What we know about radio signals spans different domains of human sensibility and literacy. The experimental research project Radio Explorations combines machine learning tools with design and development of web-based interfaces – data observatories – to facilitate search and navigation across the digital archive of radio signal recordings collected by SIGID wiki contributors. The SIGID wiki archive holds recordings of radio transmissions, as recorded by specific people, on specific locations on Earth. Radio Explorations project gives access to this incomplete but situated archive in an organized way, to the radio enthusiasts as well as researchers interested in information studies, digital humanities or media materiality.

With an interest in difference, or similarity, between the way people and machines perform intuitive pattern recognition, the Radio Explorations project proposes visual systems that enable intentional and intuitive exploration of digital information. Concrete manifestations of radio are organized into data observatories, using an unsupervised machine learning algorithm, self-organizing map (SOM) to lay out the ground. A data observatory gives access to this organisation, while always filtering the data according to a specific interest, and intention to ask a specific question. Could we identify a signal based on its similarity to other, known signals? Which signals sound the most like birds or like hip-hop?

In addition to exploring effective ways to organise digital data, the project problematizes what is sometimes referred to as ‘digital literacy’, namely how computation and networks work and what we can learn with them. This practice of experimenting with machine learning algorithms challenges instrumental categorizations of technical artefacts, such as radio signals. Inspired by the feminist critique of technoscience, the project outputs propose to become skilled in using these advanced computational techniques differently, as one possible mode of resistance.