1,726,386 research outputs found
An experimental study of client-side Spotify peering behaviour
Full text linkSpotify is a popular music-streaming service which has seen widespread use across Europe. While Spotify’s server-side behaviour has previously been studied, little is known about the client-side behaviour. In this paper, we describe an experimental study where we collect packet headers for Spotify traffic over multiple 24-hour time frames at a client host. Two distinct types of behaviour are observed, when tracks are being downloaded, and when the client is only serving requests from other peers. We also note wide variation in connection lifetimes, as seen in other studies of peer-to-peer systems. These findings are relevant for improving Spotify itself, and for the designers of other hybrid peer-to-peer and server-based distribution architectures
Is Spotify a sustainable model?
No full textopenLa presente ricerca è dedicata a Spotify. Ci concentreremo in particolare sulla sostenibilità della piattaforma dal punto di vista economico, del suo impatto ambientale e della retribuzione percepita dagli artisti.
Alla base di questo studio vi è il desiderio, da piccolo produttore musicale che utilizza Spotify per distribuire la propria musica, di approfondire il tema legato alla sostenibilità di essa.
Con questo lavoro si intende dimostrare come nonostante la piattaforma sia leader di mercato ed è quindi di fatto il servizio di streaming più utilizzato in assoluto, in realtà non è così sana come si potrebbe pensare.
Il primo capitolo introduce in maniera generale la piattaforma. Per prima cosa si parlerà delle motivazioni che hanno spinto i fondatori a creare questo tipo di servizio per poi passare a vedere brevemente il modello di business che caratterizza Spotify fornendo una panoramica attuale della sua posizione nel mercato anche rispetto ai principali concorrenti.
Il secondo capitolo, che è il corpo centrale del tema, è dedicato a spiegare i principali problemi collegati alla sostenibilità della piattaforma. Si analizzerà per prima cosa il conto economico mostrando come, da quando è stato lanciato, Spotify non sia mai riuscito a generare un utile netto annuo. Passeremo poi a trattare il tema dell’impatto ambientale dello streaming in termini di emissioni di gas serra, confrontando i dati con quelli dell’industria musicale pre-digitale ed infine cercheremo di capire se per gli artisti Spotify rappresenta una via concretamente percorribile per sostenere economicamente le proprie carriere.
La sezione finale di questo elaborato si concentra sul fornire delle soluzioni proposte da vari esperti del settore alle problematiche affrontate nel secondo capitolo. Come si vedrà queste soluzioni sono ancora in una fase embrionale ma comunque utili come spunto di riflessione per futuri studi sul tema
Explorativ bildning i strömmande medier : Spotify som ett case
No full textVilka bildningsprocesser kan urskiljas ur människors Spotifyanvändande? Det visar sig att ett sådant användande är tätt sammanflätat med musikalisk- såväl som digital kunskap. Med ett fokus på de bildningsprocesser som sker i samspelet mellan människa, teknik och musik utmanas förståelsen av vad en strömmande musiktjänst som Spotify kan erbjuda.”Evolving Bildung in the nexus of streaming services, art and users – Spotify as a case” är ett tvärdisciplinärt projekt som visar hur människans bildningsprocesser villkoras, utmanas och möjliggörs i och med den strömmande medieutvecklingen. Denna bok kan med fördel användas inom utbildningar i musikpedagogik, musikvetenskap, musikproduktion, ljudteknik, pedagogik, kulturstudier, sociologi, samt medie- och kommunikationsvetenskap.What educational processes take place in people's Spotify use? It turns out that such use is closely linked to musical as well as digital competences. With a focus on the educational processes that take place in the interplay between people, technology and music, the understanding of what a streaming music service like Spotify can offer is challenged."Evolving Bildung in the nexus of streaming services, art and users - Spotify as a case" is a cross-disciplinary project that shows how the processes of human education and self-formation are conditioned, challenged and made possible by the streaming media development. This book can be used in educations in music pedagogy, musicology, music production, audio engineering, pedagogy, cultural studies, sociology, and media and communication science.</p
User Privacy on Spotify: Predicting Personal Data from Music Preferences
Full text linkopenThe way we listen to music has changed drastically in the past decade. Now we can play any
kind of music from various artists around the world through our smart devices. Many music
streaming providers, if not most, are built with systems to track users’ music preferences and
suggest new content.
The music we listen to reveals a great deal about who we are. In general, people share their
playlists and songs of their favorite artists on the music platform; find people with common
music genres and connect with them. It is not always easy to make friends with unknown
people, but music is a good way to accomplish that. In spite of that, we must also look at other
sides of the coin from a security perspective. Is it a good idea to share music interests with
others or will it compromise our privacy? According to privacy experts and developers, there
is no purposeless data. Everything can be used to infer private information, even a single like
on social media, which seems, at first sight, meaningless, but it can reveal more information
than it promises. In the case that our musical tastes reveal our information, we may be profiled
for targeted advertisement, by surveillance agencies, or in general, become potential victims of
malicious activities Since music is part of our daily lives, and there are many providers that let
us listen to music, we are even more at risk of being profiled and having our data sold.
In this research, we demonstrate the feasibility of inferring personal data based on playlists
and songs people publicly shared on Spotify. Through an online survey, we collected a new
dataset containing the private information of 750 Spotify users and we downloaded around
402,999 songs extracted from a total of 8777 playlists. Our statistical analysis shows significant
correlations between users’ music preferences (e.g., music genre) and private information (e.g.,
age, gender, economic status).
As a consequence of significant correlations, we built several machine-learning models to
infer private information and our results demonstrated that such inference is possible, posing
a real privacy threat to all music listeners. In particular, we accurately predicted the gender
(71.7% f1-score), and several other private attributes, such as whether a person drinks (62.8%
f1-score) or smokes (60.2% f1-score) regularly.
The purpose of this project is to raise awareness about how seemingly purposeless data can
reveal personal information and educate users about how to better protect their privacy.The way we listen to music has changed drastically in the past decade. Now we can play any
kind of music from various artists around the world through our smart devices. Many music
streaming providers, if not most, are built with systems to track users’ music preferences and
suggest new content.
The music we listen to reveals a great deal about who we are. In general, people share their
playlists and songs of their favorite artists on the music platform; find people with common
music genres and connect with them. It is not always easy to make friends with unknown
people, but music is a good way to accomplish that. In spite of that, we must also look at other
sides of the coin from a security perspective. Is it a good idea to share music interests with
others or will it compromise our privacy? According to privacy experts and developers, there
is no purposeless data. Everything can be used to infer private information, even a single like
on social media, which seems, at first sight, meaningless, but it can reveal more information
than it promises. In the case that our musical tastes reveal our information, we may be profiled
for targeted advertisement, by surveillance agencies, or in general, become potential victims of
malicious activities Since music is part of our daily lives, and there are many providers that let
us listen to music, we are even more at risk of being profiled and having our data sold.
In this research, we demonstrate the feasibility of inferring personal data based on playlists
and songs people publicly shared on Spotify. Through an online survey, we collected a new
dataset containing the private information of 750 Spotify users and we downloaded around
402,999 songs extracted from a total of 8777 playlists. Our statistical analysis shows significant
correlations between users’ music preferences (e.g., music genre) and private information (e.g.,
age, gender, economic status).
As a consequence of significant correlations, we built several machine-learning models to
infer private information and our results demonstrated that such inference is possible, posing
a real privacy threat to all music listeners. In particular, we accurately predicted the gender
(71.7% f1-score), and several other private attributes, such as whether a person drinks (62.8%
f1-score) or smokes (60.2% f1-score) regularly.
The purpose of this project is to raise awareness about how seemingly purposeless data can
reveal personal information and educate users about how to better protect their privac
Measuring Information Diffusion in Code Review at Spotify
No full textBackground Code review, a core practice in software engineering, has been widely studied as a collaborative process, with prior work suggesting it functions as a communication network. Despite its popularity, this theory has not been formalized and remains untested, limiting its practical and theoretical significance. Objective This study aims to (1) formalize the theory of code review as a communication network explicit and (2) empirically test its validity by quantifying the extent of information diffusion---the spread of information---in code review across social, organizational, and software architectural boundaries. Method We conduct a large-scale empirical analysis of 220,733 code reviews by 2,246 developers at Spotify during 2019. We conceptualize information diffusion along three distinct boundaries: social (dissimilarity among review participants), organizational (involvement of developers across teams), and architectural (interconnections among the components under review). Results We find that over 99.6% of review pairs have completely distinct participant sets, indicating high diffusion across social boundaries. Approximately 18% of code reviews involve developers from multiple teams, evidencing nontrivial diffusion across organizational boundaries. Of the 5.82% of code reviews linked to others, 99.0% span distinct repositories, reflecting architectural diffusion. Conclusion The substantial diffusion of information across social, organizational, and architectural boundaries empirically supports the theory of code review as a communication network. These findings indicate that code review plays a role not only in quality assurance, but also in enabling communication and coordination in large-scale, distributed software projects. They further support its use as a measurable proxy for cross-border collaboration in the context of tax compliance, but also raise concerns about the impact of integrating LLMs on its communicative function
Measuring Information Diffusion in Code Review at Spotify
No full textBackground Code review, a core practice in software engineering, has been widely studied as a collaborative process, with prior work suggesting it functions as a communication network. Despite its popularity, this theory has not been formalized and remains untested, limiting its practical and theoretical significance. Objective This study aims to (1) formalize the theory of code review as a communication network explicit and (2) empirically test its validity by quantifying the extent of information diffusion---the spread of information---in code review across social, organizational, and software architectural boundaries. Method We conduct a large-scale empirical analysis of 220,733 code reviews by 2,246 developers at Spotify during 2019. We conceptualize information diffusion along three distinct boundaries: social (dissimilarity among review participants), organizational (involvement of developers across teams), and architectural (interconnections among the components under review). Results We find that over 99.6% of review pairs have completely distinct participant sets, indicating high diffusion across social boundaries. Approximately 18% of code reviews involve developers from multiple teams, evidencing nontrivial diffusion across organizational boundaries. Of the 5.82% of code reviews linked to others, 99.0% span distinct repositories, reflecting architectural diffusion. Conclusion The substantial diffusion of information across social, organizational, and architectural boundaries empirically supports the theory of code review as a communication network. These findings indicate that code review plays a role not only in quality assurance, but also in enabling communication and coordination in large-scale, distributed software projects. They further support its use as a measurable proxy for cross-border collaboration in the context of tax compliance, but also raise concerns about the impact of integrating LLMs on its communicative function
Spotify as a case of musical <em>Bildung</em>
No full textThis article explores the meaning and function of streaming media as a potential facilitator of musical Bildung. Taking the affordances of streaming media technologies as a starting point, the article thus focuses on the formative and cultivating dimensions a music streaming service such as Spotify might offer. The specific aim of this article is to describe and analyse how musical Bildung may evolve within a Spotify context from a user perspective. To address the aim from the point of view of music education, Spotify users’ activities and experiences of streaming media interactions were accessed, inspired by internet-related ethnography. Stimulated recall interviews, focusing on the participants’ experiences as well as their actual use of Spotify’s streaming service, were conducted, recorded, and transcribed. The generated material was subjected to co-operative hermeneutic content analysis. The results illuminate how Bildung evolves in users’ encounters with the service and with art mediated via Spotify. Relevant topics occurring in the human-art-technology relationship of Bildung from a Heideggerian perspective were Being-possible, the ability-to-be, and Spotify as the Other. In sum, it can be stated that Bildung evolves when Spotify exceeds the thingness of the Other, becoming a work of art in itself, throwing the user into Being.</p
Spotify For Dummies
No full textThe ultimate beginner guide to the groundbreaking music service, Spotify! Spotify is a free online streaming music platform that allows users to listen to songs on demand over the Internet--without having to buy or own the actual tracks. This fun and friendly guide walks you through how best to use this sweet-sounding service. Covering everything from using Spotify on selected mobile phones to creating and sharing your own playlists, Spotify For Dummies has it all. Experienced author Kim Gilmour details the ins and outs of this revolutionary music, from installing and setup to discovering ne
Evaluating Music Discovery Tools on Spotify: The Role of User Preference Characteristics
No full text本研究以Spotify為研究平台，探討音樂社交軟體的使用者使用不同音樂發掘工具進行音樂欣賞時的主觀評價和客觀推薦成效，以及與使用者偏好結構之間的關係。本研究以實驗法為主，一共有26位參與者，採用拉丁方格的組內設計，每位參與者都使用了4種音樂發掘工具（地區排行導覽工具、情境風格導覽工具、曲目電臺推薦工具、音樂追蹤導覽工具）在限定時間內探索並存取喜好的歌曲，所有參與者和系統互動的過程都以螢幕錄製的方式記錄下來。為了能從多維度、更準確地評估音樂發掘工具之效用，我們使用了主觀評價和客觀推薦成效兩個測量項目：(1)通過實證型的小型實驗來測量受試者之主觀評價，自變項為Spotify所提供的四種音樂發掘工具；中介變項為受試者的偏好結構（偏好洞見、偏好多樣性、偏好開放性）；依變項為實驗後問卷中收集的受試者主觀評價；(2)客觀推薦成效則由受試者在實驗中產生的曲目集合數量之比例決定，即以受測者所感興趣的曲目相較於工具所推薦的歌曲數目的比例。質化研究的部份，採用訪談法，通過實驗後對受試者進行針對性的訪談，為量化研究的結果提供檢定、補充和解釋。研究結果發現：一、不同音樂發掘工具的推薦效用的確有所差異。二、使用者面對不同音樂發掘工具時的主觀評價與客觀推薦成效並不一致。三、使用者的個人偏好結構的確會影響音樂發掘工具的推薦效用。An experimental study was conducted to assess the effectiveness of the four music discovery tools available on Spotify, a popular music streaming service, namely: radio recommendation, regional charts, genres and moods, as well as following Facebook friends. Both subjective judgment of user experience and objective measures of search effectiveness were used as the performance criteria. Other than comparison of these four tools, we also compared how consistent are these performance measures. The results show that user experience criteria were not necessarily corresponded to search effectiveness. Furthermore, three user preference characteristics: preference diversity, preference insight, and openness to novelty were introduced as mediating variables, with an aim to investigating how these attributes might interact with these four music discovery tools on performance. The results suggest that users’ preference characteristics did have an impact on the performance of these music discovery tools
Técnicas de agrupamiento aplicadas a sistemas de recomendación musical: un estudio académico basado en el reto de Spotify “Million Playlist Dataset Challenge”
No full textRESUMEN : La constante evolución de las plataformas digitales de entretenimiento, como Spotify, ha generado desafíos en la mejora continua de sus sistemas de recomendación musical, los cuales se ven exacerbados por la diversidad de gustos de los usuarios, los cambios en el estado de ánimo y el contexto de éstos, la necesidad de descubrir nueva música y la competencia en la industria del streaming. Con el objetivo de abordar estos desafíos, Spotify ha lanzado el concurso "The Spotify Million Playlist Dataset Challenge", promoviendo la investigación en algoritmos de recomendación musical para aumentar la retención de usuarios, la satisfacción del cliente y el tiempo de uso en la plataforma. En el marco del concurso, este estudio se apoya en los datos proporcionados por Spotify para desarrollar un modelo predictivo que permita recomendar canciones a partir de patrones de escucha de los usuarios, teniendo en cuenta variables como la duración de las canciones, los niveles de bailabilidad, energía, instrumentalidad y acústica, entre otros; con el propósito de mejorar la personalización y satisfacción del usuario al usar la plataforma. Los datos utilizados provienen del dataset proporcionado por Spotify para el concurso, el cual comprende más de un millón de playlists y dos millones de pistas. Para abordar esta problemática, se tomó una muestra de mil canciones, seleccionadas para representar una diversidad de géneros y características musicales. Cuando un usuario introduce una playlist, el sistema utiliza algoritmos de agrupamiento para analizar las características de las canciones y generar recomendaciones personalizadas. Este método permite agrupar canciones con atributos similares y sugerir nuevas pistas que se ajustan a los gustos y preferencias del usuario. Para lograr recomendaciones musicales efectivas, se optó por un modelo de agrupamiento K-Means con seis grupos, el cual ha demostrado ser eficaz, alcanzando un error cuadrático medio (MSE) de 0.08%, lo que indica una alta precisión y un bajo sesgo en las recomendaciones generadas, asegurando que éstas son tanto relevantes como personalizadas para los usuarios de la plataforma.EspecializaciónEspecialista en Analítica y Ciencia de Dato
- …
