To Protect & Serve: Enabling Responsible Software Engineering for Online Safety

Software products have been at the heart of the explosion of social computing technologies, but they are also implicated in facilitating online violence against women and girls. In this context, there is an acute need for responsible approaches to software engineering that help minimise the risks of online harms through effective consideration of how harms could arise in the context of a particular software product and selection of appropriate mitigations to minimise the likelihood and impact of these harms. In this paper, we highlight some of the key challenges faced by software engineering teams who want to ensure that software products serve the needs of users, while also protecting these users from harms. To address these challenges, we propose that there is an essential pre- requisite to develop and maintain a knowledge model of online harms. Such a knowledge model could be used by software engineers to explore how online harms link to relevant policy and regulation, the actors involved in online harm incidents; other relevant contextual factors; and product features. We propose a meta-model for capturing this knowledge and identify some key research challenges that need to be addressed for software development teams to practice responsible software engineering for online safety.