If algorithms are doing more and more seeing, then what distinguishes human engagement with images? How to resist the “object vision” dominant in algorithmic image recognition?

Algorithmic vision primarily perceives formal aspects of images – such as the tangible facts they depict – but struggles with abstract, context-dependent meanings. Despite these limitations, it is increasingly used to navigate vast collections of cultural artifacts. In Dataset of Loss, Hess explores this tension by capturing instances of loss in institutional archives and everyday life. The organising principle of the images in the publication – the theme of loss – is not a label stating what they depict. Instead, the images ask questions about origin and content while resisting formal categorisation.

The publication’s concertina format emphasises how images influence one another. Arranged in close proximity, they are not perceived as isolated facts but as a chain reaction of meaning.

Fabienne Hess is an artist working with textiles and images, reflecting on interdependence, ways of seeing, and memory. She’s graduated from the Royal College of Art in London and is an associate lecturer at the University of the Arts, London.
www.fabiennehess.com

Roland Früh is the co-director of the library at SIK-ISEA in Zurich and teaches research and writing at the Type Design Master at ECAL, Lausanne. After studying art history at the University of Zurich, he has worked for publishers in Zurich and London and has been responsible for the Art Library at Sitterwerk from 2014 to 2022.

Severin Rüegg currently oversees the collections of the SKKG (Stiftung für Kunst, Kultur und Geschichte) in Winterthur, a foundation dedicated to exploring innovative, participative ways of engaging the public with Bruno Stefanini’s collection of (almost countless) historical artifacts. Severin Rüegg has studied history and film studies at the University of Zurich.