Online networks and the production of value in electronic music

Cultural value is one of the areas in which (as the saying goes) perceptions are also realities. Thus, sociologists have argued that the production of cultural value is actually the production of a form of belief. Although popular accounts of how culture gets made tend to focus on brilliant individual creators, research has highlighted over and over again that their work typically emerges from a creative milieu, in which value (or belief in value) comes into existence. This arguably explains why cultural producers - musicians, artists, etc - are among the most committed audiences for cultural products, and why so many of them create work primarily for appreciation by other producers. This highlights the relationship between amateur, professional, and semi-professional cultural production, which is commonly assumed to have been transformed by Web 2.0 technologies that potentially making every internet user into a creator and publisher of digital content.

This project will make an incisive contribution to our understanding of how culture and its value is produced in the digital age by focusing on a specific art form: music. We will gather evidence from the SoundCloud website, which many musicians use for commenting on one another's work. And we will focus on a specific genre that has a special relationship with that website, i.e. electronic music. We will identify as 'electronic music' all audio files that are identified by their creators with synonyms of or recognised genres of electronic music (to take just a few examples: dubstep, intelligent dance music, electronica, algorave, chiptune, electronic pop, live coding). Although many registered users of SoundCloud do not produce audio content, the website's original core users consisted of musicians - especially electronic musicians - who use it for commenting on one another's work. This project will study relationships between millions of users of that website, building a series of social network graphs to represent their evaluations of audio files uploaded by other users. Social network analysis provides mathematical ways of analysing peer esteem and reflecting the intuition that the opinion of a highly-regarded creator may have special impact: for example, in a social network diagram representing positive evaluations, more highly esteemed producers will occupy more central positions, with the most central positions being occupied by those who are esteemed by other highly esteemed producers. In order to enhance this analysis, we will not only study evaluations implied by 'likes' and 'follows' on the website, but also use computational linguistic analysis to study the kinds of language used in comments and identify whether a given comment is likely to be positive or negative. Moreover, we will observe and interview musicians in the real world in order to understand how they locate value in their relationships with one another, both online and off.

Findings will be disseminated through a range of venues, including a public access event featuring lectures from academics who have carried out related studies and invited performances by musicians who have been studied. We will also write a public report explaining our findings to a general audience (including emerging musicians who may wish to understand the role that websites such as SoundCloud can play in building a career), and we will release the source code to all the applications we create, with full instructions that will enable other researchers (both inside and outside the university system) to adapt those applications to study data from other sources. Our website will include a blog providing progress updates, and knowledge will be further disseminated through professionally-produced audiovisual podcasts and press releases to specialist media.

The project will have impact on the cultural sector in three broad ways.

First, it will provide direct guidance on how social media can help emerging musicians in establishing their careers. This will be communicated through podcasts and through the first part of the public report; coverage in specialist media will also be sought (including not only conventional media such as DJ Magazine but also internet radio stations such as Brick Lane Radio). The planned public engagement event will attract journalists, musicians, and music industry professionals with its mix of performances and lectures on topics of high interest as well as presentation of project findings; a leaflet providing key details and a link to the project website will also be distributed at that point. Note that a representative of the Musician's Union with responsibility for education will be a participant in the workshop, helping the project team to bring out the findings with most direct application to emerging musicians. The website will be designed to appeal to an audience outside academia, and the podcasts will be produced by a camera operator and editor with experience not only in educational video but in music video as well.

Second, it will provide a more nuanced and less instrumental (but still empirically grounded) vocabulary with which to articulate the contribution made by particular cultural producers or institutions. Note that a leading cultural policy researcher (Belfiore), a senior consultant from a leading arts consultancy (Gibbon), and a representative of the Musicians' Union (Radcliffe) will be participants in the workshop. Further cultural professionals will be invited to the public engagement event, ensuring that the project will have the widest possible impact on discussion of cultural value even before the release of the report. By showing that reputation within a cultural field can be treated as a proxy for cultural value and studied through social network analysis, the project will provide a theoretically and empirically rigorous riposte to the 'defensive instrumentalism' (Belfiore 2012) of recent policy discourse on the arts, showing that it is possible to discuss cultural value in objective terms that do not compromise the autonomy of the cultural sphere by reducing cultural value to other forms of value. It will also demonstrate the importance of creative interactions between cultural producers, which will have direct application in advocacy for cultural institutions that foster such interactions (for example, studio organisations and mentoring schemes).

Third, the project will develop a rigorous, software-supported methodology for studying the production and circulation of value within a cultural field. This methodology is conceived not only as a means to answering the project's research questions (see Objectives) but as an output in its own right, usable not only in other academic research but in support of evidence-based advocacy and decision-making in the arts. The software developed will be designed to be re-purposed as easily as possible: cleanly coded, fully-documented, extensively commented, and written in one of the most flexible and expressive programming languages available, i.e. Python; it will also be released under a GPLv3 licence, permitting not only reuse but also rewriting and redistribution. The second part of the public report will explain how the methodology could be adapted for purely-online and purely-offline research (for example, network analysis of questionnaire data). This could be employed in consultancy work, for example to show the contribution made by a particular venue for the production of cultural value through the facilitation of interactions between producers.