Trolling Is Not Just a Art. It Is an Science: The Role of Automated Affective Content Screening in Regulating Digital Media and Reducing Risk of Trauma

Introduction

The convergence of media content is posing challenges in terms of protecting the vulnerable while also protecting free speech. Regulating abusive online media content, such as Internet trolling, is not suited to 'before-the-fact' model of current film rating agencies even if the objective rules and judgements used by those agencies are. Basing rating on age appropriateness is also ineffective for the online world, as people will have different maturity for different content at different ages with some being more prone to offense than others.

Internet trolling as a concept has transformed in definition in recent years from classical trolling, which was the posting of messages in a friendly way, to Anonymous Trolling, which is posted to harm others (Bishop, 2014b; Phillips, 2011). Internet trolling messages posted to entertain others can be seen as 'kudos trolling' and those designed to harm others can be seen as 'flame trolling' (Bishop, 2012b). But this does not mean all flame trolling is 'bad' and should be punished, nor does it mean all kudos trolling is 'good' and should be allowed. In some forums on the Internet, flame trolling is encouraged, such as the criticism of politicians, bankers, or other people who may be part of a group with dislikeable qualities. The consensual nature of these forums it could be argued should not mean their abusive comments should be prosecutable (Starmer, 2013). Equally not all kudos trolling is designed to be in the interests of people. For instance, a type of online community user called a chatroom bob, will often post friendly comments in order to seduce others. They may be a pervert looking to coax naked pictures out of the person, or a sex predator trying to groom a child or other young person (Bishop, 2012c; Jansen & James, 1995; Jansen & James, 2002). A recent example of this was the case of Daniel Perry. Daniel Perry was a 17-year-old man who was tormented into killing himself as part of an online plot to extort money from him. Perry was a popular teenager from Dunfermline, Fife, who took his own life after being targeted by a group of Internet trollers who exploited him for their own gain. Perry took his own life after a Skype conversation with a person he was led to believe was a girl the same age as him, where he shared sexualized videos of himself online known as ‘sexting’ or ‘getting naked on cam’ (GNOC). Suddenly a gang then hijacked the chat and threatened to show the video to his family and friends, unless he paid them menaces (i.e. money). Known as ‘Nigerian Chatroom Bobs,’ due to an urban myth that it is mostly people in Nigeria who extort money from people online, these trollers will use all means to gain the confidence of someone and then extort money from them, which in Daniel Perry’s case was attempted through blackmail.

It is therefore necessary to have a more technical way of looking at Internet trolling and other online misdemeanours, such as through linguistic or other forms of studying media (Bishop, 2014b; Hardaker, 2013; Hardaker, 2010). This could involve making it easier to regulate online content, so that mature users know what to expect, and Internet security software providers can better produce software with parental controls to avoid the corruption of the minds of those lacking in maturity (Haravuori, Suomalainen, Berg, Kiviruusu, & Marttunen, 2011; Roche, 2012). Such people may not have been exposed to severe or traumatic content in the past and as a result be less able to cope with it (Dutta‐Bergman, 2006). It may therefore be appropriate to refer to those Internet trolling messages which are prosecutable as electronic message faults (EMFls) and those which are not prosecutable as electronic message freedoms (EMFrs).