Digital devices and online services are increasingly embedded within our everyday lives. The growth in usage of these technologies has implications for environmental sustainability due to the energy demand from the underlying Internet infrastructure (e.g. communication networks, data centres). Energy efficiencies in the infrastructure are important, but they are made inconsequential by the sheer growth in the demand for data. We need to transition users’ Internet-connected practices and adapt HumanComputer Interaction (HCI) design in less demanding and more sustainable directions.
Yet it’s not clear what the most data demanding devices and online activities are in users’ lives, and how this demand can be intervened with most effectively through HCI design.

In this thesis, the issue of Internet demand is explored—uncovering how it is embedded into digital devices, online services and users’ everyday practices. Specifically, I conduct a series of experiments to understand Internet demand on mobile devices and in the home, involving: a large-scale quantitative analysis of 398 mobile devices; and a mixed-methods study involving month-long home router logging and interviews with 20 participants (nine households). Through these studies, I provide an in-depth understanding of how digital activities in users’ lives augment Internet demand (particularly through the practice of watching), and outline the roles for the HCI community and broader stakeholders (policy makers, businesses) in curtailing this demand. I then juxtapose these formative studies with design workshops involving 13 participants; these
discover how we can reduce Internet demand in ways that users may accept or even want. From this, I provide specific design recommendations for the HCI community aiming to alleviate the issue of Internet growth for concerns of sustainability, as well as holistically mitigate the negative impacts that digital devices and online services can create in users’ lives.