Rethinking the notion of hosting in the aftermath of Delfi: Shifting from liability to responsibility?

Liability of Internet intermediaries for a third-party content is a complex topic, especially with regards to the storage of illegal or harmful postings offered by portals. The E-Commerce Directive offered a liberal framework for handling such cases, provided that a hosting service provider has not played an active role in content management. Being passive turned out to be the key precondition for immunity under safe harbour provisions. Yet, after the Delfi ruling the legal landscape has changed radically. Although the judgment of the Strasbourg tribunal has been dismissed in some jurisdictions as an error or one-off case, the truth is that it took into account acquis communautaire and imposed liability on the news portal, which followed the guidelines of Google France and eBay rulings. Given the lack of predictability of the current legal framework, the aim of this contribution is to offer a deep-dive into the notion of hosting from a technical perspective in order to better understand why Articles 14–15 of the E-Commerce Directive may require a re-examination. It is also submitted that portals and other online service providers relying on a broad construction of safe harbours should be entitled to Good Samaritan protection akin to section 230 of the American Communications Decency Act in order not to hold them liable for being active in fighting hate speech and other forms of illegal and harmful conduct.     

When do people pay attention? Violence and non-violence in political movements and the differential media attention provided

Media can dictate what people see, but an individual’s choice in media consumption can also determine what stories the media report. This paper demonstrates that people show a greater interest in large-scale protests when the use of violence is employed, though this trend does not strongly hold in every case. Google search Trend data are analysed across five recent political movements to determine at which points in a movement media attention peaks. The data are compared to the timelines of these protests with a specific emphasis on when major violent events occurred. The level of violence that occurs in a movement is potentially correlated with how many times the region experiencing the movement is searched. Some movements however do a good job at generating attention without violence, and violence does not guarantee a large audience. The Google Trend data provide valuable information about what causes people to pay attention to world events and can be used to analyse political movements and potentially make predictions about violent and non-violent conflict.