In Flight Music

In Flight Music represented another piece where embarking on a webification bumped up against the ever-changing nature of the web, and the fact that pieces made with one API or datasource can work one day, then fall silent the next. The OpenSky Network is largely volunteer-run and, due to demands on their system, had greatly reduced access to their API by 2026 when I came to resurrect this piece. Back when I made it for my PhD the API was completely open access and I went about querying it every 15 seconds to create a near real-time piece. Investigating my options in 2026, I found the API requests were limited per 24 hours. I’m grateful to OpenSky for granting permission to use their data for this artwork.

Thankfully this small wrinkle did not affect, too badly, the implementation of the piece as it was originally conceived for my PhD. Each aircraft which comes within range — a 0.5-degree square centred on Sheffield in this version — triggers a granulated version of an organ chord. Three organ chords in fact, one for the x, y and z (altitude) axes of the aircraft in question. The aircraft’s progress across the box makes the granulator progress through the file. The files are set up so that all the notes of the chord are in the middle, with the chord building up towards the middle of the file and then breaking down again afterwards. The ‘Hendrix’ chord (dominant seventh sharp ninth, with an unresolved, rootless quality that I liked for this piece) is built up note by note, so a plane making its way to the middle of the map box creates a dissonant stacked chord. The sound world was influenced by Jon Gibson’s Cycles, an improvisation for pipe organ characterised by slowly building and modulating chords. I liked the way the entrances and exits of notes in Cycles are sometimes barely perceptible, creating a tension between stasis and movement that seemed appropriate for this piece. The x axis is in C and the y in F; the z axis features notes from both chords played one by one as a sort of wildcard — the z axis tends to remain fixed at a cruising height, maybe only fluctuating a little.

Unlike most of the other pieces in the portfolio, there is no fixed number of voices. Each aircraft effectively becomes another sound source, so the number of simultaneous voices is determined entirely by the traffic overhead. An aircraft at the edge of the box might produce only a small fragment of the chord, while one approaching the centre produces more and more of it. With several aircraft nearby these individual sounds accumulate into a larger, sometimes quite dense sound mass.

The data accessed are deliberately sparse — I’d been to a Morton Feldman piano concert around the time I wrote it and been taken with the idea of events separated by silence. Unlike many of the pieces presented in the portfolio (and now webified), the sound isn’t continuous. The piece was partly conceived as a contrast to works such as Singing in the Wires and The Beach Buoys, where the sound is continuous and unbroken. Here the presence of an aircraft overhead provides a natural mechanism for entrances and exits. There might be several planes producing a large drone, or just one, or none at all. It can fall silent for long periods. The silence is therefore as much a part of the piece as the sound, although this is not without its pitfalls.

In Flight Music was installed as part of a Sound Junction concert and I hung around the room where it was playing. In the run-up to the concert the two (moderately) famous guest composers headlining the concert came in to see what was going on. The piece was utterly silent and I looked on as they shrugged and walked out again!

In Flight Music was a rare case of one of my pieces actually having a visual already conceived to replicate on the web version. When it was first made, testing the sound generation element was tricky without actually visualising the path of the planes across the sound axes. Using Python’s super primitive turtle graphics I ended up with a sort of radar display plotting planes. As I set about recreating this for the web I went through many ‘realistic’ approaches, with interpolated movement (surprisingly difficult!), radar sweeps spinning, black-on-green colour schemes and so on. In the end I settled for something more minimalist, so as not to be a distraction from what is, after all, a piece of sound art.

The original version of The Beach Buoys is documented in my PhD thesis.

Sheffield, September 2026