5 Ways to Stop Trolls From Killing the Internet

By David Wong November 11, 2008 1,119,702 views

Add to Favorites

0

You might not be aware of this, but there are a lot of dickheads on the Internet.
Since this phenomenon seems to get worse with the size of the crowd, it is theorized that we will reach a critical mass; an Asshole Apocalypse, if you will. That's when casual Internet users--and the corporations who want their business--will step in.
There are ways to solve this crisis, but I'm telling you now, you won't like some of them.

But first, the problem...
Right away let me shut down everyone who's snorting derisively into their can of Mountain Dew and saying, "Trolls will be trolls!" You should know that there are billions of dollars at play here. The trolls are driving away business, and that simply won't be allowed to continue. I'm not saying I'm rooting for it--I'm saying that's the economic reality.
There are two huge, growing industries at stake: social networking and online gaming.
Social networking is at the heart of "Web 2.0," the future of the online world, the Facebook/MySpace/Twitter web where users create all the content and their parent companies make billions just for hosting it. It's a pretty sweet deal.

Or it would be, if they could only convince everybody to use it. But they're finding that lots of users will communicate online with people they know (virtually all use email and 37% use private text messaging), but only 8% use message boards or blogs or anything else that exposes them to the Internet's assheads.

Hell, look at this site. We just had an article that was read by 305,396 unique users in a few days ... but fewer than 100 of them joined the conversation down in the comments. That's .002%, folks. It's not that the Cracked comments are mostly retarded or nasty; it's that for a normal person, the memory of getting called a fucktard in public even one time is striking enough to make them avoid the comments forever, even if it was accompanied by 10 non-fucktard comments. It's human nature to remember the fucktard.
It's the same in gaming. There are reports that most people who give up online gaming aren't frustrated by the games themselves or technical issues. It's the sheer number of fuckwads they have to play with. Even on the most popular online multiplayer game, World of Warcraft, 70% of new players stay in modes where they don't have to interact with anybody else.
So there is a clear barrier to entry for the vast majority who haven't joined the Web 2.0 party, and that barrier is a moat full of dipshits. How can we bridge it? I see five ways:

#5.

Develop Anti-Troll Software

Imagine a world where you get in a heated argument in a hallway, but before even one sentence can get fully out of your mouth, a robot voice pipes up and tells you to cool it. Well, what sounds like really stupid science-fiction in real life is entirely possible online. Of all the futuristic movies to turn out to be cruelly accurate, who would have thought it'd be Demolition Man?
I'm talking about programs like:
StupidFilter:
This highly experimental piece of software is in beta and will some day be able to recognize comment stupidity the moment it's posted. They have a demo on their site you can play with.
You plug this code into your comment section and it's like a strap of tape over the mouth of every teenager who can't type a sentence without including the word "fail."

Robot9000:
This is a program invented by Randall Monroe, the XKCD webcomic guy that requires every post to be unique. If someone types "First!", no other post can ever consist of just that.
This sounds pointless to anybody who's never been in a chat room or message board before, but the rest of us know better. Mindless repetition of jokes (or "memes") is one of the primary tools of bored trolls who want to fill a thread with noise to drown out the signal. For once, many will find themselves using keys other than Ctrl-V.
Audio Preview:
Linguists speculate that no single body of written communication in the history of human language has ever been as collectively retarded and horrible as the comments under YouTube videos. After the aforementioned Randall Monroe suggested a feature to force users to hear their comment read aloud before they can post it, YouTube implemented that very thing (though only on an optional basis). Many a YouTuber has sat in dismayed silence after realizing that "lol wut", when spoken aloud, did not sound as clever a they had first believed.

Real-Time Voice Censor:
Now we're in the realm of the real Demolition Man-type solutions. Want to know how bad Microsoft wants to control the trolls on Xbox Live? They've patented a real-time voice censoring program. Yeah. You curse into your headset and it bleeps it in real time. How does it know the difference between "The cock crows at midnight" and "My cock grows at midnight"? With technology. Don't question it.
Of course, widespread use of this stuff will just kick off the same "DRM vs. pirates" arms race we see any time they try to control human behavior with software. The humans always win.
Also, the technology has to get a whole lot smarter before we can even try. Playing with the StupidFilter demo I linked earlier taught me that it doesn't find any stupidity in the sentence, "lol, wut your mom farts lolcats."
There are better ways. For instance, you can...

#4.

Start a Posse of Moderators, and Arm Them

Right now if you have a blog or forum or anything else with open comments, and you don't have a human moderator to watch it, you're going to wind up with a wasteland. As soon as more than one troll shows up, they will feed off each other until everyone else is gone. You have to control them. And don't start talking about free speech; the troll's goal is to shut down speech, to either fill the channel with noise until no one can talk to each other, or to get everyone talking about him instead of the subject at hand. He's a guy in a coffee shop screaming nonsense over a bullhorn.
And it's here where the marriage of creative software and human moderators can make all the difference. With things like...
Disenvoweling:
This is a bit of code that will suck all of the vowels out of a targeted post, so that this:
"What an unfunny piece of shit. Somebody should be fired for letting this guy write for the site."
Becomes:
"Wht n nfnny pc f sht. Smbdy shld b frd fr lttng ths gy wrt fr th st."
The theory is that it makes people slow down and try to parse what was being said and thus robs the post of its impact. Also it makes the troll look retarded.
Karma:
Geek megaportal SlashDot was among the first to use this, a way of allowing the community to moderate itself. Registered users can vote every post up or down, and each user winds up with a karma "score" that is just the sum total of all the "up" votes minus the "down" ones they've ever gotten.
We use this in the Cracked forums (where each member's karma score is visible to other members at all times). You can only vote once per day, so even a coordinated karma voting campaign couldn't change a score faster than the rest of the community could correct it.
Yes, it works. Everyone claims they don't care what their karma is, yet any time a person sees an unexplained drop, I get an email complaining about it. You just can't ignore a number right next to your name that announces what the community thinks of you.

But we're still thinking small, on a site-by-site basis. After all, assheads will simply migrate to places where security isn't as tight. If this is an Internet-wide problem, we need to think big. But how?