Yesterday, Julie Zhou, a product design manager at Facebook, wrote a proposal in the NYTimes titled, Online, Anonymity Breeds Contempt. This is the latest salvo in the Civility Wars meme that has been floating around the internet but this one has more severe consequences for free speech than any proposition passed by the righteous Christian voters of Oklahoma.
Zhou’s premise is that the online world is populated by trolls, which almost seems like an idea for a fantasy series world. We all know what trolling is, although Zhou’s definition, “defined as the act of posting inflammatory, derogatory or provocative messages in public forums”, is rather broad. Trolls come in many flavors. Some of them definitely have an agenda and use carefully crafted tools of the marketing trade to manipulate readers, all of this without raising the voice. Yes, trolls are nasty and unpleasant and sometimes downright creepy. That’s what spam filters and septic tanks are for.
They are also anonymous. Zhou has a problem with anonymity. In Zhou’s world, anonymity is the scourge that must be eliminated so that we can all speak openly and harmoniously, where users can get a friction free experience and share their Farmville produce with each other. All very *nice* and completely soporific. But Zhou’s explanation of the perils of anonymity had my tinfoil antenna twitching:
Psychological research has proven again and again that anonymity increases unethical behavior. Road rage bubbles up in the relative anonymity of one’s car. And in the online world, which can offer total anonymity, the effect is even more pronounced. People — even ordinary, good people — often change their behavior in radical ways. There’s even a term for it: the online disinhibition effect.
Many forums and online communities are looking for ways to strike back. Back in February, Engadget, a popular technology review blog, shut down its commenting systemfor a few days after it received a barrage of trollish comments on its iPad coverage.
Many victims are turning to legislation. All 50 states now have stalking, bullying or harassment laws that explicitly include electronic forms of communication. Last year, Liskula Cohen, a former model, persuaded a New York judge to require Google to reveal the identity of an anonymous blogger who she felt had defamed her, and she has now filed a suit against the blogger. Last month, another former model, Carla Franklin, persuaded a judge to force YouTube to reveal the identity of a troll who made a disparaging comment about her on the video-sharing site.
But the law by itself cannot do enough to disarm the Internet’s trolls. Content providers, social networking platforms and community sites must also do their part by rethinking the systems they have in place for user commentary so as to discourage — or disallow — anonymity. Reuters, for example, announced that it would start to block anonymous comments and require users to register with their names and e-mail addresses in an effort to curb “uncivil behavior.”
Some may argue that denying Internet users the ability to post anonymously is a breach of their privacy and freedom of expression. But until the age of the Internet, anonymity was a rare thing. When someone spoke in public, his audience would naturally be able to see who was talking.
Others point out that there’s no way to truly rid the Internet of anonymity. After all, names and e-mail addresses can be faked. And in any case many commenters write things that are rude or inflammatory under their real names.
I have a problem with the idea that anonymity on the internet should be denied to the casual user in a thread or that anonymity is a scourge on the internet, a necessary EEEEVVVVIL. I suspect that few commenters who post to blogs or pages on sensitive subjects such as politics, religion, homosexuality or just sex in general, would feel that anonymity is a problem that needs a solution. Without anonymity, many online personalities would stay trapped in their heads, unable to find an outlet to express their opinions, dissent or eccentricities.
And then there is the issue of what constitutes trolling. One person’s harrassment may be another person’s vigorous attempts at debate. Long time readers of The Confluence will recall that this blog was created as a result of being kicked off of DailyKos in the early stages of the primary wars of 2008. The peaceful and genteel, law abiding citizens of DailyKos ran me out of town on a rail because I had the temerity to question their consensus reality about Barack Obama and challenged them to look at their behavior as moblike, forcing Kossacks to convert to Obama or die. (See Obamaphiles carry out Jihad on DailyKos)
Zhou’s recommendations for *fixing* the troll problem are remarkably (uncannily?) similar to the DailyKos model. She would have ‘trusted users’ among other tools:
The technology blog Gizmodo is trying an audition system for new commenters, under which their first few comments would be approved by a moderator or a trusted commenter to ensure quality before anybody else could see them. After a successful audition, commenters can freely post. If over time they impress other trusted commenters with their contributions, they’d be promoted to trusted commenters, too, and their comments would henceforth be featured.
Disqus, a comments platform for bloggers, has experimented with allowing users to rate one another’s comments and feed those ratings into a global reputation system called Clout. Moderators can use a commenter’s Clout score to “help separate top commenters from trolls.”
At Facebook, where I’ve worked on the design of the public commenting widget, the approach is to try to replicate real-world social norms by emphasizing the human qualities of conversation. People’s faces, real names and brief biographies (“John Doe from Lexington”) are placed next to their public comments, to establish a baseline of responsibility.
Facebook also encourages you to share your comments with your friends. Though you’re free to opt out, the knowledge that what you say may be seen by the people you know is a big deterrent to trollish behavior.
This kind of social pressure works because, at the end of the day, most trolls wouldn’t have the gall to say to another person’s face half the things they anonymously post on the Internet.
I have to give her credit for admitting that social pressure is the goal because that is exactly what happened at DailyKos. But in this case, the trolls gained “clout” through recommendations from other users, some of whom may have been brother trolls-in-arms. After a certain amount of clout, those trolls became trusted users with the minor but utlimately significant power to upgrade or degrade another user’s clout. This became a very effective method of social control, one that David Axelrod was going to use for all it was worth. And he undoubtably did.
The result was a takeover of DailyKos through a very effective troll campaign. Here’s how it worked and how it will work on Facebook:
1.) A user writes a conversion diary or page. It has quasi religious overtones.
2.) The conversion diary is hit with massive mojo or “clout”. The diary moves up the recommended or “like” list. Those of us who were frequent DailyKos users couldn’t help but notice that the same people immediately recommended these diaries.
3.) Social pressure is used to reinforce consensus reality by rewarding the desired expression and by punishing undesirable expressions. The term used for this when referring to cultlike behavior is “love bombing“. The good user is praised, told how smart and attractive he/she is and how different they are from the rest of the world. The undesired users become targets for a campaign of decremation of their trusted user status. Yes, friends, I was a trusted user on DailyKos up until the day I wrote my last post.
4.) Bait Ball frenzies result where gangs of now motivated users decry the dissenters lack of civility, driving the user and entire modes of thought out of the public domain.
If this is the way Facebook is trending, and I see it in their ubiquitous “like” buttons, I don’t want any part of it. Zhou’s piece reads like an attempt to exert social control. But I don’t know whether Facebook will cooperate with the people who want to shape consensus reality. To the Facebook user, these entities may remain ‘anonymous’. But their force may be deployed throughout Facebook’s domain and that is where there is real danger. Because for many people, especially older or less technologically savvy users, Facebook is the entry point and the internet space where they spend most of their time. If they are exposed to the sophisticated marketing techniques that we saw in the 2008 election season, they may be unaware of how their views and opinions are being shaped and reinforced, all in the name of civility.
I don’t need Zhou to tell me how utterly obnoxious trolls are. They can be particularly hard on women who are conditioned to take insults personally. To female bloggers I have always emphasized that trolls are nothing more than black pixels on your monitor- they can not hurt you. As a blogger, you have time to formulate a snappy response to bring them to their trollish knees. But we have other tools to be used judiciously. We can filter comments automatically using trigger words, throw people into moderation or the spam filter and moderate threads to keep people from harrassing other users. But harrassment is sometimes in the eyes of the beholder. The concept of bullying can itself be used to bully people into silence. I don’t want any part of that.
I want the internet to be free to users whether they choose to remain anonymous. I have found that anonymous users are no more trollish than the identifiable user. It is through anonymity that unconventional but good ideas enter the public debate. Anonymity gives users the freedom to express themselves without exposure to real world friends and families and without social pressure to keep their fcuking mouths shut.
(Hey! Leo Laporte! this would be a great topic for This Week In Tech.)
Filed under: General | Tagged: anonymity, Julie Zhou, Leo Laporte, trolls | 97 Comments »