What we get wrong about streaming

Full script:

---

Data or it didn’t happen.

When my son was a newborn we used an app called Huckleberry to track his daily activities. How often did we change his nappy? How often and how m

uch was he drinking? When was he given medication and how much? And most importantly, when and for how long was he sleeping? In theory, a baby’s sleep pattern tends to become less erratic at three months. What Huckleberry offered was a way to make sense of these early stages through all the data we fed it, helping us to transition from the chaotic sleep of the newborn phase to a more reliable rhythm, making recommendations on everything from the optimal time to put him down for a nap, how many naps a day he should be having at each stage of development, how long we could expect him to sleep for, and how often he should be waking up at night.

In the battle that is early parenthood, Huckleberry was one of our staunch warriors, and of all the gadgets we acquired, probably ranked in the top five in terms of usefulness. For a lot of parents, those first few months are all about shift work. One parent sleeps whilst the other is on duty. And whilst we would obviously be engaged in a constant exchange of status updates about how much he ate and the colour of his shit, logging everything on Huckleberry gave us a one stop bible we could follow, an instant snapshot of our son’s vital statistics, tracking and storing everything in a reliable database was a welcome failsafe for tired, sleep deprived parents.

But despite this utility, there’s clearly an ick factor going on. Not simply that we allowed a company to harvest all this data about someone before they can even talk. But also the fact that it encouraged us to view early parenthood through a certain lens. Track enough data nodes, churn enough numbers, and of course pay a subscription to unlock some more advanced features, and you will be able to not only manage your baby’s habits but actively predict them. The prominent “get sleep” button that appeared every time you open the app exemplifying this ideology, as if sleep was a feature you could unlock if you collected enough points, reducing the erratic psychology of his fast developing brain to analysing a series of inputs to achieve the desired output. Parenthood it seems, one of the most challenging and fundamental human experiences, is not beyond the remit of a well designed app.

Further, my tired, overwhelmed brain nurtured a peripheral sense that if my son’s day to day bodily functions were not properly logged they somehow didn’t happen, they lacked reality. Data or it didn’t happen.

Intro – what does success look like?

I’ve always loved collecting data on the music I listen to. As a kid with a modest CD collection I created a rudimentary chart system, moving an album up the CD rack every time it got played. In the late 2000s I was all but addicted to lastfm, a site that integrated into your listening app of choice – in my case Windows Media Player – and logged the number of plays each artist got, allowing you to track you top artists and genres. And I continue to log my physical collection on sites like Rate Your Music and Metal Archives, the former spitting out fun stats like ownership by decade, record label, and genre. But my diligence in updating these records has reached the point that even when I’m holding an album in my hands, it doesn’t feel properly inducted into the collection until it appears in these online database. I do the same with beers on the Untapped app, I previously did so with my steps, people do it for exercise routines, films, recipes, and even the amount of time they spend using other apps, gamifying through data collection their own quest to be less online.

What streaming platforms offer, a vast library of music, a way to track and trace your listening habits, annual breakdowns of your stats, and the ability to create and curate your own playlists, is difficult to resist. To the point that one could feel discouraged from offline listening through physical media because this simple act, previously so fundamental to music fandom, now lacks reality because it wasn’t properly logged on some digital archive. Engaging in offline listening feels subversive because it exists beyond the tendrils of data collection, yet through this very fact it also takes on a liminal half life, not fully realised because it leaves no trace in the algorithms.

In light of what we already know about a platform like Spotify, the meagre compensation it offers musicians, its aggressive business practices that exploit the works of others, its hand waiving of AI content without clear signposts, and the constant moving of the algorithmic furniture, worrying about some abstract psychological change in the listener’s attitude to their consumption habits may seem trivial. But beneath the shifting economic and technological sands, I think the very bedrock of relationship to music is now on the move. A fundamental rewiring, from a spontaneous, culturally contextualised, messy, chaotic, sophisticated artform, to a simple utility, a diet plan, a source of mental nutrition.

Streaming culture is bringing this into sharper focus and exaggerating a preexisting cleavage in how people engage with music. From a purely consumptive perspective, I think these can be boiled down to two groups of people to demonstrate this point. The first I’ll call “scenic” music fans, meaning anyone engaging with (but not necessarily active in) one or more music scenes. These people could just be punters, but they probably have personal relationships within a scene (local or global) or follow them very closely. The second category is everyone else, who I’ll just call the general listener, which could include passionate and knowledgeable music fans, but they tend to be more stateless in the way they consume music, taking in a more eclectic mix in terms of both genre and era, based on the whims of personal preference. They also tend to be (tend to be) more susceptible to Spotify’s mission to turn listening into an act of pure, personalised consumption as opposed to a grassroots, communal activity.

There is crossover. On the supply side, underground artists have long worked under the assumption that there was an outside chance that they could breakout into the mainstream. And for all the decades long selling out discourse, pop music has long been nourished and refreshed by ideas bubbling up from the underground either organically or through outright appropriation.

On the demand side, listeners can fall into both categories at once. As a lifelong metal fan and orbiter/participant in my local goth scene I would consider myself scenic, with a side hustle in classical and ambient, but if I ever conceive a desire to venture out of this compound and explore a different genre I become a general listener and may well turn to a playlist or two for a primer, as I did last summer with techno.

Edge cases and overlap aside, there is a distinction here that has existed for many decades. And I think it’s becoming sharper and more brittle at the behest of datafication, the insatiable drive from online platforms to collect our data, create subscription models, and anticipate the users’ needs, all of which is sucking up any available oxygen that would otherwise be available for more commercially ambivalent forms of music to exist.

I read a book recently that clarified some of these suspicions for me, Mood Machine by Liz Pelly. Thoroughly researched, wide ranging, and accessible, it documents the history of Spotify, from scrappy disruptor to ultimate arbiter of the music industry. And as we recount the history of streaming and its various permutations I’ll be leaning heavily on her research. As a journalist and champion of indie music, Pelly is very concerned with detailing the many and various ways Spotify is flattening music, restricting the space in which challenging art can exist and even manipulating the artistic process itself. But read through a different lens, her book could be seen as the history of a strikingly successful product in terms of the service it offers to the end user.

And it’s precisely here that I want to focus. When I talk of two distinct types of listener, I don’t do so judgementally. People want different things from music at different times, and with a lot of what I’m about to say I don’t mean to denigrate anyone that’s not particularly interested in music at the granular level. We all have different interests. But it is true that for some people it would barely register if the music they were fed was made by AI, or a paid up Spotify implant making music to fill playlists with the required vibe. And although Spotify didn’t create this attitude, it has vastly expanded the space for it to be exploited, and in the process, as Pelly would argue, it is restricting the space in which scenic music can operate. It has essentially abducted music from its place within societies and cultures and placed it in a contextless, data determined nether zone to serve as a utility that can quite literally be streamed into our homes at a moment’s notice, whose chief purpose is to enhance a range of mundane daily activities.