This paper focuses on the organization and presentation of music in the visual interface of Spotify. Rather than evaluating Spotifyâs user interface (UI) for design, this paper investigates taxonomical and folksonomic ideologies behind standard music discovery features, which impact upon categorization. The UX tool of wireframes are employed deconstruct Spotifyâs layout (Figures 5 and 7) to understand pathways users have available through wayfinding features and allowing us to observe the ratio of folksonomy-friendly functions available. Interfaces are referred to throughout this paper in the same manner as Morris and Power, they are defined as âall that greets a userâ at the face of a website or app, including content organization and the navigational (wayfinding) options (Morris & Powers, 2015, p. 110). Organizational systems in music discovery, namely taxonomy and folksonomy, constitute approaches to information categorization and classification. Music discovery is the process by which users find the music they listen to. It is apparent through this research that there are few instances of folksonomy in Spotify. One potential reason for this is due to the complications that folksonomy poses for control. Taxonomy is suited for curatorial services, including the automated recommendations that Spotify provides. Simplicity for the end user often takes precedence over the richness of data, as Hogan indicates, invisible algorithms are limited by a reductionism in presentation of information evident in single column ordering and ranked lists, which are inherently taxonomical (Hogan, 2015).
The term folksonomy emerged in the early days of the Web 2.0, when dynamic and interactive websites began to replace static webpages. Coined by Thomas Vander Wal in 2004, the term recognized sites such as Flickr and DelÂ.icÂio.Âus, which use social tagging to organize and group material. Folksonomy today still involves text-based hyperlink labels but also includes hashtags. Users create tags to categorize their own content or the preexisting content they encounter online, rather than follow recommendations from administrated menus. Folksonomies are âdescribing toolsâ and can be used as personal systems of organization and sense-making as users explore catalogs (Jeorett & Watkinson, 2015). On the other hand, taxonomy allows for musical subgenres to be neatly stacked within genres, which creates useable paths for algorithmic song recommendation. This approach to genre though is becoming increasingly complex as databases continue to grow rapidly. Connections between genres also continue to grow as musicians experiment with hybrid styles. Genre maps are becoming de-centralized and look like networks more than trees, such as The Echo Nest project, âMusic Popcornâ (refer to Figure 4).
Whilst folksonomy is appropriate for the descriptive labeling of increased amounts of music data, it does have complications in implementation, particularly regarding control. With much music content shared across commercial streaming services, presentation and discovery features are major selling points of any interface, including Spotify. Such music streaming experiences are branded, with services seeking to distinguish themselves in the saturated market. Morris and Powers describe the deliberate yet subtle administrative control in music streaming interfaces, saying that services aim to give the appeal of an âeverflowingâ and endless stream of music, whilst stratifying for âdifferent levels of consumers, and different groupings of musical consumption activitiesâ (Morris & Powers, 2015, p. 118).
Folksonomy is a tool for engaging with the musical Long Tail in streaming. At the intersection of science and economics, Wired editor, Chris Anderson encountered the Long Tail. Anderson found that where digital retailers were not restricted by what they could stock (traditional supply and demand of physical stock), having a variety of niche products drew more consumers to the service (Anderson, 2010). A major development of moving to digital music has globally exposed the Long Tail, that is, diversity and things outside of mainstream culture. Anderson, noticed the appearance of the long tail in 2004 prompted by Amazon recommendations which used collaborative and content filtering (to recommend similar items, or items other users have enjoyed). The result is a new economic pattern, as consumers may wander further in there listening with increased availability. Researching the Long Tail, Gaffney and Rafferty (2009) advocate for folksonomy as it allows for exploration of the less popular ends of music catalogs and can account for rapidly growing and changing music genre vocabularies.
Some music apps use tagging to bring new approaches to the music archive, exploiting smart technology and social media to gamify online music experiences. Adjacent to this, online folksonomy has progressed beyond simple hyperlinked text-based tags to include time-based tags (annotations) and geotags (tags linking geographical data) across networked environments. Time tags and geotags let users pinpoint moments or map material geographically. Data is shared across a variety of devices promoting a deeper way to interact with media. In the 2010s, there are now many music discovery and music-based social media apps available. Many are aggregators, filtering trending music from various music streaming and media sites (Band a Day, Next Big Sound, White Label), some offer algorithm-based recommendations (iHeartRadio, Discovr,) and are social-based music sharing sites (Cymbal, 8tracks). Innovative apps take a novel approach to music discovery, for example, Songza curates playlists based on moods, WhoSampled shows who artists have sampled in their music and for a brief year Twitterâs #music pooled music from personal twitter feeds to create a playlist (Gensler, 2014).
Innovations in the presentation of data (and the increased capacity for it storage) have seen a growth in aggregation and customization services. Infomediary apps explore the data of music streaming platforms in novel ways. Spotify-linked (companion) apps provide interesting insights into user data and create novel ways to explore its large musical database. âForgotifyâ lets you listen to neglected songs; âDrinkifyâ matches a drink with your music; âSerendipityâ shows the regions of two listeners who are listening to the same song simultaneously; âClimatuneâ, a collaboration with AccuWeather, links weather, location, and mood in playlists; The Echo Nest provides a range of equally exciting apps, including the map generating âMusic Popcornâ and âThe Wreckomenderâ (providing antithetical playlists from a chosen song). Whilst companion apps use data to provide insightful curations of music or new ways to manually explore catalogs, the first of port of call is the streaming interface which greets a user with familiar wayfinding features.
Common wayfinding features
Wayfinding features in music streaming interfaces help users navigate ever-immense streamable libraries of music. Sites such as Spotify use a range of recognizable features, including menus, headers, featured content sections and customizable search bars and filters for navigation cues. Design in music streaming is faced with the particularly daunting task of making the abstract and intangible (music) into something visual and digestible. Around the same time as the release of the iPod, Maeda wrote The Laws of Simplicity, negotiating complexity and simplicity in emerging digital design: âestablishing a feeling of simplicity in design requires making complexity consciously available in some explicit formâ (Maeda, 2006, p. 45). The field of Information Architecture (IA) grew out of the problem of designing for abundant data, addressing labels and attributes, that is, the descriptive details of content and whether the nomenclature (how the labels are named) involves folksonomy. UX design focuses on the userâs overall experience of a product or interface. In extension of IA, UX design developed around a value system based upon usability and usefulness, centered on the userâs emotional response to interactive services and products (Six, 2014). Alongside UX, UI designers work with the visual elements of the interface, including layout and color, to help interfaces achieve these design principles.
One of the primary concerns of human-computer interaction has been usability, but this has grown to also include where design can promote enriching experiences (Harper, Rodden, Rogers, & Sellen, 2008). Human goals are important to understand for experience design, as Kymäläinen says, because users reveal value in experiences of sharing, creating, collaborating, and configuring (Kymäläinen, 2015). Rosenfeld and Morville wrote the iconic âPolar Bear Bookâ, defining an information ecology. The information ecology is a system of three interdependent areas: (1) Content refers to what a service hosts, be it images, music, video, or text; (2) Context, comprised of goals and resources; and (3) Users, the target audience (Morville & Rosenfeld, 2006). Due to the visual format of music streaming sites, labels and written information needs to be succinct and relatable: âthe goal of a label is to communicate information efficiently; that is, without taking up too much of a pageâs vertical space or a userâs cognitive spaceâ (Morville & Rosenfeld, 2006, p. 82). Poor labeling can destroy a userâs confidence in the site and it can expose businesses that donât have their users in mind.
Labeling systems can be planned and unplanned. Folksonomy falls into the unplanned side, whereas taxonomy is ordered and hierarchical. Drop-down menus and scrollable featured content are planned and presented in a linear order. Streaming interfaces use lines to depict logical, hierarchical and purposeful paths. Straight lines in keep with a longstanding notion of rationality (Ingold, 2016). In planned labeling systems, users only need to learn the hierarchy of system rather than the individual labels. On the ...