Review: <i>Computing Taste: Algorithms and the Makers of Music Recommendation</i>, by Nick Seaver
Bibliographic record
Abstract
On a blustery morning in December I received an email from Spotify: my personalized "wrapped" package was here.Opening the application on my phone, I scoured the splashy slideshow visualizing my listening data over the last year: my most-played songs and artists, the affective musical categories I engaged with at different moments of the day ("minimalist poignant lit" in the morning, "sorrow sentimental easygoing" in the afternoon) and a Meyer-Briggs-spoof descriptor of my "listening personality," The Adventurer.The mass of data that Spotify has collected about me over the last year was tightly presented in an easily shareable format, revealing not only quantifiable information about my individual listening habits, but the particular persona that Spotify has chosen to describe my position as a user and music fan.The relationship between music recommendation technologies, the people who create them, and the way that music recommendation companies imagine their listeners (as exemplified by my Spotify wrapped package) is the central focus of anthropologist Nick Seaver's Computing Taste.The book is the amalgamation of Seaver's years of ethnographic fieldwork at various music technology companies, during which time he worked alongside and interviewed employees at various levels of corporate hierarchy-from interns, to industry conference attendees, to CEOs.Seaver notes the tricky navigation of access in his project: like the "black box" technologies that they create, Silicon Valley startups are difficult spaces to gain entry into, let alone garner details about their business decision-making processes.But Seaver's more broad, and seemingly simple question of "[how the] people who design and build recommender systems think about music, listeners, and taste" offers an accessible starting point from which his interlocutors are able to discuss their shifting self-conception and role in the music and technology industry (17).The book is divided into six chapters, which Seaver orders to provide a chronological development of recommendation systems and arranges around themes of information overload; listener retention; the imagined user; music and sound as information; genre and musical relationships; and data maintenance.As illustrated throughout Seaver's book, music streaming and recommendation services are not guided by a simple algorithmic process, but rather by multiple overlapping and dynamic algorithms that are monitored 93
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.011 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.002 | 0.004 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.004 | 0.005 |
| Insufficient payload (model declined to judge) | 0.009 | 0.006 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".