Cultural Analysis, Volume 16, 2017
“Keep Smiling!”: Time, Functionality and Intimacy in Spotify’s Featured Playlists
Abstract: As one of the world’s largest online music providers, the streaming service Spotify has a profound capacity to shape everyday realities through digital technology. This article explores how both openness and control are embedded in Spotify’s ways of delivering recommended playlists to users. After analyzing over 500 pre-designed playlists, we argue that Spotify’s music recommendations evoke individual freedoms and flexibilities, at the same time as they prescribe normative temporalities, neoliberal subjectivities, functional approaches to music, and monetizations of intimacy. Such tensions between freedom and control speak of the dual inheritance of the digital and its potential to both liberate and constrain human action.
In June 2017, Spotify claimed to have about 140 million active users, and as people increasingly turn to streaming platforms, there is a need for a deepened understanding of the realities these services promote and materialize. Rather than neutrally channeling sounds, platforms such as Spotify take an active role in framing music, which includes the promotion of certain values and subjectivities. Like digital technologies in general, streaming music services simultaneously draw on a vision of free and unlimited access, and on regulatory practices that select and privilege certain content, collect user metrics, and deploy algorithmic ways of organizing information (Cheney-Lippold 2011). Such shifts between openness and control have surrounded Internet technologies since the very beginning. Associated with democracy, participation and emancipatory values on the one hand (Shirky 2008; cf. Turner 2006), and authoritarian control and surveillance on the other (Morozov 2011; Fuchs et al 2012), digital platforms occupy a contested position in today’s media landscape.
In this article, we set out to investigate how these dual logics of freedom and commercial-institutional power are played out through Spotify’s music recommendations. More specifically, we focus on one moment in the “social life” (Appadurai 1986)1 of streamed music files by studying how Spotify’s so-called Featured Playlists are presented to users in three different countries during one week’s time. At the time of writing this article, Featured Playlists were selections of 12 readymade, curated lists of songs delivered to users upon login, together with a short greeting—such as “Kick start this Tuesday!” or “Enjoy time with friends and family”. These are designed to cater to the expected everyday life of Spotify’s users and depend on three broad variables: users’ registered language, country and date/time.2
Occupying a central place in Spotify’s new strategy for delivering music, Featured Playlists are fascinating entities that map and provide musical context (Seaver 2015, 2012)—both in the sense of creating an affective aura around sounds, and in the sense of approximating listener behaviors and preferences. Acknowledging that such contextualizations are always socially constructed and linked to interpretation (Dilley 2002), we approach Spotify’s framing of playlists as an activity whose politics needs to be explored. Following Lev Manovich’s (2001) call to study how software interfaces organize data in particular ways and hence “privilege particular models of the world and the human subject”, we therefore ask: In what ways are musical contexts constructed through Featured Playlists and how are they entangled with the expected everyday life of users? What ideals, assumptions and subjectivities are (re)produced in Spotify’s organization and presentation of music? And how are such elements tied to the dual logics of freedom and regulation?
In answering these questions, we focus on three aspects which we view as central to the way music is contextualized in Featured Playlists: temporality, functionality and intimacy. Whereas the Spotify platform builds on a logic of user participation, flexibility and freedom of choice—meaning that people are always given the option to rearrange, select and ignore any playlist—our argument is that the specific contextualizations of Spotify’s readymade playlists are suggestive of neoliberal or radical individualist ideologies. In particular, we claim that Spotify’s promotion of prescriptive temporalities, its presentation of music as functional for productivity and well-being, and its structures for producing and monetizing intimate expressions exemplify how systems that provide freedom and flexibility for the individual user, might also be bound up with “productive constraint” (Stanfill 2015) and market-driven attempts to monitor and regulate audiences.
Spotify and the provision of “free music”
Structured playlists and content recommendations
To a large extent, Spotify’s new role as a provider of tailored musical experiences has come to center on the delivery of pre-designed playlists—that is, themed collections of songs that users can enjoy and save in their private music collections on the Spotify platform. Like lists in general, Spotify’s playlists are devices for taste-making, but also entities that carry a “dynamic capacity… to be both open and closed, to suggest both action and the ordering of action” (Phillips 2012, 97). As Phillips has put it, (play)lists are simultaneously fixations, aesthetic objects and theatrical pieces; they are “at once endless… and restrictive” (ibid., 104). While playlists can be transformed, extended, and edited according to the user’s wishes, they come in a pre-packaged format that transports affectual ideals, notions of “the good life,” and conceptions of time (and time well spent).3
In this article, we are particularly interested in exploring how Spotify’s Featured Playlists (which are non-personalized) take shape in relation to discourses of personalization on the platform.4 This focus on the non-personalized implies going in another direction than much recent research on online recommendation systems, which has mostly focused on algorithmic dimensions of content curation, and how algorithms foster certain personalized cultures (Hallinan & Striphas 2014; Galloway 2006), networks (Ananny 2016), humanities (Berry 2011), identities (Cheney-Lippold 2011), ideologies (Mager 2012), publics (Crawford 2015), performativities (Introna 2016), accountabilities (Neyland 2016), and forms of governance (Ziewitz 2015). While Spotify’s Featured Playlists partly come about by way of algorithmic data management,5 we emphasize instead the humanly curated and descriptive texts and images that frame them. These musical wrappings are central to the process of turning digital, abstract and coded music into attractive goods and something that resembles physical commodities (Morris 2011). They are also fundamental for the ways in which music becomes entangled with ideas about everyday life, since playlist descriptions actively invoke different subject positions and notions about how, when, and by whom certain sounds could (or perhaps should) be enjoyed. Rather than discussing the implications of algorithmic music recommendations, then, we focus on how such collections are presented and framed—which we believe can be traced through playlist descriptions and the greetings that surround them.
Collecting playlists: Methods and materials
For the purpose of technographically studying Featured Playlists, we created three Spotify user accounts that would allow us to make observations and collect information on the kinds of content shown to each user.7 Besides the short greetings that meet users upon login to Spotify, we were interested in playlist titles, cover images, and playlist descriptions, and in monitoring their change over time. The user accounts had identical settings: they were registered on the same day, and were listed as 25-year-old females. Because we wanted to explore and compare content delivered in different countries they were respectively assigned a Swedish, an American and an Argentinean identity. The selection of nationalities was limited by the actual countries in which Spotify is available, and motivated by an interest in exploring data from different continents. Furthermore, Spanish, Swedish and English languages are known to us, thereby facilitating the analytic process.
With the technical assistance of Roger Mähler and Fredrik Palm at Humlab, Umeå University, we used an automatic script to log into each account once every hour during one week’s time, starting at 8 am on September 1 and ending at 7 am on September 9, 2015. After logging in, we documented the presented greetings and recommended playlists, and signed out again. This process was repeated every hour, starting at 8 am on September 1 and ending at 7 am on September 9, 2015. In total, empirical data was collected on 168 different occasions during the course of this week. The data consisted of 142 unique greetings (the Argentinean user received 42 different greetings, the American user 52 and the Swedish user 48). We were further recommended 542 different playlists (the Argentinean user received 117 unique playlist recommendations, the American 213, and the Swedish user 212). Noteworthy is the fact that on average, 72% of the greetings were repeated more than once. In a similar way, 91% of the playlists were delivered on multiple occasions during the week.
The collected material—images and texts translated to English—was coded using qualitative data analysis software ATLAS.ti. This software facilitates organization and categorization of unstructured empirical data and offers functions for search and retrieval as well as for visualization of co-occurring categories. Building on a thematic approach, we analyzed the material in terms of both form and manifest and semantic content. More specifically, greetings, playlist titles, playlist descriptions and playlist covers were categorized according to content, weekday, time of day, musical genre, and mode of address. In this way, three main patterns of music contextualization began to emerge, as we further outline below.
The temporality of playlist recommendations
During weekends, the circadian rhythm slightly shifted: according to the content of messages and playlists, evenings lasted longer on Fridays and Saturdays. Hence, Saturday and Sunday mornings also began later and were typically described as lazier than weekday mornings. During this time of the week, work was not mentioned at all, and emphasis was instead put on sociability, recreation, and “chill-out.” Furthermore, certain moods and activities appeared to be particularly distinctive for weekend mornings. For instance, Spotify delivered a number of “Hangover Friendly” playlists on Saturday and Sunday.
After 4 pm, weekdays were typically associated with commuting, returning home and winding down. Playlists such as “Evening Commute”, “Long Way Home”, and “Relax & Unwind” speak to this theme, as well as exclamations like “How nice it is to get home and enjoy Home Sweet Home!”. Weekends were somewhat different in this respect, as they were clearly favored above weekdays and often characterized in more celebratory ways. On Friday afternoon, for example, Spotify in the three countries cheerfully called out: “Happy Friday with V of Viva la Vida!”, “Let the weekend begin!” and “Feel the Friday fever!”
After 11 pm, the Spotify client appeared to think it was time to sleep, at least on weekdays. Nighttime playlists were mainly devoted to help users unwind, as demonstrated in titles insisting on sleep: “Sleepify”, “Sleep Tight” and “Jazz For Sleep”. However, weekend nights seemed to stretch much longer, as users were encouraged to “shake their booties” and “dance until sunrise”. Another significant activity, only mentioned at night, had to do with sex. One playlist, for example, prompted us to “Get cozy and make time for some kissing and cuddling!” In the same vein, other playlists urged us to “Close the door… and turn off the light” or boldly asked “Getting laid?”.
Featured Playlists as Regulated Temporalities
The specific ways in which realtimeness is produced through Featured Playlists includes not only the everyday rhythms demonstrated above, but also references to seasonal changes and current events, such as the Pride parade and the death of a popular Argentinean musician. Such timely playlist deliveries can be seen as an attempt by Spotify to embed itself in the everyday lives of its users and uphold an image of being constantly up to date. However, as previously mentioned, it turned out that 72% of the greetings and 91% of the playlists were actually repeated more than once during the course of our one-week data collection. Moreover, the Argentinean user received several greetings that declared “August has never sounded better”, although the month was in fact September. These time lags, we argue, are excellent examples of how realtimeness is a precarious construction which may be undermined and contested by instances of delay.
Furthermore, we suggest that the ways in which realtimeness is produced through Featured Playlists reproduce ‘chrononormative’ assumptions related to everyday life. Freeman (2010, 3) uses the notion of chrononormativity to describe prescriptive temporalities, and more specifically “the use of time to organize individual human bodies toward maximum productivity”. She defines “temporal mechanisms” as those social and political processes that reproduce time-related norms of work, health, citizenship and family life. Such temporal regulations—expressed in the form of life course expectations—must be adhered to so that one’s embodied existence becomes socially meaningful from the point of view of capitalist and heteronormative ideologies. In our case, Spotify’s chrononormative effects were most evident in how the different dayparts did not only privilege certain updated musical content, but also included the explicit designation of activities and moods to different time slots. Together with the personal and imperative user address, this served to embed strongly prescriptive temporalities into the platform.
Music as a means: The functionality of playlists
Significant for playlist descriptions invoking this view of music was their reference to what users can achieve by listening to a particular list. The type of achievements—or the everyday areas in which music was suggested to be of help—largely corresponded to the main “genre categories” found on Spotify: workout, party, focus, chill, sleep, travel, dinner and romance. For instance, we were offered energy boosts so as to perform better when exercising, prompts for heightening our productivity at work, or repeated suggestions to use music for increased focus and concentration; Spotify recommended “rock to help you relax and concentrate”, and “Epic All-Nighters” that would help us “power through” our night time studies. Music was generally described as performative of motivation and energy—regardless of whether it had to do with dancing, getting up in the morning, or getting in the mood for partying. In part, this disciplining towards heightened performance and productivity was dependent on time regulation; hence, this too can be seen as contributing to the politics of chrononormativity, where playlists served as promises of time-bound, idealized lifestyles.
This is not the first time that music has been treated as a functional device for the purpose of increasing productivity. Much research has for example focused on the use of music in the workplace in order to increase efficiency (see Prichard, Korczynski & Elmes 2007). As Jones (2005, 724) has shown, there is a long tradition of using music as a means to “impact upon worker output” rather than appreciating it for its “aesthetic or artistic ‘value’”. Others have pointed to how workplace music upholds order in post-Fordist, capitalist societies (Jones & Schumacher 1992) or, conversely, how music consumption is a dialectical cultural practice through which workers can partly express their resistance to a routinized and alienating structure (Korczynski 2011).
We concur with the latter perspective in acknowledging that Spotify users can make sense of music in their own ways and may also freely choose between differently themed playlists. At the same time, however, we want to stress that we were repeatedly invited to view music consumption as an accompaniment to other, more significant tasks, rather than as an activity in its own right. The goal, here, was not only increased productivity but also a general improvement of one’s mental state and attitude to life, something seen in playlist descriptions like, “Get happy with this pick-me-up playlist full of feel good songs!”, “Stay focused and smart with these house tracks”, and “Nothing hurts as heartbreak. These songs will help you have a good cry.”
By providing such sketches or maps of emotional states, Spotify creates a certain type of mood environment for its users. For Anderson (2015, 838), musical mood environments “point toward a fantasy of intentional and while-you-wait mood treatments and manipulations, as if moods could be put on and shed as easily as winter hats and mittens”. In line with Anderson’s view, Spotify’s Featured Playlists can be seen to educate users on how to classify themselves according to their temper. At the same time, the playlists also encourage treating cognitive and emotional states as garments which can be pulled out of a closet, tried on, and easily be put back again. The delivery of playlists thereby serves to guide users toward particular treatments of their mental lives.
However, based on our specific study, we argue that Featured Playlists do not primarily encourage the exploration of or cycling between a wide range of moods in the way Anderson (2015) has proposed. Instead, Spotify’s promotion of music as functional primarily privileged a subject determined to strive toward well-being. This was evident in the many calls to use music for personal improvement, and in greetings and playlist descriptions such as “Conquer your morning”, “Like a boss” and “You're on top of the world. Don't forget it.” The positioning of the user as a boss, a potential conqueror, or someone on top of the world, we argue, is clearly connected to the notion of music as contributing to enhanced performance. These and other examples illustrate how Featured Playlists promote a type of subjectivity informed by positive thinking as a contemporary ideology (Ehrenreich 2009), and the related investments in unattainable fantasies of “the good life” that subjects are propelled into by neoliberal society (Berlant 2011). There were indeed a few examples in the material showing how playlists can offer accompaniment to the dark sides of life, such as when the Featured Playlists sympathetically claimed to want to “reduce insomnia and anxiety”, or provide music to cure “Morning Melancholia”. Even here, however, focus tended to be on overcoming hardships by listening to the right playlist. In this respect, Featured Playlists not only push users towards increased productivity, but also privilege a mode of entrepreneurial subjectivity in which users are encouraged to direct their desire for change inwards and “relate to themselves as if they were a business, are active, embrace risks, capably manage difficulties and hide injuries” (Scharff 2016, 108).
Intimacy and its monetization
As seen in the previous sections, the presentation of Featured Playlists often addresses users in a highly personal tone. Spotify frequently promotes collections of sounds by using imperative moods, by posing questions, or by invoking an explicit “you”. Thus, we were met with seemingly personal greetings like “What’s for dinner tonight?”, or playlist descriptions humbly asking “Trouble sleeping?”. While these questions are open for multiple answers and interpretations, they are clearly also suggestive of specific scenarios. In particular, by hailing a “you” and bringing up topics that in many other contexts would be seen as belonging to the private sphere, the client suggests a close and cordial relationship to the user(s) which in turn may produce a sense of intimacy on the platform.
Despite this personal address (and unlike much other content on the Spotify platform), Featured Playlists are not personalized but mass broadcasted elements which depend on country, time, and date. On one level, modes of address that approximate personal relationships and interactions have been part of mass-media communication for a long time (Durham Peters 2010; Horton & Wohl 2006 ). Such discursive strategies may for example be found in TV shows or radio broadcasts, and as Horton & Wohl (2006 ) noted already in the 1950s, intimate speech in these contexts has traditionally worked to simulate a sense of participation and belonging, despite the fact that communication is often one-sided. Horton and Wohl argue that intimate speech serves to establish guidelines for audience behavior, since intimate language creates a sense of sympathy, sociability and friendship, and thereby constructs audiences who behave in a similar way (even though this might certainly backfire and create detachment and rebelliousness). Spotify’s personal tone of language may thus be understood as adopting classic rhetorical strategies of mass media in order to pave way for friendly attitudes amongst its audiences.
The intimate speech that surrounds playlists also borrows from other earlier media formats that have circulated within private spheres. In particular, playlists have much in common with the intimate, eclectic tropes of mixtapes, that is, analogue cassette tapes filled with self-recorded music, which were commonly circulated amongst friends in the pre-digital era. As Drew (2005) has pointed out, the positive and emotive associations of mixtapes have frequently been picked up by creators of commercial music mixes, who capitalize off of their personal and homemade feel—and the same could be said of playlists. By being structured and named in ways that are akin to personally crafted music collections, playlists borrow their aura—a process which can be seen as a kind of appropriation of non-commercial social relations and practices of sharing amongst music fans.
Most importantly, however, Spotify’s attempts to produce and mobilize different modes of intimacy need to be understood as part of larger market logics, and in particular the company’s ways of gaining ad-based revenue. Rather than solely guiding and catering to the needs of listeners, playlists occupy a central role in Spotify’s strategy to attract advertisers and thereby keep the free—and advertisement-based—version of the service afloat. At the time of data collection, Spotify offered advertisers the possibility of reaching users according to eight different activities and moods that are found in playlists: workout, party, chill, focus, dinner, kids & family, travel and romance.9 In this way, Featured Playlists serve as preparations for behavioral marketing—an advertising strategy aimed at segmenting audiences based on their conduct, emotional states and personality traits (Connolly 2015, Peterson 2015). This targeting ensures that advertisements can be sent out to particular clients at particular points in time, based on what music they are listening to. Spotify thereby makes it possible to broadcast health and fitness ads to someone who is listening to a workout playlist, or food oriented ads to someone who is listening to a dinner playlist, for example. Through such processes, Featured Playlists become embedded in larger strategies for creating ever-more granular market divisions of audiences.
One way of understanding these strategies of turning playlists into objects for consumer segmentation, is to view them as expressions of how “public display and mediation of personal emotion and affect” becomes linked to monetary value (Hearn 2010, 429). As the Spotify machinery gears more towards the provision of playlists that evoke intimate moods, the service becomes (financially) dependent on users’ willingness to disclose their feelings by selecting a playlist that suits them. Here, we might again think about the functional, mood-oriented playlists that urged us to “get happy”, “stay focused” and “have a good cry”. On the one hand, the act of providing playlists that speak to a user’s private life is in itself constitutive of intimacy; on the other, intimacy is also produced and monetized as users are expected to share their activities and emotional states by selecting preferred playlists.
However, as demonstrated in this article, flexibility and freedom for individual Spotify users is—in a wider perspective—also bound up with control and disciplining of audiences. We have argued that such regulatory effects are primarily produced in three ways: through specific organizations of time, through the promotion of music as a means, and through connecting intimate patterns of listening with revenue through ad sales. By drawing on prescriptive temporalities in the packaging of playlists, Spotify suggests chrononormative organizations of everyday life that privilege some lifestyles over others; for example, attending to social relations in the evenings, engaging in intensive and white-collar labor during the day, and upkeeping a romantic life at night time. These normative ideas of everyday life are further advocated in the promotion of music not only as an aesthetic object but a functional tool for entrepreneurial subjects who strive to lead better, happier, and more productive lives. Further, the framings of playlists serve to produce a sense of intimacy that, in turn, generates revenue for the service. The connection between playlists and advertising strategies, reveals how capitalist logics merge with (both real and imagined) mundane life and personal spheres.
In this sense, we argue that the musical contexts discovered in this study–temporality, functionality, and intimacy–point to how the Spotify client is driven by market logics and tends to promote neoliberal or radical individualist ideologies. Featured Playlists, it seems, may work to discipline audiences and (re)produce particular subjectivities as well as normative ideas about “the good life”. While Spotify’s playlist deliveries open up free spaces for music exploration and subversion, they can thereby also be read as expressions of institutional-commercial power that users need to relate to—in one way or another.
This work was financially supported by the Swedish Research Council/Vetenskapsrådet (Framework Grant scheme, D0113901).
1 Following Appadurai’s call to analyze “the social life of things”, means we recognize that our study can only grasp a momentary slice of the cultural biographies of streamed music. [ Return to the article ]
2 See https://developer.spotify.com/web-api/get-list-featured-playlists/ (retrieved 8 April, 2016 & 3 January, 2018). [ Return to the article ]
3 Importantly, such packagings build on a much longer history of structured music deliveries that compile and frame music according to pre-established notions of taste, listening and pleasure. Some of the most relevant precedents to playlist deliveries might be found in classic jukebox technologies (Attali 2009 ), radio broadcasts (Tacchi 1998), cassette tapes (Skågeby 2011; Drew 2005), and compilation albums (Wikström & Burnett 2009). [ Return to the article ]
4 By personalization, we refer to the ways in which users’ behaviors on a platform come to underlie the content that is presented to them. [ Return to the article ]
5 In an interview from 2013, a chief product officer at Spotify described how their playlists “draws on three ‘pillars’ using data from a subscriber’s friends, personalised recommendations and real music experts” (Warman 2013). Spotify’s playlists can thereby be described as a result of the joint labor of human music editors (who describe, label, and put music in context), users (whose everyday activities and social networks are monitored and analyzed), and algorithms (whose labor consists in sifting through large amounts of data, and sketching out connections between sounds). [ Return to the article ]
6 Hence, our study says nothing about how playlists are actually experienced or put to use by Spotify listeners (for such accounts, see e.g. Nylund Hagen 2015; Werner & Johansson 2016). [ Return to the article ]
7 With the public interest in mind, we appreciate Spotify’s forbearance with any trespassings of the user agreement that this data collection involved. [ Return to the article ]
8 While the performative power of music was a major theme in the material as a whole, the extent to which it was referred to differed slightly between the three countries: it was most common in the Swedish material, and a little less so in the American and Argentinean data. [ Return to the article ]
Ananny, M. 2016. “Toward an Ethics of Algorithms: Convening, Observation, Probability, and Timeliness”. Science, Technology & Human Values 41(1): 93–117.
Anderson, P. A. 2015. “Neo-Muzak and the Business of Mood”. Critical Inquiry 41(4): 811–840.
Appadurai, A. 1986. “Introduction: Commodities and the Politics of Value”. In The Social Life of Things: Commodities in Cultural Perspective, edited by A. Appadurai, 3–63. Cambridge: Cambridge University Press.
Attali, J. 2009 . Noise: The Political Economy of Music. Minneapolis & London: University of Minnesota Press.
Barbrook, R., and A. Cameron. 2007. “The Californian Ideology”. Accessed at: http://www.imaginaryfutures.net/2007/04/17/the-californian-ideology-2/ (June 27, 2017).
Barr, K. 2013. “Theorizing Music Streaming: Preliminary Investigations”. Scottish Music Review 3: 1–20.
Berlant L.G. 2011. Cruel optimism. Durham: Duke University Press.
Berry, D.M. 2011. “Messianic Media: Notes on the Real-Time Stream”. In Stunlaw: A Critical Review of Politics, Arts and Technology. September 11. Accessed at: http://stunlaw.blogspot.se/2011/09/messianic-media-notes-on-real-time.html (June 27, 2017).
Bucher, T. 2012. Programmed Sociality: A Software Studies Perspective on Social Networking Sites. Doctoral Dissertation, University of Oslo. Accessed at: http://tainabucher.com/wp-content/uploads/2009/08/Bucher_Ph.D.diss_.pdf (June 27, 2017).
Cheney-Lippold, J. 2011. “A New Algorithmic Identity: Soft Biopolitics and the Modulation of Control”. Theory, Culture & Society 28(6): 164–181.
Connolly, A. 2015. “Spotify’s new playlist targeting means brands can tailor ads based on your mood,” post on The Next Web, April 16. Accessed at: http://thenextweb.com/apps/2015/04/16/spotifys-new-playlist-targeting-means-brands-can-tailor-ads-based-on-your-mood/#gref (June 27, 2017).
Crawford, K. 2015. “Can an Algorithm Be Agonistic: Ten Scenes from Life in Calculated Publics”. Science, Technology & Human Values 41(1): 77–92.
Dilley, R.M. 2002. “The problem of context in social and cultural anthropology”. Language & Communication 22(4): 437–456.
Dredge, S. 2013. “Spotify introduces Browse page to help people find streaming music playlists”. The Guardian, August 5. Accessed at: http://www.theguardian.com/technology/appsblog/2013/aug/05/spotify-browse-music-ios-android (June 27, 2017).
Drew, R. 2005. “Mixed Blessings: The Commercial Mix and the Future of Music Aggregation”. Popular Music and Society 20(4): 533–551.
Durham, J.P. 2010. “Broadcasting and schizophrenia”. Media, Culture & Society 32(1): 123–140.
Ehrenreich B. 2009. Smile or die: how positive thinking fooled America and the world. London: Granta.
Ellis, J. 2000. Seeing Things: Television in the Age of Uncertainty. London: I.B. Tauris.
Fleischer, R. 2015. “Towards a Postdigital Sensibility: How to get Moved by too Much Music”. Culture Unbound 7: 255–269.
Freeman E. 2010. Time Binds: Queer Temporalities, Queer Histories. London, Duke University Press.
Fuchs C, K. Boersma, A. Albrechtslund et al. (eds). 2012. Internet and surveillance: the challenges of Web 2.0 and social media. London: Routledge.
Galloway, A.R. 2006. “Language Wants to Be Overlooked: On Software and Ideology”. Journal of Visual Culture 5(3): 315–331.
Hallinan, B., and T. Striphas. 2014. “Recommended for you: The Netflix Prize and the Production of Algorithmic Culture”. New Media & Society 18(1): 117–137.
Hearn A. 2010. “Structuring feeling: Web 2.0, online ranking and rating, and the digital ‘reputation’ economy”. Ephemera - Theory & Politics in Organization 10(3-4): 421–438.
Hjorth L., and S.S. Lim. 2012. “Mobile intimacy in an age of affective mobile media”. Feminist Media Studies 12(4): 477–484.
Horton, D., and R.R. Wohl. 2006 . “Mass Communication and Para-Social Interaction: Observations on Intimacy at a Distance”. Particip@tions 3(1).
Introna, L.D. 2016. “Algorithms, Governance, and Governmentality on Governing Academic Writing”. Science, Technology & Human Values 41(1): 17–49.
Jenkins, H. 2005 . Textual Poachers: Television Fans and Participatory Culture. New York & London: Routledge.
Johansson, D. 2013. “From Products to Consumption—Changes on the Swedish Music Market as a Result of Streaming Technologies”. Working Paper, Linnaeus University, Växjö, Sweden.
Jones, K. 2005. “Music in factories: a twentieth-century technique for control of the productive self”. Social & Cultural Geography 6(5): 723–744.
Jones S.C. & Schumacher T.G. 1992. “Muzak: On functional music and power”. Critical Studies in Mass Communication 9(2): 156–169.
Korczynski M. 2011. “Stayin’ Alive on the factory floor: An ethnography of the dialectics of music use in the routinized workplace”. Poetics 39(2): 87–106.
Mager, A. 2012. “Algorithmic Ideology”. Information, Communication & Society 15(5): 769–787.
Manovich, L. 2001. The Language of New Media. Cambridge & London: MIT Press.
Morozov, E. 2013. To Save Everything, Click Here: The Folly of Technological Solutionism. New York: Public Affairs.
_______. 2011 The net delusion: the dark side of internet freedom. New York, Public Affairs.
Morris, Jeremy. 2011. “Sounds in the Cloud: Cloud Computing and the Digital Music Commodity”. First Monday 16(5).
Neyland, D. 2016. ‘‘Bearing Accountable Witness to the Ethical Algorithmic System’’. Science, Technology, & Human Values 41(1): 50–76.
“Nylund Hagen, A. 2015. “The Playlist Experience: Personal Playlists in Music Streaming Services”. Popular Music and Society, 38(5): 625-645.
Peterson, T. 2015. “Spotify to Use Playlists as Proxy for Targeting Ads to Activities, Moods”. Advertising Age, April 16. Accessed at: http://adage.com/article/digital/spotify-playlists-gauge-moods-ad-targeting/298066/ (June 27, 2017).
Phillips, A. 2012. “List”. In Inventive Methods: The Happening of the Social, edited by C. Lury & N. Wakeford, 96-109. London & New York: Routledge.
Prichard C., M. Korczynski and M. Elmes. 2007. “Music at work: An introduction”. Group &Organization Management 32(1): 4–21.
Scharff C. 2016. The Psychic Life of Neoliberalism: Mapping the Contours of Entrepreneurial Subjectivity. Theory, Culture & Society 33(6): 107–122.
Seaver, N. 2015. “The nice thing about context is that everyone has it”. Media, Culture & Society 37(7): 1101–1109.
_______. 2012. “Algorithmic Recommendations and Synaptic Functions”. Limn 2. Accessed at: https://limn.it/algorithmic-recommendations-and-synaptic-functions/ (June 27, 2017).
Shirky C. 2008. Here comes everybody: the power of organizing without organizations. New York, Penguin Press.
Skågeby, J. 2011. “Slow and Fast Music Media: Comparing Values of Cassettes and Playlists”. Transformations 20: 1–12.
Spedale S., C. Coupland and S. Tempest. 2014. “Gendered Ageism and Organizational Routines at Work: The Case of Day-Parting in Television Broadcasting”. Organization Studies 35(11): 1585–1604.
Spotify Press Release. 2015. Say Hello to the Most Entertaining Spotify Ever, May 20. Accessed at: https://press.spotify.com/us/2015/05/20/say-hello-to-the-most-entertaining-spotify-ever/ (June 27, 2017).
Stanfill M. 2015. “The interface as discourse: The production of norms through web design”. New Media & Society 17(7): 1059–1074.
Styvén, M. 2007. “The Intangibility of Music in the Internet Age”. Popular Music and Society 30(1): 53–74.
Tacchi, J. 1998. “Radio texture: between self and others”. In Material Cultures: Why Some Things Matter, edited by D. Miller, 25–45. London: UCL Press.
Turner, F. 2006. From counterculture to cyberculture: Stewart Brand, the Whole Earth Network, and the rise of digital utopianism. Chicago, Ill.: University of Chicago Press.
Warman, M. 2013. “Spotify Browse Suggests Music for Your Mood”. The Telegraph August 5. Accessed at: http://www.telegraph.co.uk/technology/news/10222908/Spotify-Browse-suggests-music-for-your-mood.html (June 27, 2017).
Weltevrede, E., A. Helmond, and C. Gerlitz. 2014. “The Politics of Real-time: A Device Perspective on Social Media Platforms and Search Engines”. Theory, Culture & Society 31(6): 125–150.
Werner, A., and S. Johansson. 2016. “Experts, dads and technology. Gendered talk about online music”. International Journal of Cultural Studies 19(2): 177–192.
Wikström, P., and R. Burnett. 2009. “Same Songs, Different Wrapping: The Rise of the Compilation Album”. Popular Music and Society 32(4): 507–522.
Zerubavel, E. 1981. Hidden rhythms: schedules and calendars in social life. Berkeley: University of California Press.
Ziewitz, M. 2015. “Governing Algorithms: Myth, Mess, and Methods”. Science, Technology, & Human Values 41(1): 3–16.