Effects of algorithmic curation in users’ music taste on Spotify

Abstract

This study addresses the potential impact of recommendation algorithms on Spotify users' musical tastes with the aim of understanding how algorithmic suggestions shape listening behaviors and preferences. A comprehensive review of the literature reveals that the presence of algorithms has contributed to reduced musical diversity and increased taste tautology among users. The results suggest that recommendation algorithms reinforce prior preferences, leading to the emergence of filter bubbles This algorithm-driven taste has obvious cultural implications and, with it, a large impact on the overall diversity of the musical experience. A qualitative methodology was used, consisting of a systematic literature review based on the PRISMA framework, identifying trends and key elements of existing studies. This study finds its limitations in the need for an additional quantitative study to delve deeper into the behavior of recommendation algorithms. Ultimately, this research underscores the need for greater awareness of the implications of music recommendation using algorithms in the digital age.

Full text

125
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
Recebido: 14-10-2024 | Aprovado: 20-12-2024 | DOI: https://doi.org/10.23882/cdig.24258
Effects of algorithmic curation
in users’ music taste on Spotify
Efeitos da seleção algorítmica
no gosto musical dos usuários do Spotify
Marta Ezquerra Fernández,
Universidad Complutense de Madrid, España
(martaezq@ucm.es)
Abstract: This study addresses the potential impact of recommendation algorithms
on Spotify users' musical tastes with the aim of understanding how algorithmic
suggestions shape listening behaviors and preferences. A comprehensive review of
the literature reveals that the presence of algorithms has contributed to reduced
musical diversity and increased taste tautology among users. The results suggest
that recommendation algorithms reinforce prior preferences, leading to the
emergence of filter bubbles This algorithm-driven taste has obvious cultural
implications and, with it, a large impact on the overall diversity of the musical
experience. A qualitative methodology was used, consisting of a systematic
literature review based on the PRISMA framework, identifying trends and key
elements of existing studies. This study finds its limitations in the need for an
additional quantitative study to delve deeper into the behavior of recommendation
algorithms. Ultimately, this research underscores the need for greater awareness of
the implications of music recommendation using algorithms in the digital age.
Keywords: Spotify, algorithmic curation, platformization, taste, calculated publics,
filter bubbles
Resumo: Este estudo aborda o impacto potencial dos algoritmos de recomendação
nos gostos musicais dos usuários do Spotify, com o objetivo de compreender como
as sugestões algorítmicas moldam os comportamentos e preferências de escuta.
Uma revisão abrangente da literatura revela que a presença de algoritmos contribuiu
para a redução da diversidade musical e o aumento da tautologia de gosto entre os
usuários. Os resultados sugerem que os algoritmos de recomendação reforçam
preferências anteriores, levando ao surgimento de filter bubbles. Esse gosto
impulsionado por algoritmos tem implicações culturais evidentes e, com isso, um

126
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
grande impacto na diversidade geral da experiência musical. Foi utilizada uma
metodologia qualitativa, composta por uma revisão sistemática da literatura
baseada no protocolo PRISMA, identificando tendências e elementos-chave dos
estudos existentes. Este estudo encontra suas limitações na necessidade de um
estudo quantitativo adicional para aprofundar a compreensão do comportamento
dos algoritmos de recomendação. Em última análise, esta pesquisa ressalta a
necessidade de maior conscientização sobre as implicações da recomendação
musical por meio de algoritmos na era digital.
Palavras-chave: Spotify, curadoria algorítmica, plataformização, gosto, públicos
calculados, bolhas de filtragem
1. Introduction
Our listening habits have changed and so has the music industry: during the last
twenty years, there has been a shift from an ownership-based to a streaming
platform-based business model. In 1999, Napster was launched, being the first
digital music platform that developed a peer to peer file sharing model. (Sifferd,
2002). The arrival of websites and applications such as Napster or YouTube shifted
our listening habits by providing an anytime availability of all kinds of music, as
later would do Spotify. Music consumption has changed radically ever since and so
has changed music curation. Traditional gatekeepers such as music journalists and
radio programmers no longer controlled what the user listened to (Bonini &
Gandini: 2019), and instead, listeners found the freedom of accessing an unlimited
library of digital music without intermediators (Hesmondhalgh & Meier, 2018).
However, years later, the user sovereignty in the music industry was proven to be
far from everlasting, but rather an illusion. The progressive acquisition of a platform
economic model, the arrival of algorithms and the crescent datafication of the
listener’s habits make it more and more difficult to elucidate who is in command of
the music we listen to. In this sense, streaming platforms and listening dynamics
are under the scope of platformization and algorithmic logics of digital platforms.
The model of music streaming platforms has been widely analyzed yet the effects
of algorithmic curation on streaming platform users remains relatively
underexplored. This study aims to analyze the possible changes in the variety of
music Spotify users listen to as a consequence of an algorithmic curation of music
in the platform.

127
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
2. Objectives & methodology
2.1. Objectives
The main objective of this research is to explore the possible impact of music
recommendation systems on the taste of Spotify listeners. Additionally, the study
aims to understand how the use of algorithms may affect the diversity of music
consumption in a music streaming platform. For these purposes, a systematic
analysis of the recent literature will be carried out.
2.2. Methodology
This research involves a systematic literature review to analyze existing literature
on the effects of Spotify’s algorithmic systems on listener’s taste. The PRISMA
(Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020
protocol was used to ensure that the review process was transparent, comprehensive
and exhaustive (Dialnet, n.d.). Eligible studies comprised journal articles,
conference papers, books, book chapters and reports. Studies were included if
● They were published between 2009 and 2024
● They were published in English or Spanish.
● They address algorithmic curation and its impact on user behavior and
music consumption or the economic and cultural effects of platformization
on the music industry.
Analyzed texts have been peer-reviewed studies on Spotify's algorithms as well as
studies focusing on the cultural, social, and economic impacts of algorithmic
music curation. In the systematic review, out of an initial pool of over 200 articles,
80 articles passed the screening phase and a final sample of 30 studies that
followed the eligibility criteria were selected for the final review, excluding those
articles that did not reflect the subject matter (38), that were written in other
languages (n=7), or were published in unverified sources (n=5). The recovered
studies were key in providing detailed insights into how Spotify's algorithmic
systems have had a meaningful impact on the curation of cultural products, and
the crescent power of platformization in the music industry.

128
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
Table 1. Systematic selection using PRISMA 2020 Protocol
Records selection
Sample
Initial records
n=206
Studies
Identification via
databases
Records deleted prior to selection
n=126
Screened studies
n=80
Excluded
studies
Reason 1 (subject)
n=38
Reason 2 (language)
n=7
Reason 3 (source)
n=5
Total number of studies included
n=30
The main authors that have guided this research are Jose Van Dijck, David Nieborg,
Tarleton Gillespie and Thomas Poell, whose theories of platformization of cultural
industries, datafication and algorithmical production of calculated publics were
fundamental in the understanding of streaming platforms. Also noteworthy is the
growing body of literature that in recent years has analyzed the influence of
algorithms on music, taking into consideration the work of Robert Prey on the
datification of listening and platformization of music, as well as the “algo-torial”
curation of music proposed by Bonini and Gandini.
3. Results
Spotify has become a key agent in the music industry over the past decades as well
as an open door to music democratization. Over the past years, Spotify has
implemented a new algorithmic system to display music under the purpose of
offering tailored content (Jacobson, 2016), which has permeated its users’ platform
experience (Björklund, 2022). The Swedish platform’s intention to provide custom
playlists has been pursued with commitment. As the user opens the app, the Explore
page offers an array of playlists made for them. “100% you” offers a series of daily
mixes based on the user’s listening habits, “Daily route: a music and news mix,
made for you”. As we scroll down, dozens of widgets display music under the same
premise: personalized radios according to the user's most listened artists, as well as
mixes categorized by listened genres, moods and even aesthetics.

129
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
This digital platform devoted to music streaming has progressively changed their
streaming model towards an almost entirely datafied recommendations model. And
due to the logic that underlies it, it becomes essential to understand its nature as a
platform and how it affects the relationship with its users.
3.1. Music streaming in platform economy
The term "platform", understood as part of the digital ecosystem, appeared not too
long ago to refer mainly to digital media intermediaries. Gillespie (2010) elaborates
a radiograph of the term, exploring the different areas on which it depends, stating
that the platform has a computational connotation, as something to build upon and
innovate from; political, a place from which to speak and be heard; figurative, in
that the opportunity is an abstract promise as much as a practical one; and
architectural, in that platforms are designed as open-armed, egalitarian facilitation
of expression, not an elitist gatekeeper with a normative and technical restriction.
One of Gillespie's most interesting contributions is the fact that a platform is used
to elevate someone from the rest. And this view is not trivial, as the concept of
digital platforms has gained particular traction among user-generated content,
streaming media, blogging, and social computing, given that they offer an
opportunity to gain visibility, communicate, interact or sell, something which
carries, according to Gillespie, a certain populist aura.
While rising above the rest, the platform offers itself as an "egalitarian space,
promising to support those who stand upon it" for its users (Gillespie, 2010). Van
Dijck has also provided some key contributions to the conceptualization of
platforms on which the research is based, arguing that, despite presenting
themselves as such, platforms “are not neutral nor are they value-free
constructions”; and neither do they reflect the social. On the contrary, “they produce
the social structures where we live”, by constructing a specific set of norms and
values inscribed in their architectures (Van Dijck et al., 2018).
Given that most prominent content platforms are user-created (Gillespie, 2010; Van
Dijck et. al. 2019), it seems that platforms cannot be completely understood without
the concept of culture, and vice versa (Van Dijck, 2014). Culture’s adaptation into

130
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
the web 2.0 has been platform-based, as digital platforms have hosted and
transformed cultural practices (Poell et. al, 2019), giving rise, for instance, to
aspirational labor, those productive activities “that hold the promise of social and
economic capital; yet the reward system for these aspirants is highly uneven”
(Duffy, 2016). This promise appears to manifest itself on Spotify, where artists and
bands frequently witness their songs achieving significant levels of popularity
without this being necessarily matched by equivalent social or economic capital
(UMAW, n.d.).
The dominance of platforms’ new digital environment requires new platform-
adapted dynamics. The term “platformization” was first coined by Anne Helmond
in 2015, referring to the dominance of platform infrastructural and economic
models on the web, as well as “the process in which third parties make their data
platform-ready” (Helmond, 2015) Later on, several scholars, including Jose Van
Dijck, Nieborg and Poell, have provided a deep and well-grounded depiction of this
process from a multi-approach perspective, that includes business studies, critical
political economy, cultural studies and software studies. (Nieborg, Poell & Van
Dijck, 2019)
3.2. Cultural platformization in music streaming platforms
As regards this research, platformization will be studied from the cultural studies
approach. It is in fact a process that affects to a large extent cultural production.
Cultural platformization is considered by Nieborg & Poell (2018) as “the
penetration of economic and infrastructural extensions of online platforms into the
web, affecting the production, distribution, and circulation of cultural content.” An
interesting concept the authors propose in order to explain the platformization of
cultural production is the term “contingent commodities”, which refers to those
products or services offered via digital platforms that are open to constant
modification and adaptable according to “datafied feedback” of the user. This term
refers therefore to platform-dependent commodities, as their own nature is
contingent on platforms. (Nieborg & Poell, 2018) In the digital ecosystem, cultural
production is a contingent commodity since it is increasingly reliant on an oligopoly
of digital platforms, that include among them Spotify.

131
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
Culture creators are considered “platform complementors”. This view on culture
production eventually reflects its effects on consumers, seen as “end users”. The
relationship between complementor and end-user is at all times controlled by platform
companies, in their role of mediator institution (Nieborg & Poell, 2018). Both
platformization of cultural production and variation of user’s consumption of music on
Spotify are the cause and the effect of the very same phenomenon. Neither of them can
be understood if not taking into account the datafication and dataveillance. The firsts
in introducing the term “datafication” were Cukier and Scho
nberger in 2013. They
described it as the -now-imposed- trend of turning every piece of information into data
to create predictive analysis, signaling that data encompasses many things that weren’t
considered “valuable information” until nowadays.
In parallel, another shift appears, baptized as “dataveillance”, and resulting from
merging data and surveillance; it consists in the monitoring of users based on their
online data. What separates this practique from surveillance is the fact that no
specific purpose is involved, dataveillance instead consists of a permanent tracking
of data whose purposes are unknown to the user (Van Dijck, 2014). Rather than the
control over one particular person, it penetrates every fiber of the social fabric
(Andrejevic, 2012), something that finds unavoidable consequences in the social
contract between citizens and corporate platforms. The goal of dataveillance is
therefore the speculation of data. There is no particular objective, but to amass data
and produce patterns.
Regarding dataveillance, Raley (2013) points out that, though it may seem, it is not
a novel formula. It already existed in electoral processes such as the U.S. census
use of data. But there have been quantitative and qualitative shifts: not only data
exchanges are growing exponentially, showing a new “appreciation” of data, but
also large-scale data-aggregation companies have augmented, with more and more
sophisticated technologies. These changes have completely rearranged our way of
consuming and our relationship with platforms.
Datafication is carried out in most platform contexts. The activity of listening to
music is also under the scope of data speculation. Prey (2016) remarked that
datafication of listening was at a very initial stage of development, at that time. In the

132
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
article “Music Analytica: the Datafication of Listening, he sums up how any piece of
information users generate while listening to a music streaming platform is
transformed into data. The number of skips, the moment in which we stop playing a
song, or the moment when we turn up the volume are just a few examples of the
almost infinite datafied actions on our everyday listening routine. Listening
datafication itself is in fact just another case of how users constantly generate a digital
trace profitable for marketers and institutions. (Nieborg, Poell & Van Dijck, 2019)
3.3. Algorithmic production of calculated publics
In the context of platformization and dataveillance, algorithms play an important
role in user interactions and consumption inside platforms. As defined by Cormen,
an algorithm is a sequence of computational steps that transform inputs into
outputs—similar to a recipe (Cormen, 2009). In “The Relevance of Algorithms”
(2013), Gillespie reflects on his critical view of algorithms as he summarizes the
main features of algorithmic logic. Some of them become especially important
when analyzing the correlation between algorithms and music listened to by users.
Algorithms produce “calculated publics”. The providing of algorithmic-chosen
content is not fully adapted to the user, nor custom-made, but only approximate. It
reshapes the public’s sense of itself, as it integrates individuals into a certain
targeted public that may correspond only partially with their sense of self but may
eventually end up reshaping it.
The algorithmic production of calculated publics is a key element of this research.
Algorithms both participate in structuring the publics that operate within a digital
environment and also create calculated publics. In the case of music streaming
platforms, algorithms create calculated publics by claiming to know their users and
by suggesting to them to develop an affinity with certain genres or songs. Despite
Cormen’s initial definition of algorithms, we should not consider them as mere
codes that produce outputs, but as a “socially constructed and institutionally
managed mechanism” which is designed to assure a new knowledge logic for the
public (Gillespie, 2014). Thus, as in any form of platform, social and institutional
interests interfere in the normative frameworks curating the music we listen to.

133
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
3.4. Algo-torial curation in Spotify playlists
Spotify’s units of organization are its playlist; they are the means by which music
is presented to the user, as a result of a selection and combination of several songs.
Playlists are not only the basic units in which Spotify organizes music, but they
have also become a distinctive feature of the platform, which aims to arrange songs
by genre, by mood, or by the current activity of the user, as a part of the platform’s
commercial aim to provide “musical experiences” (Morris & Powers, 2015). The
selection of their songs is carried either by a person or an algorithm, as part of an
“algo-torial” logic, (Bonini & Gandini, 2019) defined as a half editorial-half
algorithmic creation of “listening agendas” of music listeners worldwide. As a part
of the platform’s quest for a differential identity based on offering experiences, the
use of algorithms in Spotify’s playlist curation is increasing (Freeman et al., 2021)
playing a crucial role in user experience on the platform (Björklund et al., 2022) as
it is used to create precise and personalized experiences for the user.
Despite this, the result of an automated curation is not equal to an editorially-curated
one. Morris (2015) proposes the term “infomediaries” to address the current role of
music recommendation algorithms, being “organizational entities that monitor,
collect, process and repackage cultural and technical usage data into an
informational infrastructure that shapes the presentation and representation of
cultural goods”. As such, algorithms collect users' past behaviors and combine them
with extensive databases, something that highlights the clear human aspect of
recommendation systems: an algorithm is nothing more than an adaptation to the
tastes of a large mass of users who feed its database, and is intrinsically dependent
on them (Beer, 2009; Morris & Powers, 2015).
Spotify’s music recommendation systems are fundamentally generated by feedback
data from listener activity and user profiles (Snickars, 2017). One of the most
revealing articles on the platform’s intricacies discloses that streaming devices are
able to discriminate ‘high-value’ listeners from ‘low-value’ listeners (Prey, 2016),
just by seeing what music is in their library and through which phone model they
are accessing. Echo Nast creates a set of affinity models to segregate high-value
listeners according to their interests, which will form different targets for
advertisers. Prey concludes: “In short, music streaming space is not only

134
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
horizontally segmented via consumer categories, it is also vertically ordered via
hierarchies of listener value and projections of future worth.” In the same vein,
Maasø (2022) details how the platform model incentivizes megahits and superstar
economies.
3.5. The birth of an algorithmic music taste
Back in 2009, Beer exposed that the music we listened to had become “a consequence
of algorithms” outlining how such influence constitutes an expression of platform
power. In the same vein, a decade later, Robert Prey (2020) would point to
algorithmic music curation as a representation of the shifting point of platform power.
Since our way to listen to music migrates to the platform ecosystem, the presentation
and the selection of music also experience changes, as they get adapted to platforms
by datafication and automated curation.
Fueled by extensive databases, algorithms contribute to a uniformizing drift in the
listener's taste. This is thoroughly explored by Kyle Chayka in “Filterworld” (2024),
where he examines how peripheral cultures have been narrowed by the weight of the
algorithmic gaze. Unlike traditional human curators, algorithms prioritize patterns of
consumption, shifting the focus from an expert-driven curation to a data-driven
personalization (Beer, 2009). This can reinforce existing tastes and preferences by
suggesting content similar to what users have previously engaged with, potentially
limiting exposure to unfamiliar cultural experiences (Gillespie, 2014). Nevertheless,
algorithmic-curated music seems to meet with dissatisfaction among users. A study
on the textual framing of music playlists (Ferwerda, 2023) revealed most users opt
for playlists labeled as generic rather than as personalized due to the assumption that
personalization implies a loop of previously listened music.
As a consequence of this automated influence on what we listen to, McCaffrey (2016)
has come up with the concept of “taste tautology”, to refer to the sense that users find
themselves trapped in an eternal loop of automated generated content, which, in terms
of Spotify, is translated to homogeneous music, same artists, same particular songs,
etcetera. Indeed, algorithmically-generated recommendations were proven to
significantly reduce consumption diversity, as determined by Anderson et al. (2021)
in a quantitative study developed by Spotify Research Scientists. These findings align

135
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
with Eli Pariser's notion of “filter bubbles” (2011), which constitute echo chambers
reinforced by algorithms that isolate us intellectually and also culturally. On the other
side, musicians also experience the consequences of the recommendation systems,
working with “algorithmic precariousness” that aggravates the instability of cultural
work (Duffy, 2020) and being forced to constantly negotiate their relationship with
platforms (Morris, 2020).
Lastly, Prey (2020) considers their music selection policy as a mechanism for
uncovering Spotify’s intention to incite a dependence on the platform. This
increasing dependence emerges separate from the traditional music industry since
it deviates from past technologies. However, what seems particularly problematic
is, as Freeman (2021) noted, the fact that an algorithm reveals itself as an agent that
is able to shape listeners’ individual tastes.
4. Discussion and Conclusion
The results show that algorithmic curation on Spotify plays a pivotal role in shaping
cultural consumption by creating calculated publics—groups of listeners segmented
based on user data (Gillespie, 2014). Through Echo Nest, Spotify categorizes users
into high-value and low-value listeners based on their consumption patterns,
reinforcing commercial hierarchies where certain users are prioritized for
monetization (Prey, 2016) and dividing taste into calculated publics.
Nowadays taste has come to be algorithm-driven and thus dependent on platform’s
decisions (Beer, 2009). These algorithms do not only recommend songs; but they
actively influence listeners' preferences by promoting content that aligns with
listeners’ prior behaviors. Such systems lead to personalized but potentially
homogenous musical experiences (McCaffrey, 2016; Snickers, 2017; Anderson,
2020). While playlists curated via a blend of human and algorithmic methods
(Bonini & Gandini, 2019) offer a customized user experience, they also contribute
to a “taste tautology” (McCaffrey, 2016), trapping listeners in loops of familiar
content and reducing the diversity of their consumption. This aligns with Pariser's
(2011) notion of filter bubbles, whereby algorithms limit exposure to new content,
creating an isolated experience that constrains cultural exploration.

136
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
To conclude, datafied intermediation leads to an increasingly homogeneous culture
-and music- consumption. The power of platforms plays a role in the monetization
of listening that can be explained through the use of algorithms, having evident
consequences in culture diffusion. Future lines of research can focus on the impact
of algorithm-driven music streaming services on the opportunities for emerging
musicians. Secondly, future efforts should focus on mitigating the narrowing effects
of algorithms to promote more diverse and inclusive musical experiences.
4. References
Anderson, A. Maystre, L. Anderson, I. Mehrotra, R. and Lalmas, M. (2020).
Algorithmic Effects on the Diversity of Consumption on Spotify. In
Proceedings of The Web Conference 2020 (WWW '20). Association for
Computing Machinery, New York, NY, USA, 2155–2165.
https://doi.org/10.1145/3366423.3380281
Andrejevic, M. (2012). Exploitation in the data mine. In C. Fuchs, K. Boersma, A.
Albrechtslund, & M. Sandoval (Eds.), Internet and Surveillance: The
Challenges of Web 2.0 and Social Media (1st ed., pp. 71-88). (Routledge
Studies in Science, Technology and Society; No. 16). Routledge.
https://doi.org/10.4324/9780203806432
Beer, D. (2009). "Power through the algorithm? Participatory web cultures and the
technological unconscious." New Media & Society, 11(6), 985-1002.
https://doi.org/10.1177/1461444809336551
Björklund, G., Bohlin, M., Olander, E., Jansson, J., Walter, C. E., & Au-Yong-
Oliveira, M. (2022). An Exploratory Study on the Spotify Recommender
System. In A. Rocha, H. Adeli, G. Dzemyda, & F. Moreira (Eds.), Information
Systems and Technologies (pp. 366–378). Springer International Publishing.
https://doi.org/10.1007/978-3-031-04819-7_36
Bonini, T., & Gandini, A. (2019). “First Week Is Editorial, Second Week Is
Algorithmic”: Platform Gatekeepers and the Platformization of Music
Curation. Social Media and Society, 5(4).
https://doi.org/10.1177/2056305119880006
Cormen, T. H. (2009). Introduction to algorithms. Cambridge, MA: MIT press.
Dialnet. (n.d.). Cómo hacer una revisión sistemática siguiendo el protocolo
PRISMA. https://dialnet.unirioja.es/servlet/articulo?codigo=8583045
Duffy, B. E. (2016). The romance of work: Gender and aspirational labour in the
digital culture industries. International Journal of Cultural Studies, 19(4),
441–457. https://doi.org/10.1177/1367877915572186
Duffy, B. E. (2020). Algorithmic precarity in cultural work. Communication and
the Public, 5(3–4). https://doi.org/10.1177/2057047320959855

137
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
Ferwerda, B., Boksjö, N., Petricioiu, N., & Wollny, C. (2023). What’s in a Name?
How Perceived Music Playlist Personalization Influences Content
Expectations. In IFIP Conference on Human-Computer Interaction (pp. 585-
589). Cham: Springer Nature Switzerland.
Freeman, S., Gibbs, M., & Nansen, B. (2022). ‘Don’t mess with my algorithm’:
Exploring the relationship between listeners and automated curation and
recommendation on music streaming services. First Monday, 27(1).
https://doi.org/10.5210/fm.v27i1.11783
Gillespie, T. (2010). The politics of platforms. New Media & Society.
https://doi.org/10.1177/1461444809342738
Gillespie, T. (2014). The relevance of algorithms. In: Gillespie T, Boczkowski PJ
and Foot KA (eds) Media Technologies Essays on Communication,
Materiality, and Society. Cambridge, MA: The MIT Press, pp. 167–194.
Helmond, A. (2015). The Platformization of the Web: Making Web Data Platform
Ready. Social Media Society, 1(2), 1-11.
https://doi.org/10.1177/2056305115603080
Hesmondhalgh, D., & Meier, L. M. (2017). What the digitalisation of music tells
us about capitalism, culture and the power of the information technology
sector. Information, Communication & Society, 21(11), 1555–1570.
https://doi.org/10.1080/1369118X.2017.1340498
Jacobson, K., Murali, V., Newett, E., Whitman, B., & Yon, R. (2016). Music
personalization at Spotify. In Proceedings of the 10th ACM Conference on
Recommender Systems (pp. 373-373).
Maasø, A., & Spilker, H. S. (2022). The Streaming Paradox: Untangling the Hybrid
Gatekeeping Mechanisms of Music Streaming. Popular Music and Society,
45(3). https://doi.org/10.1080/03007766.2022.2026923
Martin, K. (2019). Ethical Implications and Accountability of Algorithms. Journal
of Business Ethics, 160(4). https://doi.org/10.1007/s10551-018-3921-3
Mayer-Scho
nberger, V., & Cukier, K. (2013). Big data: A revolution that will
transform how we live, work, and think. Houghton Mifflin Harcourt.
McCaffrey, M. (2016). Algorithmic culture and the taste tautology: How
recommendation systems reproduce musical preferences. Journal of Cultural
Studies, 30(1), 52-68. https://doi.org/10.1080/14797585.2016.1143895
Morris, J. W. (2020). Music Platforms and the Optimization of Culture. Social
Media and Society, 6(3). https://doi.org/10.1177/2056305120940690
Morris, J. W., & Powers, D. (2015). Control, curation and musical experience in
streaming music services. Creative Industries Journal, 8(2), 106-122. DOI:
10.1080/17510694.2015.1090222
Nieborg, D. B., & Poell, T. (2018). The platformization of cultural production:
Theorizing the contingent cultural commodity. New Media & Society, 20(11),
4275–4292. https://doi.org/10.1177/1461444818769694

138
RMd • revistamultidisciplinar.com • vol.6 (4) 2024 • ISSN: 2184-5492 • e202440
Pariser, E. (2011). The filter bubble: What the Internet is hiding from you. Penguin
Press.
Poell, T., Nieborg, D., & van Dijck, J. (2019). Platformisation. Internet Policy
Review, 8(4). https://doi.org/10.14763/2019.4.1425.
Prey, R. (2020). Locating Power in Platformization: Music Streaming Playlists and
Curatorial Power. Social Media and Society, 6(2).
https://doi.org/10.1177/2056305120933291
Prey, Robert. (2016). Musica Analytica: The Datafication of Listening.
10.1057/978-1-137-58290-4_3.
Raley, R. (2013). Dataveillance and Countervailance. In:‘Raw Data’ is an
Oxymoron,ed. L. Gitelman, 121-1 46. Cambridge, MA:MIT Press.
https://raley.english.ucsb.edu/wp-content/DV-uncorrected-proofs.pdf
Sifferd, J. A. (2002). The Peer-to-Peer Revolution: A Post-Napster Analysis of the
Rapidly Developing File Sharing Technology. Vanderbilt Journal of
Entertainment & Technology Law. Vol 4-1. Article 6.
https://scholarship.law.vanderbilt.edu/cgi/viewcontent.cgi?article=1500&cont
ext=jetla
Snickars, P. (2017). More of the Same – On Spotify Radio Culture Unbound.
Journal of Current Cultural Research, 9(2): 184-211
https://doi.org/10.3384/cu.2000.1525.1792
Union of Musicians and Allied Workers. (n.d.). Justice at Spotify. We Are UMAW.
https://weareumaw.org/justice-at-spotify
Van Dijck, J. (2014). Datafication, dataism and dataveillance: Big data between
scientific paradigm and ideology. Surveillance and Society, 12(2).
https://doi.org/10.24908/ss.v12i2.4776
Van Dijck, J. Poell T. & De Waal M. (2018). The platform society: public values in
a connective world. Oxford University Press USA - OSO.
https://public.ebookcentral.proquest.com/choice/publicfullrecord.aspx?p=552
0871.
Van Dijck, J., Nieborg, D., & Poell, T. (2019). Reframing platform power. Internet
Policy Review, 8(2). https://doi.org/10.14763/2019.2.1414