished. To the harasser, Twitter is a game. Making people miserable is how you win. Strong moderation doesn’t stop the game, it just changes the meta.
Yes, people can and do use “Abuse Report” type systems as a weapon. I’ve seen it happen. A couple of friends of mine were having a go at each other on Twitter. She dissed a show he liked. He said something snarky about her favorite show. She replied with an obvious (within this context) comedic threat about hunting him down. She knew it was a joke. He knew it was a joke. Nobody’s feelings were hurt. But she talks politics on Twitter, which means she has enemies on Twitter, and they reported her for making “death threats”. Sure enough, she was slapped with a three-day ban. The system to prevent abuse became another tool of abuse. Expanding Twitter’s moderation team and mandate will also increase the incidents of weaponized rules-lawyering.
Ninety percent of everything is crap. This applies to people trying to be witty on Twitter. They see a stand-up comedian or a John Stewart style political comic and they try to imitate the joke format. But the tone of voice is lost on Twitter. They overlook the self-deprecating angle that makes the joke able to reach across political lines. Because they only have a limited number of charactersAnd they’re probably typing on a phone., they can’t give it the proper comedic framing before delivering the punchline.
The thing is, this person has no idea they’re terrible at comedy. Click on a trending hashtag sometime and see how many nonsensical, run-on, out-of-context Tweets you see. Then realize that every one of those people were trying to say something that made perfect sense in their head.
This presents a problem for you if you want to moderate hundreds of millions of Tweets. Sure, some Tweets are CLEARLY monstrous, grotesque abuse that have no place in civil society. I agree. The problem is that there’s a lot of shades of grey between that point and harmless jokes, and the masses will rules-lawyer you to death if you try to figure out where to draw the line. Everyone will keep exhaustive track of slights and punishments, insisting you need to be “more fair” to their side.
Do you think that a corporation can somehow wade in and mediate all of these millions of shouting matches between emotionally charged groups who all see themselves as misunderstood victims? When people on both sides are experiencing the “strike with one punch, receive a hundred in reply” user experience?
Even if the folks running Twitter could hire an army of moderators large enough to handle this load (hard to imagine) and even if every one of those people was perfectly impartial and fair (clearly impossible) and even if there was a simple set of rules that could make a clear distinction between abuse and satire (even more impossible) and even if the moderators had the wisdom and consistency to enforce those rules across all groups in all languages across the globe (this is getting silly) it wouldn’t solve the problem, because the only weapon Twitter has is a banhammer.
If someone is being abusive when posting under an established identity, being banned is not going to magically alter their personality. They won’t suddenly develop empathy, question their life choices, or even feel bad about their behavior. They were just silenced in the middle of (to them) an “argument”, and they’re going to want to jump right back in. How much worse will they be when they return five minutes later as an anonymous egg?
It’s a system of justice where the only punishment for criminals is that they have to wear a mask. The punishment will either have no effect, or remove the last of their inhibitions and create a user with a terrible grudge and nothing left to lose.
It would be impossible for Twitter to moderate all discussions, and even if they did it would just add fuel to the fire because it wouldn’t address the core loop of negative feedback that’s creating the agitation and hate. On top of being mad at each other, we’d all be mad at those obviously biased moderators at Twitter who clearly favor the OTHER site. (My fellow Purples have shown me countless screenshot examples of Twitter punishing a Purple for a crime and then ignoring a Yellow in similar circumstances. The evidence is clear!)
Dude. Yes, it sucks. I’m sure if anyone at Twitter had time to look at it they would be offended. But this problem is larger and more complex than you can possibly imagine.
So now you’re saying, “Okay Shamus. You think you’re so clever. What’s YOUR solution?”
Yeah. You think you’re so clever. What’s YOUR solution?
Oh, I don’t have one. Yes, people are nasty to each other on Twitter and that sucks. Sometimes it’s even scary. I’m just cautioning against the “we have to do something” attitude. Because if your only options are “Do nothing” and “Make the problem worse”, then doing nothing really is your best option.
The problem is that abuse is built into the very mechanics of the system. It’s an emergent result of the mechanics of Twitter when combined with divisive topics. It’s not intended, but it’s there, and I don’t think you could fix it without tearing it all down and starting over with something very different.
When I point out how unhealthy it is to fight on Twitter, my colleagues say things like, “Well, we have to stand up against this injustice.” Or, “We can’t let THEIR ideas go unchallenged.” That sounds reasonable. We’re just fighting for the truth, and that’s a good thing, right?
And yeah, if you’re just looking to make your Twitter score go up, it is a good thing. But if you’re looking to persuade people to your position and gain a better understanding of theirs, it’s a massive failure. The result of every one of these exchanges is that the two sides hate each other even more. Nobody’s mind is changed. Nobody learns anything. Nothing is gained except that Twitter gets the traffic it’s hungry for.
Twitter is a great place to talk about the lighter matters in life, but it’s a terrible battleground for our culture wars. I’d even go so far as to say that Twitter might be a contributing factor in why politics seems a more rancorous now than it did a few years ago. I’m not saying everything is Twitter’s fault. I’m just saying Twitter was gasoline on the fire.
I’m not going to try to convince you to give up Twitter. That would be hypocritical anyway, since I still use it. And besides, if I ran around saying “You should give up Twitter because it’s bad for you!” then this would transform into a full-blown moral panic, and I have no stomach for that sort of thing. And even if I magically convinced you to give up this useful and fun thing, it wouldn’t do anything to solve the ongoing problem that Twitter is making us hate each other more intensely than ever. The problem is with human nature, and closing your Twitter account wouldn’t change that.
But if you find what I’ve said to be at all persuasive and you’re even the least bit worried that Twitter might be distorting your view of the the world or other people’s view of you, then I hope you’ll at least keep this stuff in mind when using it. Knowing the danger can go a long way to mitigating it.
Fight against confirmation bias. If the world starts looking clear cut, us-vs-them, and good-vs-evil, then you’ve probably drank more social media than is good for you. Behind all those silly names and joke icons are real human beings with real feelings. Some of them are nasty and some of them are nice, and if you condemn all for the actions of a few then you’ll lose your ability to see any of them clearly.
Note that none of this is lifting the no-politics ban in the comments. Sorry if that seems unfair. If you can’t leave a comment without taking a swing at those jackasses on the Yellow team then just give this one a pass. I’ll be back next week with programming posts.
Be careful what you put in your brain, because you can’t really take it out again. I don’t care what your politics are. If you’re good to people, then you’re my kind of people.
 I’m not suddenly braver today, I’m just out of regular content.
 For technical reasons, it’s actually pronounced “Jiggity-jeff”.
 As in: No established persona. A young account with no followers or profile pic, which suggests this is possibly someone who is starting over with a new account after being banned.
 And they’re probably typing on a phone.
Fixing Match 3
For one of the most popular casual games in existence, Match 3 is actually really broken. Until one developer fixed it.
A horrible, railroading, stupid, contrived, and painfully ill-conceived roleplaying campaign. All in good fun.
The Biggest Game Ever
Just how big IS No Man's Sky? What if you made a map of all of its landmass? How big would it be?
Starcraft 2: Rush Analysis
I write a program to simulate different strategies in Starcraft 2, to see how they compare.
Philosophy of Moderation
The comments on most sites are a sewer of hate, because we're moderating with the wrong goals in mind.