arts · digital
the instrument that needed its maker
Bespoke instruments co-designed with disabled musicians work in the study but seldom see sustained use after it: one hackathon instrument was dismantled and its software lost within a year, and a six-month build could be used at home only when one carer was available
Source: Lucas et al., “The Longevity of Bespoke, Accessible Music Technology: A Case…” Open ↗source: primaryLast updated: 2026-11-12
Problem statement
Many disabled people cannot play conventional instruments because of how those instruments are built: the NIME authors give the example of an acoustic piano whose keys sit at a fixed 73.5 cm while a typical wheelchair armrest is about 78 cm from the floor. Because disability is often specific to the person, inclusive-music organisations such as Drake Music Project Northern Ireland (DMNI) build one-off accessible instruments with individual musicians; most of DMNI's come from short-term collaborative projects such as hackathons. The instruments work during the project, but, the NIME authors write, "it is often the case that bespoke AMT does not see sustained use." After DMNI's first hackathon in 2015, a disabled artist who wanted to keep developing the gestural instrument she had co-designed found a year later that "this bespoke AMT no longer existed. The hardware had been dissembled and the software lost." A follow-up ethnographic study built a virtual-guitar instrument with a wheelchair-using musician over more than a year (about six months of design, six of development, then two months of use); it was used at least twice a week, but "when Peter [his father, one of his primary carers] was not available, Eoin was not able to use i1; other carers did not have the requisite knowledge to facilitate its use," and, "due to time constraints," the maker could not add the one feature the facilitator suggested or the adjustments he had himself seen were needed. The unsolved problem is not designing an accessible instrument, which the field does well, but keeping a bespoke one playable, maintained and supported after its maker moves on.
Why this matters
A systematic review of 113 publications describing 83 accessible digital musical instruments (ADMIs) found that "many of the systems reviewed in the current study were on prototype stages. Reports of continuous use of the ADMI after the initial study were rare." The same shortfall affects new digital instruments generally: when Morreale and McPherson surveyed the makers of instruments presented at NIME 2010–2014 (97 contacted, 70 answered), "most of the instruments have difficulties establishing themselves." For a disabled musician this is not an inconvenience; it decides whether they can practise at home and develop as an artist, as a player of a conventional instrument would. The commercial alternative is costly: the NIME authors note that Soundbeam, one of the most prominent commercial devices, has an entry price of £2,500, "likely beyond the financial means of many disabled people within the UK," and the authors suggest financial barriers may mean relatively few disabled people own accessible music technology and practise at home. The review's positive finding stands too: most ADMIs were developed with iterative or participatory design methods, and a few (MidiGrid and MidiCreator, the RHYME project's tangible instruments, Sound=Space) appear to have been used over longer periods.
What’s been tried and why it hasn’t worked
(1) Short co-design events. DMNI's first hackathon, in 2015, paired music-technology students with disabled artists for three days; each pair produced a bespoke instrument, and the event was judged a success, but at least one of the instruments was not kept afterwards (the dismantled instrument above). (2) Longer participatory design. The 2018 study used the Human Activity Assistive Technology (HAAT) model, credited with reducing abandonment due to poor design, and spent about six months on design (planned: two) and about six months on development (planned: two), for £330 in parts. It worked, but it exposed three dependencies the authors call monetary/resource, knowledge/skill and time: the carer and the workshop facilitator had a demonstration and a short manual, yet the carer needed further support and the facilitator struggled with quirks (the file format and folder structure of the backing-track memory stick), and the authors note it was unclear whether maintenance requests would arise at all without regular contact with the researcher. The study's positive side: Eoin used the instrument at least twice a week at home, in most weekly workshops and at one public performance; only one hardware fault (a loose screw) and one software bug were recorded in two months of use; and he intended to keep using it. (3) Cheap maker hardware. A £4.50 Raspberry Pi Zero turned out to be too weak for the libraries used, and a cheap motion-sensor board proved unreliable (the authors suspected a counterfeit sensor), so development time went to troubleshooting. The authors warn against assuming "that as low-cost maker technology has the potential to remove access barriers, this will naturally happen as a matter of course." (4) A community of makers. The paper's own proposal, argued in the same authors' Computer Music Journal article (which says "the provision of technical support and ongoing maintenance is required to achieve the sustained use" of such instruments), was a shared community of practice, pointing to Drake Music England's DMLab, which still runs maker meet-ups in several English cities. The author's 2023 thesis, three case studies later, still identifies his own embedded involvement as a significant factor in the instruments' continued use, so the hand-over from maker to community is not yet solved. (5) A modular, open-source toolkit. Ward's MAMI Tech Toolkit (Frontiers in Computer Science, 2023) built four accessible tools over five years with three special-needs schools and an adult day centre, with an open-source philosophy (bill of materials, code and Max/MSP patches on GitHub), affordable components and the explicit requirement of a "kit that can stay within research sites after the research is over." It is close to the common-core idea below. The paper itself says only one tool, the Noodler, "had extensive testing" (§5.3), its limitations section says the whole toolkit "could benefit from extensive testing in context," and the author found (§5.4) barriers to finding, setting up, integrating and using such technology in sessions, each with its own skills and training needs; the paper reports no follow-up of use after the research ended.
What would unlock progress
Instruments designed for hand-over from the start. A modular, open-source common core already exists in at least one form (the MAMI Tech Toolkit), so the open part is not building a kit but showing that a kit, plus support, keeps instruments in use once the researcher leaves. The NIME authors propose standardising and modularising the parts that are common to bespoke instruments (breakout boards, open-source code libraries, a chord dictionary borrowed from another bespoke guitar), so that several people use the same underlying technology and makers have a reason to maintain it. The other half is the support side that commercial products have and bespoke ones lack: documentation and setup that a carer or facilitator who is not a technologist can use, a way to adjust the instrument on a day when the musician's range of movement is smaller, and a route to repair that does not run through one person. The volunteer prosthetics network e-NABLE, which the authors cite (some 3,300 volunteers, about 800 devices distributed), is an adjacent model of standardised designs customised per user.
How we’d know it’s being solved
The measure is use after the maker leaves: the share of bespoke instruments still played 6 and 12 months after hand-over, sessions per week at home and in workshops, and the number of sessions lost because no facilitator could set the instrument up. Secondary measures: the time a non-technical carer needs to set up and change the instrument, the number of repairs that could be done without the original maker, and how many instruments share a common hardware/software core. Beware the proxy: "instruments built" or "hackathon outputs" count the starts, not the use.
Where to start
Facility-free door: take one published bespoke instrument (the NIME paper describes Instrument One's components and interaction in detail; the Device One switch interface's code is public on GitHub) and redesign it for hand-over: a modular hardware core, a plain-language setup guide, and a "carer mode" with only the controls a facilitator needs. Test setup time and errors with volunteers who have no electronics background. Design door: start from the published MAMI Tech Toolkit (code, wiring files and documentation are public on GitHub; the paper says the bill of materials is there too, though no file of that name shows in the repository listing and the repository states no licence) and Instrument One's described components, and work out what a shared parts-and-code core still lacks for hand-over: test whether a non-technical facilitator can set up and re-map a kit tool from its documentation alone, and design the maintenance arrangement (who fixes, who updates, how faults are reported) a maker community would need to keep several instruments built on that core working. Partner door (design the trial, don't run it): with an inclusive-music organisation, write a 12-month follow-up protocol for instruments built at its next co-design event, with use after hand-over as the primary outcome. Access note: co-design with disabled musicians requires a partner organisation and ethics review; the facility-free door uses published designs and non-specialist testers instead. Relevant skills: embedded systems, interaction design, documentation and service design, music technology.
Genes
Related problems
Stuck the same way, in other fields.
field: food & the living worldnew wildlife tech must prove it works in the fieldNSF funds new conservation tools to turn research into field action, on condition they plan to prove they work, against existing tools where appropriatetested in the wrong worldsource: primaryfield: food & the living worldthe pesticide that killed the wrong insectEffective neonicotinoid pest control devastates pollinator populations it was not targetingtested in the wrong worldsource: primaryfield: health & medicinekill the larvae, spare the wormRiver blindness elimination is slowed because mass-treatment drugs do not kill adult worms, and loa loa co-infection rules out ivermectin in some areastested in the wrong worldsource: primaryfield: making & materialsai needs more bandwidth than packages can deliverThe I/O performance wall — semiconductor packaging cannot scale to meet AI compute demandtested in the wrong worldsource: primarySources
- Lucas A, Schroeder F, Ortiz M. "The Longevity of Bespoke, Accessible Music Technology: A Case for Community." NIME'20 (International Conference on New Interfaces for Musical Expression), July 21-25, 2020, Royal Birmingham Conservatoire, Birmingham City University, Birmingham, United Kingdom, pp. 243–248, accessed 2026-10-22. Link ↗
- Frid E. "Accessible Digital Musical Instruments—A Review of Musical Interfaces in Inclusive Music Practice." Multimodal Technologies and Interaction 3(3):57, 2019, accessed 2026-10-22. Link ↗
- Morreale F, McPherson A. "Design for Longevity: Ongoing Use of Instruments from NIME 2010-14." Proceedings of the International Conference on New Interfaces for Musical Expression, pp. 192–197, Copenhagen, 2017, doi:10.5281/zenodo.1176218 (full text read 2026-11-12 from the full text), accessed 2026-10-22. Link ↗
- Lucas A. "Exploring and cultivating connections within ecosystems of inclusive musicking: moving towards sustained access to music." PhD thesis, Queen's University Belfast, July 2023 (record and abstract), accessed 2026-10-22. Link ↗
- Drake Music, "DMLab", accessed 2026-10-22. Link ↗
- Ward A (2023) The development of a Modular Accessible Musical Instrument Technology Toolkit using action research. Front. Comput. Sci. 5:1113078. doi: 10.3389/fcomp.2023.1113078, accessed 2026-10-22. Link ↗
- Lucas A, Schroeder F, Ortiz M. "Enabling Communities of Practice Surrounding the Design and Use of Custom Accessible Music Technology." Computer Music Journal 44(2-3):9-23, 2020, doi:10.1162/comj_a_00567, accessed 2026-11-12. Link ↗
Also cited in the brief
Revision notes
This brief gets checked against its sources and contains the following revisions. Future revisions get added here.
Worth knowing
- The survey of 70 instrument makers covered new digital instruments for solo live performance and left out instruments made for specific ages or medical conditions, so it speaks for new instruments in general, not for accessible ones.
Full check
Sources read
- Lucas, Schroeder and Ortiz, the longevity of bespoke accessible music technology (NIME, 2020): full text
- Frid, review of accessible digital musical instruments (Multimodal Technologies and Interaction, 2019): full text
- Morreale and McPherson, design for longevity of new musical instruments (NIME, 2017): full text
- Lucas, doctoral thesis on inclusive music ecosystems (Queen's University Belfast, 2023): abstract only
- Drake Music, DMLab community page: full text
- Ward, the MAMI Tech Toolkit (Frontiers in Computer Science, 2023): full text
- Lucas, Schroeder and Ortiz, communities of practice for custom accessible music technology (Computer Music Journal, 2020): full text
What changed
- Stated more strongly than the source: that nothing at the 2015 hackathon kept the instruments afterwards; the paper documents one instrument that was not kept and says nothing about the others, and the page now says so.
- Quotations: a quotation about the six-month build now gives the paper's exact words; words shown in quotation marks as coming from the thesis abstract are not in it; the point is now stated without quotation marks.
Left open
- the volunteer and device counts for the e-NABLE hand network (taken from a study that was not read)
- that the toolkit's bill of materials is public (the paper says so, but no such file shows in its online files)
Separate check of specific claims: The same authors' journal article was added as the place their community proposal is argued, and the page now says what the toolkit's online files hold.
Earlier full check: A gloss inside a quotation, a claim that some instruments were used for years, a description of the hackathons and where a toolkit paper says what it says were corrected.
Separate check of specific claims: The long title was narrowed (display title unchanged), several claims were brought back to the paper's wording, the instrument's good results were added, and an open toolkit was added as an earlier attempt.