Social networks must learn the difference between user experience and user welfare, writes Niall Kitson.

Last weekend we saw social media at its worst. Its kicking, shrieking, indifferent, head-in-the-sand worst. Two incompatible value systems - happy clappy Twitter and a rabble of 'trolls' sending hate messages for kicks - hiding behind the same free speech defence to absolve themselves of responsibility for a sustained campaign of abuse that, in the 'real world' would have resulted in wholesale arrests for harassment, stalking, and attempted sexual assault.

Following a successful campaign spearheaded by Caroline Criado-Perez to have Jane Austen adopted as the new face of the £10 note, the freelance journalist and feminist blogger became inundated with rape threats over a 24-hour period. At one stage Criado-Perez was receiving up to 50 threats per hour from what she later referred to as a "nest of men who co-ordinate attacks on women". The abuse has tapered off but at the time of writing, the campaign has spread to include MP Stella Creasy and Laura Bates of the Everday Sexism Project.

Whether it was a co-ordinated attack or a random swarm of the bitter and the easily lead, the content of the tweets points to a disturbing culture of misogyny and a 'gamification' of offence where the winning is in inflicting the greatest emotional damage with the least possible effort. Twitter, with its 100 million strong echochamber and emphasis on users’ being responsible for the content they post (the ‘safe harbour’ defence adopted by YouTube in the past to avoid copyright litigation), is the perfect arena for this behaviour - and so it was last weekend.

When Criado-Perez brought the threats against her to the attention of Twitter she was told to go through the usual channels for reporting objectionable content - filling out a detailed form on a support page. With abuse running at a rate of 50 messages an hour this route was just not feasible. Attempts to contact Twitter's manager of journalism and news Mark S. Luckie (based in New York) resulted in the locking down of his account and blocking Criado-Perez. Luckie unlocked his account yesterday morning and thanked his followers for their "support" after he was "singled out by a group in the UK". Luckie - the author of a book on digital journalism - made no mention of the controversy and has not engaged with users who messaged him about it.

It was up to the Twitter's user base to fight back against the trolls. Criado-Perez posted regular screen grabs of threatening tweets, created the hashtag #shoutingback to rally support and a petition demanding Twitter act by introducing a 'report tweet' button similar to that on the service's current iOS app. At time of writing the petition had reached nearly 60,000 signatures and Tony Wang, head of Twitter UK, confirmed that such a measure was already being tested. In the meantime, journalist and author Caitlin Moran has suggested everyone take a “trolliday” on 4 August and avoid Twitter for a full 24 hours. Good luck with that.

That Twitter had to be forced into acting by a groundswell of anger was bad enough, but to dismiss the severity of Criado-Perez' situation by asking her to fill out a standard form and wait for something to happen showed how completely unprepared Twitter was to deal with the situation.


According to Tony Wang, head of Twitter UK, a report button is being tested and should be rolled out in the near future. In May, Facebook was forced into adopting a more assertive stance on trolling and hate speech on its platform after a series of misogynistic groups began posting image memes making fun of domestic violence.

Social networks' main error in dealing with hate speech is to prioritise user experience (ie making it easy and pleasant to say things) over user welfare (ensuring users aren’t damaged by content on the service) - assuming people will use their platforms responsibly. The rise of cyberbullying has proven this is simply not the case and as services gain momentum, they have to evolve to protect individuals from anonymous, unjustified personal attacks., for example, have managed to avoid problems with cyberbullying by adopting a simple policy of “don’t be a dick” enforced by a community of moderators. It’s an unsophisticated approach that works. Newspaper websites like the Irish Times use self-moderation for below-the-line comments which are also monitored by a community manager creating a dual-reporting system. If forums and newspapers can be this proactive, why have social networks been dragging their feet?

In the absence of a system that works for all, it’s best to look at what users can do to protect themselves. According to child welfare groups like the best way to deal with online harassment is to log all incidents and report them to the authorities, not to engage with trolls and block offenders. Criado-Perez has opted to log and make a statement to police but is being vocal about the abuse. This won't work for everyone but who knows, maybe the twittersphere needs to engage in some troll shaming in the absence of a fit for purpose reporting function to make such behaviour socially unacceptable.

As an addendum, the first arrest in relation to the weekend's events was made on Sunday in Manchester. A man has been bailed until September.

Niall Kitson is editor of