Algorithms that Matter

Algorithms are everywhere today, and at the same time they are hidden. We perceive them when we are presented with related items of interest in an online store or on a social media site, whenever we query a simple term on the web. Nonetheless, we would find it hard to exactly describe how any of these algorithms work. We take them for granted, accept that we are subject to their analysis and decision-making. How can art contribute to understand the increasing influence of algorithms and translate them into aesthetic positions? Algorithms have been used in music even before the emergence of computer music in the 1950s, but today we witness an entire new wave of interest, reflected in festivals, genres, publications and research projects. Interactive and real-time control of compositions has been possible already for two decades, so the reason must be sought elsewhere. It is the very notion of algorithms that is shifting. They are no longer an abstract formalisation, the image of thought, immaterial, static and timeless. Instead of being givens, they emerge from artistic praxis and experimentation, they become entangled in material processes that produce space and time. The project Algorithms That Matter (ALMAT) is grounded in this new idea that algorithms are agents that co-determine the boundary between an artistic machine or apparatus and the object produced through this machine. The central question is: How do algorithmic processes emerge and structure the praxis of experimental computer music? The hypothesis is that, instead of being separated from the composer generators and transformers of infinite shapesthey exhibit a specific force that retroacts and changes the very praxis of composition and performance. A series of methodical artistic experiments is carried out by the project team together with guest composers. Over defined periods of time, they develop series of interrelated sound pieces. The work process is observed and transcribed into complementary forms of presentation and discourse on which future research projects can build. This includes concerts and exhibitions, an online public continuous exposition, and gatherings and symposia that connect researchers across various institutions in Europe. The project not only aims at extending the praxis of experimental computer music using algorithmic processes, but also at contributing to the scope and methodology of artistic research, introducing a field of research so far disregarded. ALMAT is hosted by the Institute of Electronic Music and Acoustics (IEM) at the University of Music and Performing Arts Graz, where it is integrated with ongoing research and teaching and performing activities. It seeks not only to be visible in the related research community, but also to reach out to young researchers and a general audience across different fields, raising the awareness of artistic research praxis.

Algorithms are everywhere today, but they are often hidden. We perceive their effects when we are presented with related items of interest in an online store or on a social media site, whenever we query a simple term on the web. We would find it often hard to exactly describe how any of these algorithms work. Irrespective of who exerts power through algorithms, we can say they have become part of our nature. Rather than looking at specific algorithms and their specification in isolation, to understand the way they shape us, we must look beyond the single intention of a designer, purpose and use. From algorithms we move to the algorithmic, a fabric that determines how algorithms work, interconnecting them, even into the world beyond the boundaries of computers, inducing intrinsic rhythms and strategies that yield consequences deeply rooted in the medium of computation itself. The project Algorithms that Matter focused on one type of consequence: the aesthetic impact when working with sound and music, something not only visible and audible in the artefacts, but changing the way we understand the artistic process and our role as artists. Here, we move beyond the traditional narrow view of algorithms as abstract thought, governed by formalised logic. We understand the algorithmic as a continuous process, emerging from praxis and experimentation, as algorithms become entangled in material processes and contexts that produce space and time. To look at something so difficult to grasp as process, we combined practical work with observation and discourse, targetting the complex exchange between artists and the algorithmic ensembles that they build and code. The questions we ask include: how do we select specific ways of working with algorithms, how do they open up specific affordances to steer the aesthetic development, how do we intervene in this steering, and how is the interface with the physical and sensual world created; most importantly, how can we share experiences that seem so inseparable from personal practices with other artists-researchers? We carried out a series of iterations, the project team coming together with guest artists. We developed interrelated sound pieces, particularly focusing on the processes in which decisions and adjustments were made, and communicating these both on the non -verbal level, through the practical sound work, as well as discussions in the lab, creating a vast trace of exchanges on an online platform, which constituted a living document or virtual workspace called the continuous exposition. The experiments were validated through exposure in concerts, exhibitions, festivals, conferences, and we invited peer groups in the form of workshops and symposia across various institutions in Europe. The continuous exposition was finally transformed into a public database.