How to kill a troll

A new Pew Research Center survey on online harassment – the first such of its kind undertaken by Pew – confirms what most of us already know: the internet can be a vicious, frightening place, especially for young people, and most particularly so for young women.

Pew surveyed 2849 internet users. Here are some of the results:

• One out of every four women between 18 years old and 24 years old reports having been stalked or sexually harassed online.
• Two out of five people reported having been victims of some form of online harassment.
• One out of four had seen someone being physically threatened.

Out of those people who’ve reportedly been the targets of online harassment, these are the rates at which they’ve suffered particular forms of abuse:

• 27% of internet users have been called offensive names
• 22% have had someone try to purposefully embarrass them
• 8% have been physically threatened
• 8% have been stalked
• 7% have been harassed for a sustained period
• 6% have been sexually harassed

Pew focused on two distinct but overlapping categories of online harassment: the somewhat less severe abuses, including name-calling and embarrassment (so common that many simply ignore it), and a second category that includes more severe attacks, such as threats of physical harm, harassment over a sustained period of time, stalking, and sexual harassment.

45% of harassment victims – that’s 18% of all internet users – have been plagued by the “more severe” forms of harassment.

Those more severe forms of harassment tend to be directed disproportionately at women – particularly young women, who see high rates of sexual harassment and stalking.

One of the most common pieces of advice given out to harassment victims has been to ignore their persecutors. That apparently seems to work: 83% of those who chose to ignore the abuse thought their decision made things better.

For the 40% of victims who chose to respond rather than to ignore their harassers (60%), these are the methods they employed:

• 47% of those who responded to their most recent incident with online harassment confronted the person online
• 44% unfriended or blocked the person responsible
• 22% reported the person responsible to the website or online service
• 18% discussed the problem online to draw support for themselves
• 13% changed their username or deleted their profile
• 10% withdrew from an online forum
• 8% stopped attending certain offline events or places
• 5% reported the problem to law enforcement

Automated troll killing

Technological approaches to policing trolls have been suggested.

Specifically, iOS developer Danilo Campos in July published a no-nonsense post, titled The least Twitter could do, in which he agreed that yes, policing a global community of Twitter’s scale “is tough work”, but no matter: it’s not a question of policy, at any rate.

Rather, it’s a software problem, he said, and went on to list some programmatic options that could be offered to users.

To wit:

• Block all users whose accounts are less than x days old. This would stop serial harassers cold – or, at least, for as long as x days. No more playing Whack-A-Mole with users whose response to account shutdown is to just go churn out a new one, which is then shut down, after which they go create another, on and on and on, ad nauseum.
• Block all users whose follow counts are less than whatever threshold users set. As Campos noted, this technique to establish credibility is old hat by now. Google’s been using the social proof of “back links” to rank content and credibility for 16 years. If somebody can’t convince others to follow them, users should be given the ability to block that person (though this option would have to also block rings of those who follow their Pied Piper from account to account).
• Block new users whose @replies include any words the user decides. Blocking new accounts that have no followers and/or who use the “c-” or the “n-” word: sounds familiar doesn’t it? Sounds like “if-then” programming, of course. Indeed: it’s “stupidly easy to express algorithmically”, Campos says.
• Block any user who has been blocked by more than n people I’m following. Nobody has to go it alone. This can be a team effort. Some of the most notorious trolls have long lists of targets. If we all get together and block the known trolls, then we’ve potentially spared future victims from their toxins.
• Auto-blocks are opaque. Campos doesn’t think that harassers should be made aware of being blocked and should instead believe that everything is working as normal. This is far from a clearcut issue, however. Twitter users themselves put up a storm of protest when Twitter last year decided to allow blocked users to continue to follow their targets, oblivious to the fact that they’d been blocked as they continued to interact with blockers’ tweets, receive their timeline updates, and thereby, critics said, enable blocked users’ friends to continue the harassment. A common thread in signatories’ comments at the time was that the change empowered stalkers whose activity was then hidden from potential victims. Twitter rolled back the change as a result of the backlash.

Campos’s proposals have already inspired a third-party tool for use with Twitter.

The free program, called Block Together, was created by Electronic Frontier Foundation technologist Jacob Hoffman-Andrews after he read Campos’s post.

Block Together lets users automatically block the accounts of new Twitter users who send them direct “@” replies and enables sharing of lists of blocked accounts with friends.

Another troll-blocking app, KnownTroll App, was released earlier this week.

If any of you try them out, please do let us all know, in the comments section below, how they work out.

Crushing trolls with embarrassment

While apps such as those two will hopefully go a long way toward shielding victims of harassment online, there’s nothing about them that serves to reach out to the abusers and shake some sense into them or cause them to stop their activities.

What would?

Danielle Citron, law professor at the University of Maryland and author of Hate Crimes in Cyberspace, draws a parallel between how sexual harassment was treated in the workplace decades ago and the current, harassment-heavy atmosphere of the internet.

The Atlantic quotes her:

Think about in the 1960s and 1970s, what we said to women in the workplace. 'This is just flirting.'

That a sexually hostile environment was just a perk for men to enjoy, it's just what the environment is like. If you don't like it, leave and get a new job.

Years of activism, court cases, and Title VII protection have changed that, she said:

Here we are today, and sexual harassment in the workplace is not normal. Our norms and how we understand it are different now.

Laws against sexual harassment are “a blunt instrument”, she said. The fact that there’s also a stigma attached to those who harass their coworkers has helped tremendously.

The Atlantic’s Jake Swearingen ponders whether a combination of legal action, market pressure, and societal taboo can work together to curb online harassment in a similar fashion, but he acknowledges one obvious difference between online vs. offline harassment: namely, anonymity.

Or, at least, the illusion that what one does or says online is hidden by a cloak of anonymity – an illusion that vanishes pretty quickly when you start to read a blog such as Naked Security, where headlines featuring prosecutions of trolls, revenge porn publishers and the like are far from sparse.

Granted, it can be difficult to unmask a troll: besides social media venues’ duty to protect the anonymity of users and their subsequent resistance to victims’ pleas to unmask their tormentors, targets of harassment often cite law enforcement who seem unwilling, or powerless, to protect them – some police, as victims relate, don’t even know what Twitter is.

OK. So. How about this: we rat out the trolls to their mothers.

That tactic is employed by trollslayer Mary Beard, a Cambridge professor of classics in the UK who’s endured a career’s worth of misogynistic attacks – this having having to do with the fact that she speaks her mind on TV “without being conventionally attractive”, as Yes Magazine puts it.

From the article:

[Beard] rallies her many supporters to strike back at the more vicious trolls. When she discovered an image of herself that was altered to show genitals where her face should be, she featured the image on her blog and suggested her followers flood the message board it came from with Latin poetry. Said message board was shut down soon after.

On Twitter, she'll sometimes retweet a vile comment to expose the harasser’s nasty behavior. Put into the spotlight - and occasionally shamed when Beard's followers contact trolls' mothers about their behavior - these men often end up apologizing, showing that the common troll is easily slain if you have the support of your own anti-troll army.

Take it all together, and our arsenal includes:

• Societal change that turns trolling into a stigma.
• Legal change, such as the UK’s stiffening of prison sentences.
• Software that blocks trolls.
• People who, like Charlotte Laws, persist in pursuing prosecutions against revenge-porn kings and their ilk.
• Public call-outs.
• Mum call-tos.

Add them all together, and you get a multi-pronged approach.

That seems appropriate, given that this is a beast with a thousand heads.

If we band together, and if venues afflicted with trolls choose to take the problem seriously, then maybe, someday, the most egregious of the beast’s effluence can be dammed up.

Image of Don’t feed the troll courtesy of Shutterstock.