Lines of Flight

Lines of Flight is a series of volumetric digital poems generated using tracked flying data from the author's gliding flights over the course of 2023/24. This data is visualised using three.js and parsed using a Word2Vec language model that converts the tracking points into vector queries for generating interrelated strings of words (as derived from a training set of early works of aviation history). This technique is characterised in the piece as a form of "Aeolian AI" - a speculative depiction of what AI might represent if it was not so expressly rooted in the kinds of extractive approaches that characterise it currently.

The essence of this project is an attempt at depicting the human and more-than-human registers of atmospheric flight, while gesturing towards the greater contextualising histories, infrastructures, and environments that entwine them. A running critical commentary is provided alongside the poems, as well as a full bibliography.

Lines of Flight has been inspired by my own, very particular outlook on the airborne encounter as a glider pilot, as well as an artist and an academic in digital culture - an uncommon space in which my ability to negotiate regularly with the fluctuating sky is a critical matter of flight safety, as well as bringing a range of visceral experiences into dialogue with my conceptual awareness.