What Is the Best Way to Stop Internet Trolls?
bbc.com
bbc.com
The secret to dealing with internet trolling is some combination of ignoring it, turning your devices off now and then, and realizing that the character you are interacting with online is not actually a real person. Though -- like a character in your favorite movie or TV show -- it is of course created by a real person, it is still a fictional character.
With the exception of actual actions that come into the real world, such as fake reporting to the police, interfering with someone's ability to have a job, etc, the vast vast majority of online "trolling" is solved by ignoring it. Same as it ever was.
At a higher level, though, sites and moderators of discussion have to participate in the reduction of trolling. Failure to do so dooms a community. I've left my share of communities, mailing lists, usenet message boards and the like due to it being tiring to participate in a discussion that devolves into name-calling and wasteful trolling. My favorite example was Digg about a decade ago. At some point the trolling problem got so bad that you couldn't leave a comment of any kind without encountering a land-mine. I remember the "straw that broke the camel's back" was when I spent a few minutes leaving a multi-sentence comment and had copy/pasted part of a sentence to a place where it made more sense. I had accidentally doubled an "and". I was met with "Learn fucking English, asshole!" and was like "huh?!". The guy read my entire comment and grammar trolled me over an obvious typo? I deleted my account shortly after that. I wasn't angry -- I just came to realize that the site stopped providing me any value.
Hacker News has been pretty rigid with its community standards and though there's (legitimate) complaints about the rigidity, the trade-off encoutered from a "downvote because I disagree" vs "downvote because comment is hostile" is well within my tolerances here.
90% of human interactions are crap ;) Pre-Internet people lived in isolated bubbles and the mess we have now is just a demonstration what happens when too many people from one bubble start interacting with another.
Right now, the login/password field is bugged so you have to type in the URL string to get to your account. ICU/brain dead is about right for K5 now, unfortunately.
Many media companies spend money moderating internet forums because their users and/or advertisers have expectations. Automatically detecting or predicting bad behavior decreases the cost of running these forums, whose purpose is profit rather than dialog.
It's not a change in the nature of trolling that makes moderation more necessary than it was in the past; rather, it's the monetization of forum-posting audiences.
They sign up for services, post all their details and make the sort of off-hand comments that wouldn't be picked up in close circles, except they are using a global publishing platform (social media).
USENET is dead because they couldn't keep the troll/spammer population in check. Smaller forums with more control survived because they were more resilient and didn't offer as big a target.
Usenet died because of 2 reasons: The burden of carrying alt.binaries The burden of pirate media on distributed Usenet servers from alt.binaries
They made a great point of attack for the MPAA and RIAA member associations. And traditionally, ISPs provided these services. I think it was 2006 when Comcast finally pulled the plug. Most other ISPs doesn't even offer a connection at all, which is a shame.
But that's why it died.
The infamous "Green Card Spammers" (http://www.wired.com/1999/04/the-spam-that-started-it-all/) shifted the tone dramatically, and from there it was inevitable that the whole community would collapse. There was very limited control over what people could post, USENET was intended to be free and open. This succeeded wildly in an academic environment where if someone was seriously out of line you could address their university and get their account pulled.
As soon as the general population came on and there was no accountability to anyone it was simply a matter of time before unchecked abuse started to happen.
Most providers pulled the plug because the only thing of any value left on USENET was binaries. The discussions were virtually dead.
What you are missing is that human beings are social creatures, and our well-being is enormously dependent on how well our communities work. Every community has norms of behavior and ways of enforcing them. If trolling and cyberbullying are allowed, then either people in the community suffer unnecessarily, or the community collapses.
Let me add that you are disempowering yourself when you don't try to ensure that your communities are good.
Unfortunately, far more cases are less extreme than yours. The line gets blurrier when you get down to persistent low-level harassment, where individual examples will look relatively harmless to an outsider but the victim is forced to face them as a whole. In those instances, it's the responsibility of the social media platform to allow the victim to properly document and report the harassers, and most definitely put up the appropriate barriers to prevent future harassment. Twitter's biggest failing on this point is in how easy it is to create "sockpuppet" accounts, such that as soon as one is banned they can continue uninterrupted from the next.
Where was that said? I don't see this comment at all. "turning your devices off now and then" is akin to stepping out of a discussion when you start to take things personally. That doesn't mean you are banned from all conversations. It's healthy to take a break.
This isn't about someone "[taking] things personally" -- this is about organized groups making existing platforms unusable for their targets.
Who is targeted by that and why? Who organizes that - I mean, why would thousands of people suddenly have reason to hate you?
Criado Perez started a petition to ask Bank of England to consider including a woman on the notes, and to remember to include women in future re-designs.
That's not much.
For that she was subjected to months long campaigns across all her social media accounts, extending to her AFK addresses, with severe threats of sexual violence.
One man was sending 50 messages an hour over 12 hours across multiple accounts.
It's disappointing that some people are defending those who make death threats. It's common over a keyboard, but AFK a death threat is serious business.
https://www.judiciary.gov.uk/wp-content/uploads/JCO/Document...
> Here there were a series of communications (not just a single one as referred to in the guideline) where extreme language was used and substantial fear was caused.
> Indeed it is hard to imagine more extreme threats (death, rape, and worse [Sorley]) and rape [Nimmo]. The harm is, and (despite what is said on your behalf) must have been intended to be, very high.
> The fact that they were anonymous heightened the fear. The victims had no way of knowing how dangerous the people making the threats were, whether they had just come out of prison, or how to recognise and avoid them if they came across them in public.
And about blocking:
> The messages were posted on a number of different sites, and continued after being blocked and or warned
As I said in response to your other post, this is something you report to the police. And I bet she did.
And about blocking:
> The messages were posted on a number of different sites, and continued after being blocked and or warned*
In that case, it's not really "organized" by anyone in particular.
For example, in your analogy, they're not being manhandled out of the venue. Someone at the venue is taking their picture and following them home, distributing that information, then random people are showing up at their house for the next eighteen months or more. Then people are showing up at their family's house with threats, and their boss's house demanding that they be fired.
Is this the type of "shit [flying]" that you think people should be prepared to expect when they post things online?
(Btw yes, that sort of reactionary persecution can happen and does indeed happen in all sorts of places. It's part and parcel of meaningful political activism anywhere in the world.)
Why should the consequences be so completely out of proportion to what was originally said? "In public" covers more situations than it used to, and the size of your audience isn't always intuitive.
> Can I walk around Harlem with a racist sign without expecting to be roughed up at some point?
Probably not. Will you be roughed up by thousands, be followed home, and be stalked and harassed for years? Probably not.
The fact that "in public" situations might have expanded (and sometimes replaced more private ones) is a problem that can (and should) be fixed with education. NBD.
Just because some people seem unable to comprehend the (massive) scale of internet space, it doesn't mean that we can magically make it a paradise where real-world behaviour honed in thousands of years simply does not happen. Internet space is a new all-encompassing dimension, the scale of which we've never seen. One should think carefully before acting in such an environment. It's not like interacting with press or peers in a given niche; it's like broadcasting on all channels in all countries at all hours. The chances of getting reactions you might not like are higher than with any other media or venue; but on the positive side, the chances of getting reactions you will like are also disproportionately higher than anywhere else. Continuously complaining that the glass is half empty will just provoke someone in smashing the glass to solve the problem -- because "you didn't like it much anyway, always complaining".
Because you're a woman on the internet, and said something.
In the good old days, people used to send you such threats by anonymous snail mail instead of twitter.
It's probably a crime, yes. Unfortunately, you'll be prioritized accordingly. When prioritizing a crime that involves some anonymous troll responding with a death-threat or threat of harm that was most likely done in a moment of stupidity/weakness rather than with an actual intent to commit a crime, you're going to show up pretty low against the list of rapes, murders, assaults and thefts that are currently being handled. Once you've identified that the threat is such that it actually is a crime and can be successfully prosecuted, you have to go through the trouble of finding out who actually left that threat. Should they successfully uncover that information, the likely penalty might be so minimal that it's not even worth pursuing. Because of all of those circumstances, all but the most extreme cases of these get completely ignored (and many of the extreme ones).
But that's probably (definitely) a terrible thing to do, so maybe it's better to find other ways to solve these problems.
Furthermore, from some anecdotal evidence, I think most victims in such cases already know who is harassing them.
Some online behavior that crosses into the real world will and has always been illegal and actionable of course, as I noted in the original comment. Most of what people consider "online abuse" isn't in that category.
If you don't want to see rippling abs and cleavage you have to avoid nearly all advertising these days. If you don't like guns you can't watch movie trailers. The modern world has many such dilemmas. Ignoring the things you find objectionable shouldn't be discounted as a simple solution.
As I said, I realize I'm out of the mainstream. But I've been called horrible things and even doxxed and stalked and threatened in the real world by online commenters a couple times over the past 20 years. I've found that "not caring about it" has been a shockingly useful and effective tool.
For younger kids today, a overwhelming majority of their lives are online, so it's so much harder to escape - both for work/education purposes and entertainment, so it's so much easier to keep up the waves of harrassment, so simply saying "just ignore it" doesn't really work these days.
Kick the jerks. Ban them if they are repeat offenders. Provide mechanisms for people to shrink or expand how much information is shared and how easy it is to contact them.
I'm actually dismayed that to this day very, very few sites enforce a system where you must be recommended by someone in order to join, or by a friend to ask to be friends/communicate with someone.
Those that get perpetually harassed would be far less likely to have to deal with trolls if those trolls had to be a "friend of a friend". Is vouching so technically complicated we haven't done it yet?
There are social structures and conventions that haven't really seen much traction in software. Someone's totally out of control at a party? Kick them out and kick out the person that invited them too.
I know services like Slack are kinda helping this along a little bit - I'm a part of one or two Slack groups that I've been invited to by other people, with a bit of a pep-talk at the start of like "here's the sort of shit that WON'T fly here..." that are just small single-interest focussed groups that have kinda organically grown, but it's definitely a rare situation.
I think it's just education for new users (kids), and making sure they fully understand how the internet works and being mindful of what information they share where and with whom. Trolls (in all their levels of extremity) will always exist - anyone claiming they know how to "get rid" of them are fooling themselves
If people feel like they have tools to push back against trolls they'll use them. If they feel like they're powerless, they'll leave. If they leave the troll to user ratio skews a bit higher and might end up poisoning the entire community.
Which is right?
Thousands of reports are made to police each year. http://www.bbc.co.uk/news/uk-england-london-24160004
> About 2,000 crimes related to online abuse are being reported to the police in London each year, according to new figures.
Over a thousand reach the threshold for court action: http://www.bbc.co.uk/news/technology-23502291
> More than 1,700 cases involving abusive messages sent online or via text message reached English and Welsh courts in 2012, the BBC has learned after a Freedom of Information request.
The criado-perez trolls were enabled by the Internet. They saw other people trolling and thought it was okay; they were able to create many accounts rapidly; they were able to send very many messages.
The hardest part of dealing with online harassment and "trolls" seems to be drawing that line, or even acknowledging that it exists.
Since when does that help? http://www.adistantsoil.com/2011/12/18/stalked-someones-watc... http://www.themarysue.com/zoe-quinn-drops-case/
No. Gamergate has proven that "ignoring it" and "turning your devices off" is NOT the way to fix it. Neither of those things does anything.
This solution seems to have worked for some people :)
The problem with is then @alice controls whose tweets I can read if I want to follow her content. :/ Just because I want to hear what @alice has to say doesn't mean I want @alice to control whether I can talk to @bob.
I think solid ignoring features is really the limit to this approach due to that fact. [e.g. @alice ignores @bob, @bob never shows up in her feed no matter what]
I agree with you, the main point is that @alice doesn't have to smell @bob's shit anymore.
The point isn't to nail it down exactly right here and now. The point is that this is a discussion that should have taken place many times at Twitter, and we should see some concrete results by now, rather than fluffy PR-focused fake-internal memos to the effect of, "ok guys now we gotta try hard!"
That could work well enough. :)
Yeah, my main concern was a power user who people wouldn't want to unfollow abusing the feature to silence critics.
Personally, I think it starts first with community standards and moderation. Hacker News, Stack Overflow and such do a pretty good job already. You get a share of jerks here and there but by and large the community self-filters the tone[2] as much as the quality of the comments.
When the inevitable trolls appears, one of the best approaches I've seen is to silence them without their knowing it. I wonder how often this practice is figured out and remediated by creating a new account. It'd be possible to extend this further by categorizing troll posts and selectively providing them to each other (as well as providing all replies) so that they think "it's working" and don't bother switching accounts, all the while hiding all of the trash from everyone else. Perhaps that's a lot of wasted energy but I'd be interested in hearing if something like that has been tried anywhere.
[1] That word always seems to describe the behavior best, to me. I'll use "jerk" from now on. :)
[2] I've always believed that you can say "someone/something is wrong on the internet!" without having to be a jerk about it and have seen many examples of such on Hacker News and the StackExchange family of sites. Those who cannot lack creativity and are, unfortunately, less likely to have their (possibly) valid arguments heard.
Shadowbanning does work well but requires (somewhat obviously) fair admins/moderators. Otherwise it becomes something like a censorship tool.
Good fun.
Joining a dungeon in World of Warcraft and wiping the group is trolling.
Sending someone genuine sounding death threats on twitter isn't.
There's a spectrum there somewhere, but more often than not it feels like we've lost another useful term to hyperbole.
I think, though, that it's a pretty huge difference. The speed with which someone can crap all over a discussion is instantaneous and psychologically a lot easier due to not being "in person". Trolling in the form that it would have existed a couple of centuries ago[1] still exists today (without the sword, perhaps, but the threat of violence in retaliation to trolling still exists today -- we just have better technology and better enforcement). I would use Donald Trump as an example. He gets his share of trolling (and non-trolling protesting) at his in-person events, however, I'd hazard a guess that the amount of such activity during his entire campaign would be about as much as is seen in an hour on Reddit.
[1] We don't have a lot of historical evidence that dates back 2500 years, so I'm assuming that was a typo or an embellishment used to make a point (or, perhaps, something I missed in my speed reading of the article).
I'd rather just live with the trolls. Is it that bad?
And losing anonymity also reduces someone's willingness to talk openly (and civilly) about a topic that they may otherwise be unwilling to discuss (sexuality, medical issues, etc). Heck, I won't even talk directly about my political opinion after reading articles about how people with strong political views are willing to reject a candidate for hiring entirely because of them not aligning politically (not really a problem for me since I generally hate politics).
This also lowers the number of available voices and it might reduce the ratio as it relates to trolls. In a sensitive topic do we lose more trolls this way or do we lose more civil conversation? I'm not sure.
Elimination of anonymity helps, yes, but isn't appropriate for every kind of discussion and must be combined with other methods.
See the Sikh who advertised Facebook ("What's this raghead terrorist doing on my wall?") or comments under any news article (especially Daily Fucking Mail ("Kill the immigrants")).
Anonymous apps tend to bring out the worst in people, especially with groups like teenagers and cyberbullying.
Seems kind of insane that people are willing to dox themselves on the internet.
1. Disagreement is inherent to a lot of discourse. Social networks are not research papers -- some of the disagreement is going to be accompanied with strong displays of emotion. Where does the line between strong disagreement and bullying lie? If Anne tells Bob, "Bullshit, your arguments don't make any sense and you are just an arrogant cyberbro I DRINK MALE TEARS", is she cyberbullying him? Should her access to Twitter be cut off, or would voluntarily blocking her on the part of Bob address this problem in a sufficient manner?
2. How to balance prevention of trolling with limiting self-expression? The outside world is not a perfect place where everyone engages in reasoned discourse either. Given that pub-sub systems like Twitter are the equivalent of everyone shouting their opinions loudly in the same street, how much of the unreasonableness inherent in human discourse do we allow to penetrate these social networks to keep them essentially human in nature and not turn into a LinkedIn style list of platitudes?
I wish the article went some ways towards addressing these. Instead it seems to focus on ad-hoc initiatives set up by a lot of people who (in my opinion) seem to not have any concrete ideas of what they are trying to accomplish. For example, one of the research papers linked [1] describes cyberbullying as including "...posting mean or hateful comments, aggressive captions or hashtags". In my opinion this is a rather extravagant definition of the term and should be examined more carefully and revised per the points above.
--------
[1] "Detection of Cyberbullying Incidents on the Instagram Social Network" http://arxiv.org/pdf/1503.03909v1.pdf
Gregory Alan Elliot is drawing that line right now in Canada: https://en.wikipedia.org/wiki/R_v_Elliott
It's a disingenuous argument at best.
That is an awful misrepresentation.
The paragraph in question basically pushes the conversation toward, "Where are the brightlines?"
What would this be like? Before seeing something, you'd be asked if you want to see "An advertisement for a home care product." Or "An angry personal attack" (Maybe with some indication of who it's against.)
This solution allows everyone to say whatever they want, and adds the ability for everyone to not be exposed to what they don't want to be exposed to.
How do you know I want to be exposed to a consent related question?
Kinda like how asking certain questions about consent can be classified as harassment if you have already been told no, even if the actions of which consent was being requested for were never carried out.
I don't use Twitter and likely never will. It's a cesspool of discrimination and abuse. I hope it dies as a medium.
Unfortunately, this has a few issues...
1. You need the manpower to actually check what's being posted on a regular basis. Or enough active members willing to help out that they'll do it for you on a voluntary basis as moderators. This is obviously easier for small sites like customer support forums and the average fan site.
2, You need to really, really limit bias as much as possible, so that you're booting out people or bots because of their behaviour rather than their political/social stances. This is what Twitter doesn't do at the moment, to the point being a 'progressive' troll is a good way to never get banned while being a conservative or having an unpopular opinion will get your account suspended for the smallest trangression.
3. A willingness to ban popular, rich or 'powerful' users as much as anyone else, if they break the rules. A good forum/community manager will just as easily kick out a troll with ten thousand posts as one with ten, but social media sites don't seem to have figured this out yet.
So for a community manager or site owner, moderation is the best way to stop internet trolls (and every other type of troublemaker or nuisance as well).
Making a user's post invisible to everyone but himself (shadowbanning or as vBulletin called it, Tachy goes to Coventry) and making their time on the site a nightmare due to fake errors (hell banning/miserable users) used to work. Note the word used to, since the internet is a lot more savvy about this sort of stuff and will usually figure out these systems pretty quickly now.
As for how to avoid internet trolls as a user? Well, stay off large social sites with few rules (or in some cases, extremely badly enforced ones) and go to places where the people tend to be a bit nicer to each other due to a shared interest or two. Less Twitter, more forums.
Stay fairly anonymous wherever possible. Seriously, trolls love knowing every detail about someone's offline existence and what annoys or upsets them. By giving less personal information, you force people to either REALLY dig for information to use or to attack your arguments rather than your background/appearance/personality.
Don't engage in arguments with trolls/idiots. The more you make it seem to affect you, the more pleasure a troll will get from their actions. If you've ever watched The Simpsons, you probably remember the scenes where Bart prank calls Moe's Tavern, right? Or how often, Moe would basically scream back down the phone in anger afterwards? That's what a troll wants to see. They want people who are easy to annoy.
So yeah. Moderated community platforms tend to be the best way to stop internet trolls, since they mean such users and comments are removed as quickly as possible.
This is a political non-starter right now, which is why Twitter doesn't do it. Seriously, when they accidentally ban someone for threatening violence against other Twitter users in the name of left-wing activism a huge swathe of the tech and security community launches a campaign to reinstate them and accuse Twitter of all matters of evils fro banning them in the first place.
http://freethoughtblogs.com/almostdiamonds/2012/02/28/dont-f...
It's an interesting idea (and I'm 100% sure it's been done before), but it doesn't really work out on any site with a significant audience. I mean, all you need is one of those trolls to access the site at work/school/wherever and notice their posts are being shown if they're not logged in, and then suspicions are raised.
Too easy to bypass and figure out. It's like how people figured out Twitter's shadowbans, because people who followed them said off site that they weren't seeing their messages.
The article hopes to offer an advance in troll detection technology based more on reason than hope. ;-)
At some point I think a wet team is the only solution.
There is such a thing as people being presented with content that's irrelevant to their interests.
I better way to semantically express what we need, as well as a recommendation system that goes with it, is what we lack today.
The people who matter to trolls are those who can't take a joke, who are super serious all the time; they want to be mad about something. Trolls merely give those people what they want. Flame wars ensue. Purposeful trolls are social pranksters and are conscious of their trolling; these can be reasoned with and are thus not serious issues to fora.
The actual trolls worth worrying about are those who troll yet know not their trollish behavior. These people begin arguments over trivialities and get called jerks/dicks/assholes. These people are much harder to reason with, because they will only listen when they are emotionally sober; bringing up their trolling just gets them deep into another argument. You can't stop trolls; they have to grow out of their own emotional immaturity.
Thus, as the unconscious trolls are themselves a subset of those harmed by trolling, they can and do harm others without understanding what harm they cause.