Hacker News new | comments | ask | show | jobs | submit login
Study finds web trolls get a feeling of abandon similar to drunks (news.com.au)
27 points by nreece on June 24, 2011 | hide | past | web | favorite | 13 comments

The best way to shut these people down is not to give them an outlet. Few people seem to step back enough from the problem to ask, "do we need comments?" Rather, it ends up being yet another half-assed feature that usually adds intellectual and visual noise to a page.

But, hey, it drives visitors to return, so it has to be good, right?

> The best way to shut these people down is not to give them an outlet.

I disagree. Give them a non-persistent outlet. Something like a shoutbox, where the trolling disappear after 5 minutes.

I don't think the concept of contributing anonymously is necessarily a bad thing. Your online personality might be polite and contribute positively where as your real world personality might very shy and unable to participate in discussions at all.

> "People believe the myth that they can say things that ordinarily they wouldn't be able to say just because they are online."

Is it really a myth?

I don't think it is. I frequently voice opinions in irc that I wouldn't dare mention to most people I know offline.

To provide anecdotal evidence of the contrary, anonymous discussions have helped me to become more honest about what I actually think and feel in real life. This comes with a side effect where one's social circle needs to be redrawn to reflect the change in attitude, since old friends only know the old you and habits and complacency and all that good stuff will end up holding you back.

So, a question for you epochwolf. Why would you not say these things to people face to face?

I think there's a converse to the Greater Internet Fuckwad Theory—let's call it the Greater Reality Stick-up-the-butt Theory. Basically, in real life, we enforce social norms on one another that we don't actually see the merit in ourselves, because others expect us to enforce those norms and punish/shun us if we don't enforce those norms (think of the monkey/water story: http://freekvermeulen.blogspot.com/2008/08/monkey-story-expe...). Anonymity allows for the relaxation of those norms, because nobody's around to catch you not calling someone out for doing something that you would normally call them on, but which you actually see no rational reason to be calling them on, besides some vague sense of "propriety."

Basically, anonymity allows people to treat everyone around them as utter foreigners—you don't expect me to have any grounding in your culture, and I don't expect you to have any grounding in mine, so neither of us "needs" to be upset when the other does something that is uncouth. Because of this, discussion can concentrate more on facts, and less on signalling, which I (and I think many introverted people) find a less stressful mode of communication.

Tangentially, I think this explains why many people on the internet have a annoyance-trigger at the undefined name-dropping of specific locales within, say, the US (with the usual response of "we're not all from the US", even if the person stating that is, in fact, from the US)—it grounds the discussion in a specific culture, which increases the likelihood of some subset of the people in the conversation finding a valid means of conveying subtext—which then leaves others out of that subtext, creating a kind of I-can't-enforce-norms-I'm-not-aware-of anxiety which is projected into a problem with the context being introduced in the first place.

In a lesser way, this is also why people tend to overreact to women who identify themselves as such on the Internet: they're introducing an ambient cultural context that increases norm-enforcement anxiety. (Interestingly, there's a specific name for Internet-age gender-role-relative-norm-enforcement: "white knighting." It's quite common to see people calling the practice out, though usually even the people themselves aren't sure precisely why they're unhappy about it.)

I'd say that's an overly internet-communication-favoured assessment of the situation. Note that people from the same culture talking over the phone also tend to pay less attention to whether they're offending the other person involved. At least part of the difference in communication style is simply because it's easier to not care about how you're hurting the other person when you can't see them, or directly observe their reaction.

Anonymous communication over the internet (for all its many, many virtues) depersonalises communication, and that has both good and bad results.

There is less risk in being honest when I'm online. If I voice an opinion around my coworkers or family that they strongly disagree with[1], there is a chance that could change my relationship with them for the worse. Some people hold grudges for a very long time over minor things.

On irc, if someone decided to hold a grudge I can simply add them to the short list of people I've put on permanent /ignore. In real life I have to deal with them on a daily basis.

There is also the added benefit of irc having scroll back and delays in communication. I can read what I've said before so I don't forget what I'm talking about. I can also walk away for 5 minutes to grab a drink and cool off if the debate gets heated. It would be very difficult to do either gracefully in a live conversation.

1: Example topic: The death penalty.

One of my coworkers has voiced a strong opinion on putting to death anyone that kills another human being except in a clear cut case of self-defense. I disagree with his opinion in the strongest way but I keep my mouth shut because inviting a civilized debate over that topic is a good way to ruin an otherwise enjoyable lunchbreak. On IRC, I would be happy to invite debate because the cost of the debate going bad is far less.

This study only adds new data to a fairly old (and rather simple) model that falls back on persistent character traits ("true character") and consistent individual perceptions to explain the behavorial effects of visual anonymity.

The model seems to be based on an archaic idea of man, much like the late 1960's concepts of deindividuation--the idea that underneath our civilized personality, there resides an instinctual force that is inherently wild, dangerous and destructive (neglecting any peaceful, friendly or cooperative traits that we also seem to possess).

More recent models like the 'Social Identity model of Deindividuation Effects' take a more systematic approach to considering further individual and situational variables and how they work together.

Owing to the enhanced complexity, these newer models do a much better job at explaining the factors and processes at play in computer-mediated communication, but they of course make for a much less catchy story.

I'm not sure I see how an alternate model changes much. In the case of SIDE, it is because they are actively viewing the out-group (the site's users) with enough contempt that they need to sow discord. OK, I can buy that.

But how does it help us know how to stop trolling? Simply saying, "it's complicated!" without offering workable solutions doesn't help us here.

I think my comment was more academic than pragmatic: Once you let go of the assumption that people will invariably do bad things in the absence of external control, the whole model falls apart.

As for practical recommendations, I've seen very few come out of academic research, unfortunately. From SIDE you could derive that you should aim to create a community or an idea that people identify with in a positive way (thus creating a social identity), and try to facilitate the salience of that identity by adjusting the situational setting.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact