Last week, HUMlab at Umeå University organized a conference with the somewhat elusive title Genres of Scholarly Knowledge Production. This was a three-day event that experimented with – and critiqued – the concept of the academic conference itself, and mostly drew scholars from various fields within the digital humanities. I, Maria Eriksson, was there as a result of my affiliation with HUMlab, and presented some initial results from a research experiment I’ve been conducting during the past month, alongside the works of Rasmus Fleischer and Anna Johansson.
In particular, this experiment has been geared towards trying to find a way to open up the “black box” of streaming music, and to try to follow – or perhaps even stalk – streamed music files.
Now, one way of trying to follow streaming music files as they move about within the Eco-systems of digital streams, would be to try to talk to the people who use or build these services.
Instead of doing this, however, we’ve decided to adopt a kind of DIY approach. That is, we produce and distribute music as a way to critically engage with streaming platforms. These experiments are grounded in ethnographic methods and participant observation. By enacting and setting streaming music and artistry in motion, we want to develop a deeper understanding of the software workings, organizational structures and forms of meaning making that take place behind streaming platforms. In the process of doing so, we study both ourselves (our choices and experiences), and the institutions, actors, gate-keepers and programs that we face as our music gets released.
At the conference, my contribution to the discussions was therefore to tell the story of how I became an “artist” as part of these explorations. My talk was meant to give listeners an insight into what it takes for something to become streamed music, and to make a case argument for the kinds of insights that can grow out of these kinds of interventionist methods.
In order to recoup what was said, my modest “music career” began at my kitchen table where I recorded sounds from my breakfast, took a picture of my kitchen carpet to use as an album cover, and in the end produced three tracks which I jointly labeled “Frukost” (or “breakfast” in English).
I then proceeded to try to get my music distributed by a music aggregator. To keep it short, this is a new kind of intermediary actor that has popped up since music distribution has gone online. What an aggregator does, is to take music and push it out onto various digital platforms such as Spotify and iTunes, either for “free” (I’ll shortly come back to what the concept of “free” means in this context), or against a fee. The advent of aggregators means that artists are no longer forced to have a record contract before their music get the chance to be distributed across vast distances. What remains for an artist that wants to have their music released, however, is to convince an aggregator to take their music under its wings.
I decided to first try my luck with aggregators that market themselves as “free”, which means that no initial cost is demanded in order for the music to be distributed. Signing a contract with one of these aggregators does, however, mean that the aggregator holds rights to a certain percentage of any future earnings.
To my great disappointment, it turned out that these initial attempts were not a great success. My Frukost-tracks got rejected twice. One aggregator told me that my sounds were not “music content”, and another aggregator that tried to put things a bit more nicely, gently replied that my music was not the kind of content they were looking to sign up at the moment. If these rejections were the result of a machine or a human actually listening to my sounds, remains unknown. What is clear, however, is that aggregators play a crucial role in defining what counts as “music” today.
Rather than giving up, however, I decided to pay for myself, and it turned out this made things a lot easier. I aimed to sign up for a one-year contract with a subscription based international aggregator, and then started going through a long process of answering questions about myself and my music in order for my music to be released.
For example, I was asked to put genre labels on my sounds – something which forms the basis for the kinds of metadata materials that recommendation algorithms base their suggestions on. I was also asked to decide whether I was aiming for “world domination” (which would mean having my music distributed onto 17 different platforms online), or if I was more of a “top dog” (which would mean signing up for a package that would make my music appear on five different platforms). I decided that a top dog was probably enough in my case.
I also had to decide on which royalty levels I wanted to maintain for my sounds, were a higher royalty level meant paying more money up front. My tracks also had to be re-formatted in order to fit the standards of this particular aggregator. In this case, this meant that I deleted one of the three tracks I had first produced, in order to squeeze my music into the EP-release package the aggregator had on offer. In the end, I paid $20 to have my music released on 5 different platforms (Spotify, iTunes, Deezer, Amazon MP3 and Google Play).
After 5 days, my sounds first appeared on Deezer. After six days, they were available on Spotify and iTunes, and after about 30 days I’m still waiting for my tracks to appear on Amazon MP3 and Google Play.
After my sounds had taken their first steps out into the world, I was given access to tools that helped me monitor the success of my “music”, and by extension also the success of myself as an artist. These tools for example contained daily statistics regarding my Spotify fan base, broken down into categories of age, gender and geographical location. I was also faced with graphs that displayed my sales incomes and royalty revenues, and the aggregator continuously offered me promotion opportunities in order to boost my fame.
Having gone through this process of signing up and engaging with many different aggregators, I started realizing something important: my music, and my artist self, had started migrating onto various platforms. Rather than being confined to the five platforms I had originally signed up for, I was now also represented on sites such as Twitter, Facebook, Paypal and imgur. Partly, this was the case because I had been forced to use some of these services in order for my music to be released. In other cases, I had just been strongly advised to create such accounts, and decided to follow suit. Rather than thinking of streaming music as objects which are confined to particular streaming platforms, this revealed how streaming music is entangled in complex networks of actors, services and practices of self representation.
In the end, I argued that this experiment has pointed to three important results:
– First, it has offered access to the backsides of streaming platforms, and thereby areas to which it otherwise would have been difficult to gain entry.
– Second, it has offered insights regarding the kinds of expectations that are put on contemporary artists looking to merge into the stream.
– And last, the experiment has also provided great navigational guidance regarding which actors and platforms that are important to have a look at in the future.
As a starting point, this experiment has also put in place a structure from which the pathways and movements of streaming music can be explored. It hasn’t allowed me to open up the black box of streaming music completely, but it has at least given us a chance to peek into what might lie inside.
For those who are interested, here is a sample from my songs on Spotify, and Deezer. All talks from the HUMlab conference were also recorded and published here.