Many social networks have attempted to deal with harassment and trolling on their services by offering reporting tools to users, and using automated systems to lock trolls out. However, these automated systems continually cause problems for users, especially users who wish to remain anonymous. This is the case with Twitter’s implementation of phone verification.
The problem is that Tor users are being flagged by Twitter’s new system. As far as Twitter is concerned, flagging spam or sockpuppet accounts is important for reducing undesirable content. But requiring users to provide a phone number presents a hardship for some people, resulting in their accounts being banned. Other users might simply prefer to use the site anonymously, rather than exposing themselves to government snoops when providing their phone number.
Thing is the telephone verification is unlikely stop harassment in any significant way. According to Electronic Frontier Foundation activist Parker Higgins, Twitter trolls are too diverse to be thwarted by one simple system:
Abuse on Twitter comes from accounts using real names, and from accounts using pseudonyms. It comes from accounts with massive follower bases, and from so-called ‘egg’ accounts that are freshly made. It can come from a user with a handful of sock puppets, or a small army of one troll’s dedicated fans.
Higgins also notes that Twitter, in trying to fix its problems, has fallen into a common trap – the Politician’s Syllogism. The fallacy is essentially “We must do something; this is something; therefore, we must do this.”
Both Google and Facebook tried to combat harassment with real names policies. Google was met with resistance and eventually gave up on the real names requirement. Facebook’s real name policy was similarly disastrous. Everyone from pagans, to drag queens, to professional wrestlers and Native Americans were caught up in a dragnet that was supposed to make the site more civil by forcing users to use their real name.
There may be a simple, and more elegant solution to this problem: Shadow bans. Reddit has been using this system for some time. The banned user thinks they are still able to interact, comment, vote, and post on the site, but their activity is hidden from others. Such bans could leave anonymous users in peace, while cutting retaliation, and reducing the visibility of harassment.
The trolling problem isn’t really one of access, or anonymity. It may be a demographic problem that’s currently at the forefront of the national debate. We may never be free of trolls, and punishing honest users because of bad actors isn’t the way to engender trust in your site.