ERROR! ADM+S workshop - 21st May at RMIT, proposals due April 10th
Christopher O'Neill
An ADM+S workshop at RMIT University (Narrm/Melbourne), with Associate Professor Mike Ananny (University of Southern California, Annenberg School of Communication and Journalism) (online keynote), Dr Mardi Reardon-Smith (Monash University) and Dr Christopher O’Neill (Monash University).
The ‘biases’ and mistakes of algorithmic culture and machine learning have often been subject to critique. In this workshop we hope to reframe this kind of critical analysis through a new investigation of the status of ‘error’ in digital technology. Specifically, we are interested in two questions:
1, What is it that error ‘does’? That is, how do contemporary technologies make errors and also use errors; how do they iteratively reprocess their own failures, how does the imperative to solve errors spur new hype and investment cycles, new distributions of authority, and new attributions of blame for when things go wrong? How do the people who work with contemporary technologies make sense of the errors that they solve, work around, and live with? And how do people deal with the effects of error on their lives and livelihoods?
2, What is it that we can ‘do’ with error? That is, how can we perhaps use error methodologically to better understand the operations of opaque digital technologies? As Mike Ananny and Simone Hodgkins have recently argued, “Errors show not only how systems succeed or fail but how people want or expect systems to do so—what counts as “good enough” to whom, when, and why.”
Our wager is that something of the ‘nature’ of error is changing. Whether we are examining machine learning, generative AI, sensor technologies, state surveillance, or digital twins, how can a focus on ‘error’ reframe and sharpen an analysis of contemporary digital power and how it is experienced by people?
We will convene a workshop at the ADM+S headquarters (Building 97, RMIT University, 106-108 Victoria Street, Carlton) on Thursday 21st May for short (roughly 10 minute) papers that take up ‘error’ as a problematic in a range of contexts and settings, including:
the politics of the ‘good enough’ and of ‘normal accidents’ in an error-stricken world;
error, tinkering and care-ful data practices;
linkrot and indexing errors in the digital archive;
parsing differences between errors, glitches, failures, and mistakes within STS, media studies, and anthropology frameworks;
errors and backpropagation in machine learning;
error, responsibility, and the role of the ‘human-in-the-loop’;
error and misclassification in the (automated) workplace;
the failures of biometric authentication and their consequence in digital infrastructure;
the use of sensing and automated technologies for environmental management;
error as a productive friction for understanding the interplay of technologies and techniques;
and more!
If interested, please submit a title and abstract (200 words) here by 10th April.