When Vanessa* found a Facebook group dedicated to horror and fantasy movies a couple of months ago, she was thrilled to discuss her favorite topic with nearly 5,000 people who shared her enthusiasm.
But one day not long after, she noticed that a thread had gotten off topic. One of the group's members was threatening to attack people, even throwing out the word "rape." Vanessa, 26, informed the group's administrator and the troll was banned. The next morning, Vanessa found that the troll had somehow tracked her down and sent her private messages.
In two of the messages, the troll had included personal information about Vanessa and her family -- the former mailing addresses of her father and mother, as well as Vanessa's own current phone number. In another message, the troll threatened to personally attack Vanessa, saying she was "next" and telling her to mind her own business. She still has no idea how she was singled out.
Vanessa got a call from her mother that same morning. Her mother said she'd received two voice mails at 3 a.m. The first one was just "tapping noises and breathing," but in the second, someone said Vanessa's name multiple times before asking, "What are you going to do?" Vanessa did a quick Google search and found that all of her personal information could be easily tracked down using the White Pages, so she immediately hid the information from the public and went to her local police station to file a report.
Things quieted down after that, but one day, she checked the Facebook group's page and saw a post that included pictures of her father taken from his Facebook profile, along with his old street address. It was a new account, but clearly the same individual was behind it. The poster stated that he or she had multiple accounts and wouldn't stop, telling Vanessa that he or she would go after her mother next. The post was taken down immediately, but Vanessa was understandably shaken.
"I didn't sleep the entire night," she told The Huffington Post in an email. "I just kept obsessively checking the page, terrified that a post about my mom would come next. The next morning, I publicly left the group, hoping that the troll would see and would leave me alone."
Vanessa contacted family members, asking them to heighten their privacy settings to protect themselves. All posts on Facebook include a "report" button, and Vanessa tried to reach someone from the company so that extra attention could be paid to her case, but she found it was "virtually impossible to contact a live human being from Facebook."
Then her brother's ex-girlfriend got an anonymous voice mail from someone describing Vanessa as a "cunt." So Vanessa went back to the police. The cops' only response was to tell her that there was nothing they could do -- the troll hadn't done anything technically "illegal."
"My mom nervously joked that the troll would have to kill someone before the police would get involved," said Vanessa. "The woman at the police station hesitantly confirmed that was pretty much true."
It's now been nearly a month since she's heard from her harasser, so Vanessa is hoping that he or she has lost interest. But she's still pretty scared. This is the first time she's encountered a troll, and she hadn't realized just how easy it is for an anonymous harasser to latch on to a helpless, unsuspecting victim.
If you're a woman and you've experienced online trolling, you're not alone.
After New York actress Shoshana Roberts appeared in a video about street harassment last month, she was receiving rape threats from commenters within 24 hours. Earlier this year, gaming vlogger Anita Sarkeesian -- who calls out sexism in the video game industry on her Kickstarter and her site, Feminist Frequency -- was forced to leave her home after the rape and death threats became too much to handle. Then there was Caroline Criado-Perez, who dared to suggest a female face appear on English currency, and Shaunna Lane, who was a victim of revenge porn. They, too, heard from a deluge of faceless users promising brutal sexual violence.
This is not an exhaustive list. There are many, many more names.
One need not rely on anecdotal evidence. The Pew Research Center has found that women are more likely than men to be tormented on the Internet, with about one in four women between the ages of 18 and 24 reporting being stalked or sexually harassed online -- rates two to three times higher than among men of the same age. Men were slightly more likely than women to report online harassment, but they usually experienced less severe forms of it. In fact, while a 2014 survey found that men receive more negative messages on Twitter in general, the group that received even more harassment was female journalists -- that is, some of the women who rely most on the Internet as a way to share their thoughts and simply do their jobs. Women of color and members of the LGBT community tend to face especially fierce online harassment.
In a 2013 piece for New York magazine, Eric Benson defined a troll as "an online user who posts provocative items on an Internet forum in the hope of inciting a hostile, naïve or corrective response." There are virtually no barriers to trollhood -- all you need is an Internet connection and a willingness to sling negativity -- which is part of the reason trolling has become so pervasive.
Websites are doing what they can to fight trolls, but no one's found a perfect solution.
So what's being done about this? Well, that's where things get confusing.
Different online platforms have different monitoring and reporting systems, but a determined troll can often work around them. If a troll gets blocked on Facebook, OKCupid, Twitter or Instagram, he or she can simply create a new account and go right back to harassing their target.
And even with clear policies against trolling, a site can still mess up. Take the case of Thorlaug Agustsdottir, a woman from Iceland who got into an online disagreement in late 2011 with a user of a Facebook group called "Men are better than women." Soon, a new picture had been added to the page: Agustsdottir's face, Photoshopped to look beaten and bloody, on the body of another woman. Agustsdottir immediately reported the issue to Facebook, but was told the image "does not violate Facebook’s Community Standards on hate speech, which includes posts or photos that attack a person based on their race, ethnicity, national origin, religion, sex, gender, sexual orientation, disability, or medical condition.” It wasn't until Agustsdottir took her story to the media that Facebook removed the photo and apologized for the mistake.
As many have noted, a lot of these social media companies employ a largely male workforce. About 70 percent of Facebook's employees are male, as well as 83 percent of Google's tech employees and a whopping 90 percent of Twitter's tech employees. This may be a reason some of these sites aren't more closely attuned to this particularly gendered brand of harassment. But a Facebook spokesperson quoted in an October piece for The Atlantic said that the company receives millions of complaints each day, and "it's not easy to keep up with requests."
Facebook doesn't rely only on user reports and its Community Standards to protect women. It has partnered with the National Network to End Domestic Violence to create a guide of privacy and safety features for survivors of domestic violence who are concerned about online harassment.
In a statement to HuffPost, Facebook outlined its current stance on trolling:
At Facebook we have no tolerance for hate speech or content that is threatening, or incites violence, and we will not tolerate material deemed to be directly harmful to anyone. We've built industry leading technical and human systems to encourage people using Facebook to report violations of our terms, and we have developed sophisticated tools to help our teams evaluate the reports we receive.
Human error aside, filtering systems aren't yet sophisticated enough to keep all vicious content out of inboxes or off comment boards. Some systems can be programmed to flag obscenities, but they're often unable to guard against messages where the abusive language is more subtle. (Not to mention the fact that plenty of people use curse words and obscenities in friendly messages to each other.) And since several of these platforms allow anonymity, tracing and policing the trolls can be nearly impossible.
Still, people are trying. There are third-party sites like Trolldor, which monitors trolling activity on Twitter and blacklists trolls to "combat the defenselessness" of the platform's users and encourage a "friendly" environment, as the Trolldor FAQ page puts it.
Twitter is also taking action to further decrease trolling on the platform. This month, Twitter announced that it would be collaborating with Women, Action & the Media for a project that's currently in a pilot test phase. Users can use an online form to report any trolling incidents. Twitter will then use that data to "better understand how gendered harassment functions on their platform, and to improve their responses to it."
Cahn started the site after going through a divorce and finding no online outlet where she could discuss her feelings that offered both anonymity and protection from nasty comments. On VProud, users can toggle back and forth between public and private accounts, depending on the topic, so they can decide who sees their opinions on different topics. The idea is to make it easier for users to share their honest opinions.
"I think that’s what's lacking on Facebook, frankly," Cahn told The Huffington Post. "Everybody's putting out their curated, happy lives. It's really not a place where you're going to talk about the intimate issues in your life, like parenting your kid with special needs and the frustrations around that, or your sexless marriage."
At the top of every discussion thread on VProud is a video of some kind -- either culled from the Internet, submitted by users or created by the site's editors themselves. Cahn said that her team has crafted special comment moderation technology that weeds out "trollish" comments and includes an artificially intelligent database that understands context. Users can also click a handy "troll" button (see image below) to flag any offensive comments, and Cahn said that the site's editors "live and breathe in their conversations making sure that the dialogue is productive." All new conversations are approved by editors to make sure they're free of hate.
"When we talk about 'trolls,' I want to be very clear," Cahn said. "Our definition of a troll is all about rude, hateful, mean and unproductive comments. These comments can come from men or women."
That said, Cahn wants to keep VProud women-only. As of now, the only way to ensure that no men enter the site is by constant email verification. Cahn said that VProud periodically sends emails to every single account to make sure they represent active users, and not just "a one-and-done situation" where a man has created a new Gmail account to pose as a woman. Cahn admits the system isn't airtight, but for now, that's all the site can do until a user posts a hateful comment that gets flagged by either the moderation technology or another VProud user.
Cahn said that anonymity is a huge part of the troll problem: People feel OK about harassing women when they know their words can't be traced to them. On the other hand, sites that tie their users' accounts to Facebook, or some other platform that uses people's real names, tend to see better behavior from their readers. Cahn cited Reddit and Gawker -- particularly the latter, which uses Kinja, a commenting platform that allows for near-total anonymity -- as some of the most dangerous places on the Internet for women, simply because of the lack of accountability among their users.
"Look what's going on with Gawker and Jezebel and their Kinja product, where you don't even need to have an email address and you can just go and get a Kinja account and you can just comment and say whatever you want," said Cahn. "You can pick a screen name and literally have no email, no authentication -- that is the most trollish tool out there, and the poor Jezebel editors have to live with that."
As of Oct. 29, Kinja has enacted a comment flagging system. The news was announced in a blog post, which read in part: "We kicked it off in response to the feedback we heard from our editorial staff and the greater Kinja community." Kinja also has a feature that allows users to click to see comments that have yet to be approved.
Most major news sites, like The New York Times and The Washington Post, have on-staff moderators as well as an account registering system that authenticates the email addresses associated with commenters. Even though The Huffington Post has living, breathing moderators to facilitate productive conversations, this site also struggles to deal with trolls, despite enforcing a policy in which users must tie commenting accounts to their Facebook profiles. While not foolproof, this Facebook strategy has been fairly successful for other news outlets, like the Los Angeles Times and The Business Journals.
What's clear is that no website has yet come up with a totally troll-proof solution. Finding a happy medium between "free-for-all" and censorship can be nearly impossible -- especially when you're asking your users to post their honest feelings.
When popular sites can't offer real protection against trolls, some women find alternative ways to defend themselves.
So what are women to do -- hide out in our own corner of the Internet on carefully policed niche sites? We still need to find a way to exist on popular platforms -- like major news sites, online dating pages and social apps -- that don't necessarily have rigid policies in place for combating trolls.
Enter vigilante justice.
Alexandra Tweten, an operations coordinator in Los Angeles, was particularly offended by the hostile messages she was getting on OKCupid. At some point she realized that it was happening to other women, too -- a lot of them. So she started an Instagram account for women to share their experiences and laugh about them.
"Someone [on Facebook] posted a screenshot of a message that she had gotten from someone on OKCupid," Tweten, 27, told HuffPost. "It was this guy trying to start a conversation with her. She hadn't responded and he just messaged back 12 hours later with 'asshole.'"
So Tweten decided to ask her friends to send her their similar experiences of men turning malicious, and turned it into an Instagram account called ByeFelipe. (The name is a play on "bye Felicia," a bit of dialogue from the 1995 movie "Friday" that lives on to this day as an all-purpose dismissive farewell.)
"The standard formula is when a guy hits on a girl and the girl doesn't respond or she rejects him and then he lashes out," Tweten told HuffPost.
Here are a couple of the exchanges that have been sent to ByeFelipe:
The approach of ByeFelipe seems to be: If you can't beat 'em, shame 'em. Tweten will include the offenders' photos or first names in her Instagram posts, but she won't post private information, such as phone numbers or last names. She said that the type of trolling featured on ByeFelipe happens all over the Internet, but she has focused on online dating sites because they're the "worst."
So do these sites have a responsibility to safeguard against these types of messages? Tweten isn't so sure.
"I struggle with that question, because I don't know how they could censor these messages before users get them unless someone's going to go through and read all of them or they have language alerts," she said. "I really think the problem is deeply embedded in our society. Until we change that, women are going to always get messages like this."
The most important thing women can do, Tweten said, is to not take these messages personally, since the men who lash out are probably doing the same thing to a lot of other women. A common occurrence on ByeFelipe is for a man to viciously attack the appearance of the woman who rejected him, even if he was hitting on her minutes earlier. The very same woman the man was just trying to court will suddenly have a "receding hairline and weird bent nose."
"I like to make fun of them on the Instagram, because it's like taking away that power that they think they have over women," Tweten said. "The first thing that they think they can strike with or that they think is most powerful is, 'Oh, you're ugly.' Don't take it personally. Make fun of them."
As much as the Internet enables trolls, it's also a great way to publicly embarrass them. After one male Reddit user posed as a female on a dating site earlier this year, he wrote about how he received "super aggressive," "increasingly sexual" messages within minutes -- even after telling the men he wasn't interested. Even though the Redditor went into his experiment thinking "girls have it easy on dating sites," he came away from the experience with the opposite perspective.
"I would be lying if I said it didn't get to me," he wrote. "I thought it would be some fun thing, something where I would do it and worse case scenario say 'lol I was a guy I trolle you lulz'etc. but within a 2 hour span it got me really down and I was feeling really uncomfortable with everything."
Sites like Straight White Boys Texting, Creepy White Guys and the now-defunct Nice Guys of OKCupid have turned male hostility on its head, exposing the trolls while allowing other people to get a good laugh at their expense. But since these exchanges are mostly private, some people, like Tweten, feel conflicted about shaming the message senders in public.
Figuring out whether or not we need anti-troll legislation (and what it would even look like) has proved tricky.
This brings up another question: Should online trolls be subjected to legal action? That's a delicate subject, one that Cahn herself doesn't want to discuss publicly.
"That's a great question and we have a conversation going on about that now," she told HuffPost. "I have commented privately about my views on VProud, so I'll leave it at that."
In the U.K., Justice Secretary Chris Grayling has proposed a law that would make Internet trolling punishable by face to two years in jail -- four times the current maximum sentence of six months.
"These Internet trolls are cowards who are poisoning our national life," Grayling told the Daily Mail last month. "No one would permit such venom in person, so there should be no place for it on social media."
In the U.S., there's also been some push to criminalize online trolling. California made revenge porn a crime in 2013, and states like New York and Arizona have also proposed measures to crack down on trolls.
But as former attorney Brett Snyder recently pointed out at the blog FindLaw, people who spew bile over the Internet often do so anonymously -- which would make it hard to pin down a defendant for an in-person court appearance. Snyder said that you can, however, use the power of the court to force platforms to reveal anonymous users if you can prove you've been wronged. Even Christopher Poole of 4chan, the controversial site that recently hosted a slew of leaked nude photos of celebrities, told Details magazine in 2010 that he complies with court subpoenas and provides information on anonymous users to the FBI -- at least when the alternative is going to jail.
Many types of trolling, however, don't violate any laws, as Snyder notes. If you're not threatening someone or trying to extort them, then exposing their private information or photos is only despicable -- not actually illegal. And defamation can be pretty hard to prove, since many of the offensive opinions expressed online are just that: opinions.
If you don't feel that a troll is threatening your safety**, you can go Cahn's route and find a safe corner of the Internet for the times when you want to share your thoughts on sensitive topics. But until the Internet at large finds a viable solution to combat trolls, perhaps Tweten has it right: Yes, there will always be hateful, aggressive people online (and offline), but if there's nothing you can do about it, you might as well rise above it.
"I get where they’re coming from," Tweten said of the trolls she exposes. "You can just see the desperation and the frustration they're having. But it doesn't make it OK to lash out at people."
*This name has been changed to protect the person's privacy.