Part of Temporal Constructs Appendix, which contains details behind the research: dates, quotations, descriptions of instruments, and sources.
General definition
The page records how the questions of this research formed, by reading the two documents of 2025 beside the record of what the doctorate has done since. Both were written and applied for in 2025, before the doctorate began, and they aimed at different things.
Relation to the practice
The reading below is an inference drawn from those two documents and from the record of this archive, and it is marked as such. It exists because the questions the research now asks arrived over more than a year, and because a practice-based doctorate is judged on what it found as much as on what it set out to do.
Documents of 2025
The application of April 2025 to the SIA grant Artistiek en Ontwerpend Onderzoek is the outward document, Stompff, Levelling the playing field (2025). It argues from the state of a sector: generative systems strengthen the unequal position of emerging artists. The means to run them belong to whoever can invest in them. It responds with a tool for other people, an open-source artificial intelligence acting as a maker interface to MIDI. Its four work packages build that tool, learn what musicians need, demonstrate it in a performance and an album, and share what is learned.
The proposal of June 2025 is the inward document, Gardner, PD proposal (2025). It argues from a practice: creative practitioners are losing the ability to think with their tools. The collaborative principles written into the instruments are being forgotten at the moment they would be most useful.
It responds with a laboratory of the researcher’s own, mediated by artificial intelligence, and with digital heritage kept alive by use. The two documents share a subject and a concern. Both use artificial intelligence and music as their material. Both name the loss of autonomy and of skill among musicians, and both look to the instruments of the MIDI era for principles worth keeping.
Document contributions
The interface of the application became the material base of the doctorate. MIDI, its protocols and the languages of the period are what the research works with, and the MIDI era is the period its argument rests on. The consortium of the application became the network of the doctorate. V2_Lab for the Unstable Media and Thunderboom Records were partners in the application of April 2025, and both became research partners of the doctorate in late 2025.
The technical core of the application arrived ready-made. Its first work package was to build an open-source interface between a model and MIDI. What the doctorate received in April 2026 was an open-source agent framework that could write its own tools. Within weeks it was porting HMSL and writing KML into Trip-A-Tron. The thing the application would have built became the platform the practice runs on, and not the deliverable it produced.
The frame of the proposal became the frame of the research. The question the doctorate now works on is how creative agency is shared once a person and a model work inside one instrument. The pages that set out that argument come from the line of the proposal about thinking with tools and about heritage. The laboratory of the proposal became CONDUCTOR. The objective of the proposal is an audiovisual research laboratory mediated by artificial intelligence, and the system built in the first year is that laboratory in a working form.
Deferred application
The outward half of the application is the part that is deferred, and it is not abandoned. Its aim is access for musicians who cannot invest in their own systems. That aim belongs to the later modes of the plan, which are the workshops, the transfer to educational settings, and adoption by five institutions.
The archive records the deferral in its own words. The educational commitments were set aside in April 2026, because the technical conditions for responsible transfer did not yet exist. The assessment of the first year names Change agent, the role that reaches beyond the field, as the weakest of the four roles. The first occasion on the calendar is the panel with the Kunstenbond in Utrecht on 2 December 2026.
Missing capability
Neither document anticipated the capability that arrived in April 2026. The plan of 2025 was written without an agent that could build its own tools. The order in which the modes were worked is an order the plan did not set out.
The findings of the first year are not the findings either document aimed at. The interfaces the agents built, the mapping of the presets of the synthesiser and the unattended run at MU are results of the arrangement, and not of the plan. The Discoveries page contains them.
Verification
Both documents are held in full, the application in Zotero and in this archive, and the proposal in this archive. The dates, the work packages, the commitments and the partnerships stated here are taken from them. The reading of what each supplied to the present is an inference, drawn from those documents and from the pages of this archive. It is offered as a reading, and not as a fact of the record.
Unforeseen
Here the unforeseen situations, sometimes difficult, sometimes unexpectedly interesting, will be described as the project runs. Many of these unforeseen situations have already occurred and have been recorded. They will be added in the time to come.