live coding // musical instruments // July 2026

phd

What makes a good live-coding musical instrument, and what does it mean to practise a programmable system? My research asks how instrumental understanding, intuition, and action can become embodied, fluent, and musically absorbing when the instrument itself expects and encourages transformation, even over the span of a single performance.

abstract

This practice-research PhD investigates the design and practice of live-coding musical instruments: programmable systems whose vocabulary, representations, mappings, and operative logic remain open to change while they are being used. It asks both what can make such an instrument good and what it means to practise and gradually master one when transformation is not an exceptional interruption but an expected part of the practice, potentially within a single performance. The central problem is plasticity versus sedimentation: how instrumental understanding, intuition, and action can become embodied, fluent, and musically absorbing without sealing the conditions of future action against revision.

The research combines research-through-design with longitudinal practitioner inquiry and performance-as-research across TimeLines, uSEQ/ModuLisp, and an emerging gamepad structural editor. Selective genealogy and comparison supply counterexamples and test the resulting vocabulary. The proposed ergo triad— ergologics, ergonomics, and ergodynamics—traces how operative logic, distributed labour, and perceived potentialities feed back into one another through practice. The intended contribution is an account of mutable instrumental mastery and design principles for sustaining musical agency, embodied fluency, and meta-instrumental change over short and long timescales.

the_question

“Good” is not a context-free property of an artefact; instruments are learned over time, appropriated by particular people and cultures, and evaluated in relation to musical, social, and aesthetic purposes. I am therefore not seeking a universal checklist or a single measure of flexibility, but rather a framework for understanding how design, implementation, practice, and perceived potentialities feed back into one another and shape a player's instrumental practice over both short and long timescales.

A good live-coding instrument sustains durable but revisable coordination among player-relevant computational logic, embodied instrumental fluency, and musical agency. It supports conditions for flow while allowing performance to transform some of the conditions of its own continuation; each change creates the environment within which the next one may arise.

The narrower centre is how live modification of player-relevant operative logic mediates the relationship among instrumental fluency, musical agency, and meta-instrumental agency. The central tension is not freedom versus constraint or code versus body, but plasticity versus sedimentation: action must become stable and trainable without sealing the instrument's future action space against revision.

This makes mastery itself a research problem. How does a player cultivate durable skill in an instrument whose vocabulary, representations, mappings, abstractions, and temporal rules are not merely available for revision, but actively expected to change as part of the practice? What kind of mastery can include changing the instrument without continually destroying the fluency already developed with it, and how do the instrument's design and implementation facilitate or challenge this?

why_live_coding_matters

A universal computer, in the sense described by Alan Turing, is not one fixed machine; it can interpret descriptions of other machines and temporarily assume their organisation and functionality. Alan Kay and Adele Goldberg's metamedium carries this from formal universality into personal and interactive computing: the computer can incorporate existing media and support media not yet invented.

Universality does not automatically make a computer transparent or malleable to a person. Practical plasticity is achieved through languages, representations, interfaces, and feedback. Live coding makes some of that mutability performable: a musician can change the current sound and the structures through which later actions will acquire consequences. Those changes must remain meaningful while time, sound, bodies, skill, and expectations continue to move.

three_questions_not_one

live_coding

Operative representations of musically consequential computation are modified within the temporal horizon and stakes of an ongoing activity, leaving consequences available for situated interpretation and response.

instrumentality

An enacted and cultivated relation among computational substrate, current configuration, learned practice, player, culture, and situated performance—not a property of software alone.

goodness

A situated judgment that names its player, practice, values, evidence, situation, and timescale. Bad and unlearnable live-coding instruments must remain conceptually possible.

Musical practice has an autotelic dimension: the quality of playing, exploring, practising, and improvising matters, not only the efficient production of a predetermined result. Situated goodness is not arbitrary, however. My commitments include sustained and developing musical agency, accountable action and consequence, appropriation, recoverable risk, and the formation and revision of skilled practice. Flexibility alone is not goodness; useful constraints give a possibility space a shape within which to evolve.

second-order_instrumental_action

A conventional instrumental action changes the sounding result while leaving the instrument's action space mostly intact. Live coding can also change the representations, mappings, abstractions, or temporal rules through which later actions will be made. The performer acts within the instrument and, to some degree, on it.

The performer “completes the circuit” by configuring, learning, and enacting a substrate already full of temporal commitments, musical theories, permissions, and prohibitions. Completion does not mean closure. The boundary between lutherie and playing remains mobile: a performed idiom can become a library or interface, while a stable abstraction can be reopened. To design a live-coding instrument is also to practise it.

practising_a_mutable_instrument

Making, practising, and performing are related but distinct parts of the inquiry. Making introduces a change and externalises a design proposition. Sustained practice is where routines become reliable, frictions recur, mappings sediment, and previously unavailable possibilities are progressively disclosed. Performance tests that developing practice within musical, temporal, technical, and social stakes.

Mastery here cannot mean only command of a settled vocabulary. It may also involve knowing which parts of an instrument to stabilise, which to reopen, how to migrate learned action across a change, and how to move skilfully between absorbed playing and reflective reconstruction. This is not mastery despite mutability, but mastery partly constituted by the capacity to manage mutability without losing musical continuity.

three_coupled_capacities

instrumental_fluency

Reliable action with this instrument: its mappings, routines, abstractions, failure modes, and possibilities. Practice makes actions available without every component remaining in focal attention.

musical_agency

The capacity to hear, imagine, choose, enact, and evaluate musical possibilities. It includes timing, form, phrasing, listening, and aesthetic judgment; technical command alone does not guarantee it.

meta-instrumental_agency

The capacity to alter the instrument's vocabulary, mappings, abstractions, or action space. It can extend musical agency while disrupting accumulated instrumental fluency.

These capacities are analytically distinct but developmentally coupled. Practice chunks action, changes perception, redistributes attention, and discloses previously unavailable musical aims. Fluency is both an achievement and a possible constraint because it can sediment assumptions until they become difficult to perceive or question.

embodied_fluency_and_flow

“Muscle memory” is shorthand for a wider achievement involving proprioception, tactile expectation, perceptual chunking, timing, prediction, recovery, and attention. Physical interfaces can give computational action stable spatial and tactile form—several fingers poised over possible actions, movement guided by touch rather than visual search—but embodiment develops through practice, not from physicality alone.

Flow is a central experiential stake: deep absorption in which action and awareness can become less divided, often supported by an answerable relation among challenge, skill, goals, and feedback. It is not an artefact property or a permanent target. Musical flow, operational editing flow, and reflective meta-instrumental work can align or compete. The design problem is to support skilful transitions among absorbed playing, reconfiguration, failure, and return.

a_revisable_sensorimotor_grammar

I have begun building a gamepad-controlled structural editor for Lisp. Lisp is particularly useful because actions can operate on explicit program structure and relationships rather than only on a stream of characters.

The aim is not to design one supposedly optimal mapping and fix it in stone. The gamepad supplies a bounded set of tactile controls; the editor supplies structural operations and ways to compose them. Together they form a playground in which a player can explore the materials and cultivate a sensorimotor grammar appropriate to their instrument, body, and practice.

sedimentation
Stable physical primitives and recurring structural actions can become composable units of muscle memory.
appropriation
The player forms a personally meaningful grammar rather than merely memorising designer-fixed mappings.
plasticity
The grammar remains revisable as practice discloses friction, possibility, and changing musical aims.

This is an emerging case of embodied meta-instrumental fluency: structural reach need not be accessed only through discursive text entry; the means of navigating and transforming code can themselves become physical, trainable, and player-configurable.

how_the_project_arrived_here

  1. 2018–2019

    TimeLines: signals of time

    A Haskell live-coding eDSL influenced by TidalCycles but organised around time → value. Continuous signals controlled SuperCollider parameters, establishing “music as algebra over time.”

  2. 2019–2021

    From expressions to languages

    Attempts to change TimeLines' notation exposed the eDSL boundary: I could provide building blocks for expressions, but not easily the building blocks from which players might construct new languages.

  3. 2023–2025

    uSEQ and ModuLisp

    Lisp supplied language plasticity; Eurorack supplied a tactile, spatial, electrical, and fault-tolerant grammar. The two threads converged in a purpose-built live-coding module.

  4. 2026

    Longitudinal reconstruction and embodied editing

    uSEQ's three execution regimes provide a case for tracing which changes reached practice. The gamepad editor now makes revisable muscle memory and flow explicit constructive concerns.

the_ergo_triad

ergologics
Player-relevant operative logic: representations, semantics, temporal rules, abstractions, and modes of modification.
ergonomics
The distribution of physical, perceptual, and cognitive labour among performer, interface, notation, runtime, and other agents.
ergodynamics
The potential, resistance, characteristic feel, and possibility that progressively disclose themselves through practice.

These are reciprocal relations, not independent scores. Operative logic conditions action; repeated action becomes practice; practice discloses possibility and resistance; those discoveries motivate retention or redesign of operative logic. The gamepad case compresses all three into one interface.

working_research_questions

  1. Constitution. How is instrumentality enacted across computational substrate, instrumental configuration, learned practice, culture, and performance situation?
  2. Practice and mastery. What does it mean to practise and gradually master a live-coding instrument as a programmable system whose operative logic and action space are expected to remain open to revision, including during performance?
  3. Coupling. How do player-relevant operative logics support instrument-specific and embodied fluency, and how does that fluency enable or constrain musical agency and flow?
  4. Design. How can live-coding systems preserve enough continuity for fluency to develop while leaving future action possibilities open to performative reconstruction?

research_method

genealogy_and_comparison

A selective history and varied corpus of systems and boundary cases, used to supply contrasts and pressure-test the vocabulary rather than narrate a path towards my own instruments.

making

Research-through-design with TimeLines, uSEQ/ModuLisp, and the gamepad editor, using artefacts and iterations to investigate propositions about operative logic, fluency, and musical action.

practising_and_performing

Longitudinal practitioner inquiry into how routines sediment, resistance and possibility are progressively disclosed, and mastery develops or is disrupted as the instrument changes.

reflection_and_analysis

Reflexive analysis tracing design intervention → changed condition of action → practice or performance → consequence → redesign, while considering rival explanations and the limits of first-person evidence.

This is practice research combining research-through-design, longitudinal practitioner inquiry, and performance-as-research. I occupy the related roles of instrument designer, practising player, performer, and researcher. Earlier projects are reconstructed retrospectively through versioned artefacts, source history, specifications, recordings, and reflective records; current work can be documented prospectively through design rationales, practice logs, performance recordings, interaction traces, immediate reflection, and stimulated recall.

The comparative chart is descriptive before evaluative. Its dimensions are not quantities to maximise or collapse into a score. A profile becomes normative only relative to a named player, practice, telos, situation, timescale, and body of evidence. Different records support different claims: implementation traces establish what changed, while claims about fluency, attention, flow, or musical agency require evidence from sustained practice and situated performance. Constructive cases propose mechanisms; sequence alone does not establish causality. Other players and hostile comparative cases provide external challenge rather than simple validation.

expected_contributions

instrumentality
An enacted account spanning substrate, configuration, practice, culture, and situated performance.
explanation
The ergo triad as an account of reciprocal relations among operative logic, distributed labour, and disclosed potential.
practice
An account of mutable instrumental mastery: how practising can sediment reliable action while cultivating the capacity to revise, migrate, and reconstruct it.
evidence
Longitudinal constructive cases from TimeLines, uSEQ/ModuLisp, and embodied structural editing.
design
Principles for embodied meta-instrumental fluency, including revisable sensorimotor grammars that can sustain flow without becoming fixed.

evidence_and_route_ahead

evidence_available

  • TimeLines implementation, SuperCollider integration, editor, demonstrations, and design history.
  • uSEQ/ModuLisp hardware, language prototypes, performances, and three runtime architectures.
  • Two peer-reviewed ICLC papers and a public talk on the TimeLines-to-Clojure trajectory.
  • Source history, specifications, benchmarks, diagnostics, rehearsal traces, and reflective records.
  • The developing gamepad structural editor and its evolving mapping history.

work_ahead

  • Build the selective genealogy and varied comparative corpus.
  • Trace specific design changes through conditions of action to observed consequences and rival explanations.
  • Document current design, practice, and performance prospectively while keeping retrospective reconstruction methodologically distinct.
  • Study mutable mastery, embodied fluency, and flow longitudinally, involving other performers as an external challenge where feasible.
  • Test the vocabulary against closed, opaque, collective, gestural, and non-Lisp-centred cases.
  • Separate general live-coding claims from my situated normative programme.

current_phase_and_open_questions

The project is moving from exploratory construction towards consolidation. The immediate task is to turn accumulated practice into bounded, traceable evidence, document present work prospectively, and determine which claims each form of evidence can support.

The final ontology and evaluative theory remain open: whether live coding is a strict category or family of practices; whether ergologics earns its place beside semantics, mental models, cognitive dimensions, and epistemic tools; how flow can be investigated without reducing it to smoothness; and which claims survive beyond my own instrument-building trajectory.

intellectual_anchors

selected_outputs

project_at_a_glance

programme
Part-time PhD, approximately halfway through
institution
University of Sussex
research_home
Experimental Music Technologies Lab

contact

d.kyriakoudis@sussex.ac.uk github.com/monkey-w1n5t0n