Manufacturing the Quantified Self

An Ethnography Inside a Chinese Fitness App

Resumo

The adoption of fitness apps for tracking daily activity has become increasingly normalized and routinized among a growing public. The data generated by tracking devices and apps enable forms of self-observation in which mundane activities are rendered measurable, comparable, and visible. The quantified self, as a human-algorithm assemblage, emerges through the constant interplay among app developers, users, and software infrastructures. Because the quantified self is made visible through dashboards, visualizations, and other forms of materialization, it has become relatively accessible to researchers; many studies have examined this phenomenon and evaluated its implications for society and for individuals who participate in lifelogging practices. Yet the public rarely has the opportunity to move beyond the interface and glimpse the interplay among actors and software infrastructures that co-produce the datafied self. Following Latour's actor-network theory, this paper investigates the algorithmic black box through a six-month ethnography of a Chinese fitness app that allows users to log daily workouts, body measurements, and caloric intake. It examines the inconspicuous but complex procedures through which the final data product is produced. Through the lens of the duality of translation, the paper explores how heterogeneous actants - including patents, workflow applications, workplace chat applications, and user feedback tables - are interwoven in the manufacture of the quantified self.