This thesis examines the relationship between contemporary digital media architectures, gender culture, and online self-presentation. Moving beyond the early techno-utopian idea of a fluid, “post-gender” internet, it argues that digital platforms do not dissolve gender norms. Instead, they function as probabilistic systems that amplify and structurally reproduce gender bias. Drawing on Erving Goffman’s dramaturgical theory, Judith Butler’s concept of performativity, and Laura Mulvey’s theory of the gaze, this research introduces the ideas of “High-Definition Traditionalism” and the “Algorithmic Gaze.” Driven by engagement metrics and categorization efficiency, platforms tend to reward highly recognizable and stereotypical performances of gender. As a result, complex identities are reduced to simplified, monetizable labels. Based on cross-cultural user surveys and contemporary media case studies, the study identifies a structural imbalance of agency between platforms and users. Many users engage in continuous and exhausting identity management on the digital “front stage,” navigating both peer surveillance and algorithmic misrecognition. The phenomenon described as the “One-Click Effect” further demonstrates how recommendation systems often interpret a momentary interest as a fixed identity trait, reinforcing ideological filter bubbles and limiting exposure to diverse perspectives. The thesis concludes that, because gender bias is embedded in database structures, taxonomies, and interface design, it can also be addressed through structural redesign. This research proposes practical interventions, including reducing the prominence of quantifiable metrics, introducing algorithmic reset mechanisms, and implementing more flexible classification systems to mitigate forms of “Administrative Violence.” Ultimately, it advocates for transforming digital platforms from systems of surveillance into inclusive environments that support more fluid, safe, and authentic forms of self-representation.
Questa tesi esamina la relazione tra le architetture dei media digitali contemporanei, la cultura di genere e l'autopresentazione online. Superando la prima idea tecno-utopica di un internet fluido e "post-genere", la ricerca sostiene che le piattaforme digitali non dissolvono le norme di genere. Al contrario, esse funzionano come sistemi probabilistici che amplificano e riproducono strutturalmente i bias di genere. Attingendo alla teoria drammaturgica di Erving Goffman, al concetto di performatività di Judith Butler e alla teoria dello sguardo di Laura Mulvey, questa ricerca introduce i concetti di "High-Definition Traditionalism" (Tradizionalismo ad Alta Definizione) e "Algorithmic Gaze" (Sguardo Algoritmico). Guidate dalle metriche di engagement e dall'efficienza di categorizzazione, le piattaforme tendono a premiare performance di genere altamente riconoscibili e stereotipate. Di conseguenza, le identità complesse vengono ridotte a etichette semplificate e monetizzabili. Basandosi su sondaggi interculturali rivolti agli utenti e su casi studio di media contemporanei, lo studio identifica uno squilibrio strutturale di agency tra piattaforme e utenti. Molti utenti sono coinvolti in una continua ed estenuante gestione dell'identità sulla "ribalta" digitale (front stage), navigando tra la sorveglianza dei pari e il mancato riconoscimento algoritmico (algorithmic misrecognition). Il fenomeno descritto come "One-Click Effect" dimostra inoltre come i sistemi di raccomandazione interpretino spesso un interesse momentaneo come un tratto identitario fisso, rinforzando le filter bubbles ideologiche e limitando l'esposizione a prospettive diverse. La tesi conclude che, poiché il bias di genere è radicato nelle strutture dei database, nelle tassonomie e nel design delle interfacce, esso può anche essere affrontato attraverso una riprogettazione strutturale. Questa ricerca propone interventi pratici, tra cui la riduzione della preminenza delle metriche quantificabili, l'introduzione di meccanismi di reset algoritmico e l'implementazione di sistemi di classificazione più flessibili per mitigare forme di "Administrative Violence" (Violenza Amministrativa). In definitiva, essa promuove la trasformazione delle piattaforme digitali da sistemi di sorveglianza in ambienti inclusivi che supportino forme di autopresentazione più fluide, sicure e autentiche.
Digital stage and algorithmic cage: performing gender in contemporary media environments
CHEN, RUI
2025/2026
Abstract
This thesis examines the relationship between contemporary digital media architectures, gender culture, and online self-presentation. Moving beyond the early techno-utopian idea of a fluid, “post-gender” internet, it argues that digital platforms do not dissolve gender norms. Instead, they function as probabilistic systems that amplify and structurally reproduce gender bias. Drawing on Erving Goffman’s dramaturgical theory, Judith Butler’s concept of performativity, and Laura Mulvey’s theory of the gaze, this research introduces the ideas of “High-Definition Traditionalism” and the “Algorithmic Gaze.” Driven by engagement metrics and categorization efficiency, platforms tend to reward highly recognizable and stereotypical performances of gender. As a result, complex identities are reduced to simplified, monetizable labels. Based on cross-cultural user surveys and contemporary media case studies, the study identifies a structural imbalance of agency between platforms and users. Many users engage in continuous and exhausting identity management on the digital “front stage,” navigating both peer surveillance and algorithmic misrecognition. The phenomenon described as the “One-Click Effect” further demonstrates how recommendation systems often interpret a momentary interest as a fixed identity trait, reinforcing ideological filter bubbles and limiting exposure to diverse perspectives. The thesis concludes that, because gender bias is embedded in database structures, taxonomies, and interface design, it can also be addressed through structural redesign. This research proposes practical interventions, including reducing the prominence of quantifiable metrics, introducing algorithmic reset mechanisms, and implementing more flexible classification systems to mitigate forms of “Administrative Violence.” Ultimately, it advocates for transforming digital platforms from systems of surveillance into inclusive environments that support more fluid, safe, and authentic forms of self-representation.| File | Dimensione | Formato | |
|---|---|---|---|
|
Digital Stage and Algorithmic Cage.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Digital Stage and Algorithmic Cage
Dimensione
95.42 MB
Formato
Adobe PDF
|
95.42 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/252367