Dr. Brandon Farnsworth
Executive Summary
Background: The European live music sector operates without systematic data, a key barrier to evidence-based cultural policy and sustainable artist development. While the tracking of revenue from recorded music has now become well-established, live performances are a major blind spot in our understanding of the music industry. This ‘data gap’ makes it difficult to understand artists’ local and international touring patterns, identify musical communities, and assess the impact of funding programmes.
This report presents the results from the Tourbus Pilot project, a new approach to structuring and analysing information about live music performances. By gathering information from public websites, and processing it using AI-based tools, this approach builds a ‘bottom-up’ map of artists and venues at a scale previously impossible. It differentiates itself from existing mapping and archiving approaches through its low cost and high scalability, which make it feasible to undertake large-scale music sector data analysis. The pilot validated that the method can create comprehensive, accurate data about where musicians perform, with whom they collaborate, and how they move between venues and countries. Tourbus was developed by Brandon Farnsworth based on musicological research, and tested in a project with Art Music Denmark.
The live performances that are the backbone of most musicians’ careers remains unmapped and unmeasured
The project collected data from Denmark, Finland, Iceland, Norway, and Sweden. Analysing ca. 500 000 web pages, it identified approx. 1 600 venues across, exceeding experts’ knowledge of the sector. Copenhagen was found to have the most venues where live music takes place in the Nordics at 147 unique venues, nearly double that of Stockholm. Surprisingly, many performance venues were found outside major cities, reaffirming the potential viability of more sustainable regional and ‘slow’ touring. The data shows an expected recency bias toward events from 2024-2025, and some potential false positives remain, but represent a trade-off with the ‘bottom-up’ ability to capture occasional and unconventional venues that traditional approaches miss.
Data was used to create a social graph connecting the people, groups, venues, and festivals active in Nordic musical life today across multiple countries and genres. This made it possible to analyse the flow of artists between countries. For example, the greatest amount of artistic exchange occurred between Denmark and Sweden. This social graph also revealed how musicians tend to group together into larger communities around certain kinds of music, finding distinct jazz and experimental music communities stretching across the Nordics.
The pilot demonstrated the feasibility of bottom-up live mapping of live music events, as well as its potential applications to augment expert decision-making in the music sector through access to new quantitative tools and data. Looking ahead, this approach can be expanded internationally, and to more performing arts such as dance and theatre at a fraction of the cost of existing approaches, with more comprehensive results. With appropriate investment, Tourbus can provide the data infrastructure to help create an integrated European music market that supports artistic excellence, sustainable careers, and evidence-based cultural policy.
Music Export
- Target the right people and venues in new markets using data intelligence, strengthening the pan-European domestic music market
- Measure the impact of various kinds of export support on artists’ careers
Funding Bodies
- Access detailed information about artists’ careers and performance history, helping to give more context to funding applications and ensure fairness of outcome
- See what musical communities exist in your region
Artists and Venues
- Artists can be recommended venues in new cities that match their style, reducing booking costs and supporting regional circuits
- Enable data-driven collaboration through “unlikely venue networks” that program similar artists but do not know each other yet.
Introduction
Tourbus is a software system that uses AI-powered data collection to create and analyse systematic data about when and how live music happens. This report details the findings of the Tourbus Pilot project, which aimed to validate the method’s viability using real industry data from the Øresund region. The project was a collaboration between musicologist Brandon Farnsworth, Art Music Denmark, the Danish music support organisation, with assistance from Lund University, Sweden. Data collection and processing took place in Spring 2025, culminating in a final public presentation in April 2025 organised by Lund University’s Centre for Øresund Studies and Art Music Denmark, held at the residence of the Swedish Ambassador to Denmark.
This report briefly outlines the significance and use cases for collecting this kind of data by surveying existing reports and research. This is followed by an introduction to the case study and how data was collected. After discussing how data was collected, it demonstrates findings through a series of visualisations and discusses current limitations of this approach.
We set out to solve the challenge of collecting data about when, where, and with whom live music happens. Whilethere are many experts with a fantastic overview of various areas of music-making, this lack of data means there are no reliable ways to identify patterns that exceed our own experience. This lack of the ability to undertake a ‘distant reading’ of live musical activity makes live music strangely unknowable at a certain scale. This is a barrier to understanding what venues close or survive, how touring patterns evolve, what communities of musicians exist, who moves between them, what spaces they share, etc. Answering questions on this scale becomes particularly urgent when faced with challenges that impact live music across many different communities, such as how live music has recovered post-pandemic, the impact of changing cultural policies, or the impact of inflation (see Baumol’s cost disease). Recent music industry reports demonstrate that this lack of data is widespread, finding that “there is no aggregated data on the pan-European live music sector listing the value of the market, the number and size of venues and shows, number of festivals, share of European artists”[i], and arguing that “the absence of comprehensive data analytics for live music creates a significant barrier to a functioning pan-European music market”[ii] (2019 “Music Moves Europe: A European Music Export Strategy”).
The absence of systematic data collection renders live music unknowable at scale. This prevents the music sector from understanding itself: what venues close or survive, how touring patterns evolve, what communities of musicians exist and how they interconnect, etc. Right now, what should be measurable information is mostly available as small mapping studies, with most relying on anecdotal impressions and intuition. This is an obstacle to quantifying the impact of funding programmes, internationalisation and music export strategies, and changes in musical activity, such as during the ongoing process of pandemic recovery. Recent music industry reports confirm this, finding that “there is no aggregated data on the pan-European live music sector listing the value of the market, the number and size of venues and shows, number of festivals, share of European artists”[iii], and arguing that “the absence of comprehensive data analytics for live music creates a significant barrier to a functioning pan-European music market”[iv].
Existing approaches to systematic data collection about live music face cost and siloing challenges. The challenge of any mapping process is the work time needed to undertake data entry, cleaning, as well as to maintain it over time as new performances occur, changing the map. The high labour cost of doing this work, combined with the delimited nature of project-based funding have led to an archipelago of frozen in time datasets. In contrast, music rights organisations such as the German GEMA or the Swedish STIM have successfully developed extensive bottom-up approaches, where performers and organisers have monetarily and legal incentives to self-report concert activity to the central authority. Yet data from these organisations is also siloed by country, and limited by what is considered and reported as live music activity. Finally, this data is not made accessible outside the organisations which collect it.
Tourbus Pilot closes this gap by applying digital humanities methods to studying the live music sector. The pilottested the viability of using a combination of web scraping and AI models to create data. The approach consists of the web-crawling of born-digital sources about music performances on public websites (those belonging to artists and ensembles, venues, and festivals), and extracting semi-structured metadata about musical performances into a centralised database. This metadata provides a source of detailed information about who is performing where, and with whom. It is regularly updated in order to advertise upcoming concerts and to document accomplished work, available for a wide variety of international artists of many different professional backgrounds, and therefore a potentially suitable source of live music data. This approach mirrors similar approaches currently being developed in other fields such as economic history and library science.[v]
The high level of detail of this data makes new forms of live music analysis possible. By viewing the connections between people and places as a proxy for musical activity, they can be used to reveal the structure of musical life. This approach draws on the concept of ‘music worlds’ proposed by Bottero and Crossley.[vi] Viewing music as a dynamic social network allows for the application of computational social network analysis methods. This approach facilitates the analysis of network structure and artists’ positions within it to further the understanding of the physical and social spaces they share, and how these relations shape distinct-yet-interconnected music worlds. Drawing on the concept of art worlds also acknowledges that the affordances of physical spaces are fundamental to music communities. These spaces are the sites where social interactions happen, resources are shared, connections are made, and influence is exchanged. Following the trajectories of artists who perform in different spaces can therefore reveal the functional social maps that make up music worlds, revealing the many diverse spaces where music is performed, rather than only those most established.
Analysing live music in terms of networks is also suitable for developing a more complex understanding of how musical genre continues to influence music-making. Belonging to any specific genre becomes slippery and fraught when subjected to rigorous study. Popular music studies of genre show that music’s claims to be ‘post-genre’, coinciding with audiences’ increasingly omnivorous musical tastes, are misleading.[vii] While tastes may have broadened, persistent forms of intersectional discrimination can recreate genre divisions by regulating along racialised lines what artists are ‘permitted’ to be post-genre[viii]. Early quantitative research on this topic supports this interpretation: a recent study of reviews from Pitchfork.com found that ‘Black artist’ was used so frequently as a descriptor that it functioned as a genre category itself[ix].
Data Collection
Data was collected from public websites, including the websites of music venues, festivals, and artists’ individual portfolios. Websites were found by crawling links, starting from a seed of 400 websites and following links up to two domains removed from the seed. This led to us find a total of about 500 000 webpages forming the raw data. The initial seed URLs were compiled from multiple sources to make sure we had a balance between the two countries. Art Music Denmark provided a list of websites of artists who had applied to their funding programmes. This was combined with websites related to the 203 active members of the Swedish group Konstmusiksystrar (Sisters in Contemporary Music), the subject of Tourbus’ very first pilot, as well as a handful of venue webpages added manually.
The pilot focused on musical activity in Denmark and Sweden, with certain visualisations widened to include Iceland, Finland, and Norway. This delineated a manageable geographical area for the pilot to test the feasibility of the approach. The longstanding interest in connection across the Nordics allowed us to build on existing research to ground our expectations. For example, a recent report from the Nordic Council of Ministers found that the Øresund region we focused on is home to the largest number of cross-border workers in the cultural and creative industries in the Nordics, emphasising its central importance in the Nordic region.[i]1 Such studies also helped illustrate the added value of this new approach: the Nordic Council’s report relied on tax and register data to measure exchange, resulting in a picture limited to Nordic taxpayers, and lacking more detailed data on groups, repertoire, venues, etc. We argue that in order to fully understand musical exchange in a region like the Øresund, an international and more content-aware view is needed. Such a method would reveal how the many different professional musical communities existing in this region are connected to influences, audiences, and markets in a broader European and global perspective.
Data Processing
The data was processed using the Tourbus software platform I developed. The system used an AI model to extract details from the webpages, including venues, dates, and times of performances, performers’ roles and groups they are members of. The data was extracted as a social graph, allowing for advanced analysis of musical communities and flows of artists based on the principles of social network analysis and music world analysis, as detailed in the findings.
The processed data shows an expected bias towards recent events, with most performance events found in the years 2023-2025. This is logical given the source of the data. There is substantial data going back to about 2011 before dropping off. This supports the idea that an ongoing collection of data would be ideal for maintaining a detailed record of live music activity. This likely could have been mitigated by having access to Facebook Event data that Meta does not make readily accessible.
Basic data cleaning was applied to the dataset before analysis and visualisation. This included methods for deduplication and error correction. The social graph was also used for data cleaning: false positive nodes were often clustered together (hotels, sports arenas, restaurants without a clear connection to live music) and could therefore be easily eliminated through human review.
Developing appropriate metrics is an ongoing challenge. The novelty of the Tourbus method means that developing useful metrics for measuring live music activity is a subject for future research. Drawing from existing research on music world and social network analysis allows us to build on a solid theoretical foundation, but the appropriate algorithms and quantitative metrics still need to be developed together with the field.
Representing the Network
Our method uses network analysis to represent entities like artists and venues as ‘nodes’ and their relationships as ‘edges’. For example, in Figure 1, a line connects two venues on a map if the same artist has performed at both. This visualises artist travel patterns at a high level. In the findings section, this is also scaled by bundling all venues in a country to compare the flow of artists between countries, using line thickness to represent the number of shared artists.
For all visualisations created for the report, when an artist is part of a group such as an ensemble or band, this has been represented by disaggregating the group into its (listed) members and registering them as individual connections. This ensures the same units are used throughout. However, note that in this pilot, large ensembles, specifically orchestras with a large and generally fixed number of members, have not been disaggregated. Because the main focus of this pilot has been to test this method, these inconsistencies have been left as they are and will be fixed as the system itself improves. Studying how groups form and break apart is also possible using this data, and opens another interesting area of study.
A further limitation of this pilot was the decision to make undirected connections, meaning that the visualisations do not say anything about whether a ’Swedish’ artist travels to Denmark or vice versa. This was done to avoid the problem of determining algorithmically where people are ‘from’. Gender has been excluded for similar reasons.
Analysing connections using social network analysis algorithms reveals communities consisting of strong interconnections that are significant for analysis. When artists frequently collaborate or when venues frequently programme different members of a small group, this is represented by a higher number of connections within this small group than with the rest of the network. Using community detection algorithms, it is possible to identify clusters of increased connection between a small group. Communities are significant because they begin to reveal similarities that can tell us a lot about what different music worlds exist, and how they are interconnected. As we zoom out, we could imagine that the experimental music world is connected to, but somewhat distinct from, classical music,[i] that jazz is somewhat of a distinct group but has people who ‘crossover’ to other genres, etc. But it also can be a way of identifying disconnection: if there are two distinct communities of experimental musicians with little overlap, then this could represent a failure to adequately exchange between geographical regions, or else a faction who disagrees with a more broadly accepted style and is developing their own unique approach. It should be noted, though, that the final option is that a disconnection exists because of a lack of data, which can have a significant effect on network analysis. This means that community analysis should always be taken with a large grain of salt: it is very helpful for identifying large-scale tendencies, but its precise boundaries should never be taken as absolute fact.
Data Quality
The data has a recency bias because it relies on publicly available data at the time of the snapshot. Older information about music events has likely either been taken offline, or buried deeper in website archives than the crawler reached. While access to Facebook Event data would have been a strategy for mitigating this recency bias, access to this data is limited by Meta and therefore it was not able to be processed for Tourbus Pilot.
Data cleaning remained a challenge. Close inspection of the dataset revealed some potential false positives like a gas station in Denmark or a forest in Småland. This can be attributed to two main factors. First, errors in data processing can lead to locations irrelevant for the performing arts to be ingested as venues. We expect the lessons learned during Tourbus Pilot to help us lower the false positive rate in the next iteration. Second, there are many occasional venues that do in fact host music events. We were often surprised when looking up venues we thought were false positives on the web, finding they did in fact occasionally host concerts (such as the forest in Småland). These can range from bars, restaurants and hotels to museums and gallery spaces, and highlights how musical activity extends beyond the most central and well-known venues. As will be demonstrated in the section on community analysis, this is likely pronounced within experimental music, which often makes use of eclectic and unusual venues to hold concerts.
There is a persistent gap between quantitative network analysis and lived professional experience. Those with existing references and experiences in live music can be understandably resistant when presented with empirical data about live music. In the final public presentation of Tourbus Pilot, it was demonstrated that the dataset contains a total 61 artists who have performed in both Malmö and Copenhagen in the past 10 years. An established Danish organiser argued this was impossible, as in their career they had personally programmed more Swedish artists than that at their venue, and that the total number surely must be higher. This illustrates the core tension between data that is highly accurate but can always be more comprehensive and detailed, and professional experience developed over many years, but which cannot be fully quantified.
Findings
The findings from Tourbus Pilot demonstrate the comprehensiveness of the collected dataset, and reveal a first sketch of musical mobility between Denmark and Sweden, as well as across the Nordics. In total, ca. 1 600 venues were identified across the Nordic countries where music takes place, linked by artists performing across the region. This section explores this data from numerous perspectives to reveal the depth of data collected by Tourbus Pilot.
The first graphic shows the number of venues identified per country. We have chosen to include all Nordic countries to demonstrate the breadth of results, but note that because data collection focused on Denmark and Sweden, this is where data will be the most comprehensive.
We can further break this down to the number of venues per city, filtered here to the top 15 cities. The number of venues identified in Copenhagen is striking: at 147 it is nearly double the amount found in Stockholm, demonstrating its centrality in the Nordic live music landscape. While the number of venues in cities at the bottom of the list is low, there is a long tail of cities outside the top 15 with just one or two detected venues.
Number of venues per city
The number of venues outside of the major centres is demonstrated by geocoding and plotting venues on a map, allowing for more in-depth exploration. This reveals the expected density of venues in major urban centres and the somewhat unexpectedly high number of performance venues in Copenhagen. It also demonstrates the surprising spread of venues across the Nordics not concentrated only in major cities. It should be noted that this does not say anything about the kind of music being performed, only that performances have likely taken place there.
Venues plotted on Map
The next map demonstrates how the Tourbus Pilot has also constructed a social graph that can be used to link venues and performances together by matching artists’ names. In the next map, each arc connects two venues in the dataset, and represents the number of distinct artists who have performed in both places. While this creates a very noisy graph, it demonstrates the data that can be used to analyse how artists move across the continent as they perform at different venues over their career.
Flows between Denmark, Sweden, Norway
These arcs can be simplified into a chord diagram to visualise how artists move between countries. The diagram demonstrates that the strongest interconnection was found between Denmark and Sweden. This matches the Nordic Council of Ministers’ findings that Sweden, the largest country in the Nordics and located at its centre, dominates the flow of cultural and creative workers[x]. However, it should be noted that because data collection began with lists from these countries, more research would be needed to confirm these preliminary findings. This chart is valuable for demonstrating the kind of high level analysis made possible by the Tourbus Pilot. In a future study, this could be used to help inform music export policy, for example by identifying countries or areas with which the flow of musicians could be increased, and then measuring how these flows develop over time to measure the effect of policy changes.
Art Worlds and Community Detection
The social graph linking artists and the venues they perform at can also be used to identify clusters of activity where a subset of artists has all performed at a number of shared venues. Following music world analysis, the hypothesis is that specific venues are important in the creation of distinctly separate music worlds, and that these distinct clusters could be used to advance our understanding of the interrelationship between social structures and musical genres.
These clusters are identified using ’community detection’ algorithms that are designed to find groups within a network that are more connected to each other than to the rest of the network. The clearest cluster detected was a number of densely connected jazz-related venues who programmed a large number of artists who have also performed at other jazz-related events, creating a strong community cluster. It is also interesting that this cluster spans multiple countries, meaning that artists are likely travelling internationally with some frequency for performances. The following diagram shows venue names that have been identified in this cluster.
The data is less conclusive in the identification of one single set of venues that host experimental music. Only when decreasing the sensitivity of the community detection algorithm could a coherent set of venues be potentially identified. This is caused either by a lack of data, or possibly because experimental music often takes place in venues normally used for other purposes, as mentioned earlier. A future study could test this hypothesis by constructing an artist to artist network based on shared event participation and investigating how tightly connected such an experimental music network is.
Several of the communities could not be easily or directly attributed to one specific genre of music-making. We should note that labelling of a cluster as related to ‘jazz’ is only based on venue names, artist profiles, and event titles, and did not consider discrete musical characteristics. Within other clusters, the absence of a clear genre label may indicate emergent scenes that operate outside of conventional classifications, that lack the established vocabulary to describe them, or simply artists working in existing styles that do not have many contacts to other Nordic countries. Another possibility is that there is some kind of general ‘folk/pop’ category that is not easily assigned a specific genre designation. Examining these clusters in light of discussions about post-genre musical practices is an interesting potential area of future research. Community-based analysis has promising applications for changing how we understand musical genre and musical exchange. Knowing how genre labels can be used for inclusion as well as exclusion, it becomes possible to quantify how totally separate music worlds can coexist just a few streets apart, even in small cities.
Community detection findings can also be used for networking and building connections. By combining these findings with human expertise and domain knowledge, it is possible to identify separate communities that stylistically may have much in common but which because of a lack of exchange have remained isolated from each other. Such insights can then be used to inform concert programming by drawing exciting voices from totally different communities, to better determine whose career could benefit from a residency period by seeing if they will be connecting to an entirely new context or continuing existing collaborations. It can even be used to programmatically create recommendations for where artists could find relevant venues to perform at, whose audiences are likely receptive and interested in their fresh voice.
Conclusions
The goal of this report has been to demonstrate the potential of this new method for systematic data collection and analysis of the music sector. The report has demonstrated the feasibility of this method by detailing these preliminary findings based on a snapshot of performing arts data focused on the Nordic countries.
It was found that Tourbus Pilot far surpassed individual expert domain knowledge about where musical performances happen and how they are interconnected. The method is highly suitable for identifying venues where music takes place, including occasional venues as well as those outside of major urban centres.
The data was found to have a recency bias, with the number of events detected dropping off after one year. Challenges arose in data cleaning, with trade-offs made between data quality and quantity. The next version of the software is expected to significantly improve data quality by applying these lessons learned.
Analysing how artists’ performances connected different venues confirmed the strong interconnection and artistic exchange between Denmark and Sweden, and agreed with existing findings from other studies (but because of the limitations of the pilot did not confirm) that this may be the most significant area of musical exchange in the Nordic countries. Finally, the use of community detection algorithms demonstrated it was possible to identify a set of music venues distinctly dedicated to jazz, as well as a cluster of venues more generally connected to contemporary and classical music.
Tourbus Pilot also demonstrates how AI can close the European music sector’s data gap. The method works, scales affordably, and produces insights impossible through traditional approaches. The next step is to expand data collection across Europe and to work with the music sector on the tools for building an integrated European music market.
These new methods make possible more complex forms of analysis that can augment expert decision-making in live music. Gaining access to new quantitative tools can help areas such as tour planning and coproduction development, artist support and mentoring, music policy, and music export. This brings with it the potential to solve major challenges facing the sector, enabling experts to step out of our own shoes and get an overview of how cultural exchange is working ‘at a distance’ and across more events than one person could ever feasibly listen to.
Music export can be transformed by using network structure to enhance decision-making, transforming areas like artist touring, co-production, and block booking. Algorithmic recommendations for where artists perform can draw from a larger amount of data than individual experts, and can recommend venues around the world. Venues can also discover new artists or potentially find artists already planning to perform nearby and fill ‘dark days’ by adding to a tour. This can be combined with adopting social innovations around networking from other performing arts such as dance or theatre, where co-produced tours and block booking seem to be more widespread practices, at least in Europe.
The effects of cultural policy changes can be measured by examining the uptake or impact of funding programmes, showing the effects of subsidy on the number of live arts accessible in given regions, or by highlighting the importance of smaller and local venues in supporting national music ecosystems.
Providing a detailed map of where music happens can contribute to closing the rural-urban divide in access to culture. Demonstrating both the wealth and interconnectedness of existing rural venues can help make them more discoverable for other artists looking for places to perform, and can also help argue for their importance within sustainable local music worlds.
These efficiency gains can improve the sustainability of the music sector. Easy access to information about artists coming to town and venues in a given place can be used to incentivise artists to tour geographically closer cities that can more easily be reached using lower-carbon travel. For example, artists could now easily find places to perform all along the future Fehmarn Belt route between Germany and the Nordic countries. It can also improve career sustainability through improving opportunities for performance and musical exchange, helping artists quickly access new music markets.
[i] Antal, Daniel, Mária Kmety Barteková, and Katarína Remeňová. 2023. ‘Economy of Music in Europe: Novel Data Collection Methods and Indicators’. Open Music Observatory. https://music.dataobservatory.eu/documents/open_music_europe/economy/report/report_music_europe_data_collection.html#data-gaps.
[ii] European Commission: Directorate-General for Education, Youth, Sport and Culture, First Music Contact, KEA, Le bureau export, Music Austria, Music Estonia, Why, Jacquemet, B., Le Gall, A., Saraiva, N., Kern, P., Hergovich, F., Smidt, P., Winkel, D., Sadki, C., Sillamaa, V., Pletosu, T., Dorgan, A., Demeersseman, M.Miclet, F. 2019. ‘Music Moves Europe: A European Music Export Strategy : Final Report.’ Publications Office. https://data.europa.eu/doi/10.2766/68347.
[iii] Yates, Margaret, Anna Campbell, and Mark Casson. 2017. ‘Local Property Values in Fourteenth- and Fifteenth-Century England’. In Large Databases in Economic History: Research Methods and Case Studies, edited by Mark Casson and Nigar Hashimzade, 124–45. Routledge Explorations in Economic History. London New York: Routledge, Taylor & Francis Group.
[iv] Bottero, Wendy, and Nick Crossley. 2011. ‘Worlds, Fields and Networks: Becker, Bourdieu and the Structures of Social Relations’. Cultural Sociology 5 (1): 99–119. https://doi.org/10.1177/1749975510389726.
See also: Bourdieu, Pierre. 2015. Les règles de l’art: genèse et structure du champ littéraire. Éd. revue et corrigée. Points 370. Paris: Éditions du Seuil; Becker, Howard Saul. 2008. Art Worlds. 25 th anniversary ed. updated and Expanded. Berkeley (Calif.): University of California press.
[v] Peterson, Richard A., and Roger M. Kern. 1996. ‘Changing Highbrow Taste: From Snob to Omnivore’. American Sociological Review 61 (5): 900–907. https://doi.org/10.2307/2096460.
[vi] James, Robin. 2017. ‘Is the Post- in Post-Identity the Post- in Post-Genre?’ Popular Music 36 (1): 21–32. https://doi.org/10.1017/S0261143016000647. See also: Muchitsch, Veronika, and Ann Werner. 2024. ‘The Mediation of Genre, Identity, and Difference in Contemporary (Popular) Music Streaming’. Twentieth-Century Music, February, 1–27. https://doi.org/10.1017/S1478572223000270.
[vii] ‘Minor Labels | Price Lab for Digital Humanities’. n.d. Accessed 11 June 2025. https://pricelab.sas.upenn.edu/projects/minor-lables.
[xiii] ‘Nordiska Gränsutbyten Och Mobilitet Inom Kulturområdet’. 2024. Nordic Council of Ministers, 9. https://doi.org/10.6027/nord2024-008.
[ix] On experimental music’s partial embedding within classical music, see Emerson, Gina. 2023. Audience Experience and Contemporary Classical Music: Negotiating the Experimental and the Accessible in a High Art Subculture. 1. Audience Research. Abingdon, Oxon New York: Routledge. https://doi.org/10.4324/9781003142874.
[x] Nordic Council of Ministers, ‘Nordiska Gränsutbyten’.