Sonic Hypermirror: Attuning to Hyperobjects

Abstract

Long-running pandemic without an end in sight, climate crisis encroaching on our everyday lives—global crises are collective events, but they take on multiple forms and scales, leading to radically different experiences for people. The inter-scalar, inter-temporal representations gained dire urgency due to the crises surfacing simultaneously at a global scale. Hyperobject, as defined by ecological philosopher Timothy Mor- ton, is the in-experienceable object that is vastly distributed in time and space that easily exceeds human’s perceptive capability. I start with a hypothesis: hyperobjects are better heard than seen. This thesis is focused on the critical approach to data representation, by bringing forward listening as a primary modality of interaction. I present Sonic Hypermirror, a custom tool that allows data probing of large-scale audio data based on vocal interaction, accompanied by a visual interface that utilizes computational tools to assemble a soft, continuous semantic space of multiple audio streams. It is an experiment to build a data sensorium where the listeners enter into, inhabit, and learn from. Through the thesis, I propose the system of data representation that is continuous, non-referential, and exploratory; and revisit the affordances of architectural space as a data storage and an interactive datascape.