1 Archiving as a “Future Making” Practice
Even after the so-called archival turn (Robertson 2011), television has remained somewhat marginal within memory studies, often overshadowed by literature, cinema, photography, and museum-based heritage discourses. Yet recent research and international scholarly initiatives have increasingly recognized the centrality of television in the everyday practices of remembering and forgetting. As Assmann (2008: 98) observes, memory and oblivion are not opposites but interwoven processes—active and passive, personal and collective—where “the tension between the pastness of the past and its presence becomes a key to understanding the dynamics of cultural memory”.
In this contemporary context, television no longer refers to a single technological medium, but rather to a dynamic constellation of screen-based practices, distribution formats, and access modes. From streaming and mobile platforms to digital channels, television now circulates through a hybrid media ecology in which traditional broadcast logics cohabit with new forms of algorithmic personalisation. As Harrison (2021: 24) has argued, archiving should not be understood solely as a backward-looking practice, but as a “future-making” operation: a process that shapes how societies imagine, organize, and access their collective temporalities. This is particularly evident in the popularity of biopics, reboots, and docuseries built around archival footage, which circulate mediated forms of audiovisual history across generational and national boundaries.
Similarly, streaming platforms can no longer be understood as mere “repositories”, but rather as active agents in the construction and circulation of cultural memory. The notion of a repository, commonly linked to an idea of passivity, tends to obscure the complexity of contemporary digital environments. As will be discussed, both archiving practices and access models—particularly those embedded in digital platforms—are shaped by a selective power, a form of mediation that determine what becomes visible, accessible, or narratively central. In this context, streaming platforms operate as sites of negotiation between editorial or institutional curation, carried out by human agents, and algorithmic recommendation and tagging infrastructures. They repackage, classify, and selectively re-circulate audiovisual fragments of the past.
In this context, public service broadcasters are called to renegotiate their relationship with the past: not only as custodians of national heritage, but as actors within a platform-driven environment. As Grasso (2021) notes, streaming platforms now operate as new loci of “public history”, extending historical access beyond the classroom or museum and into the algorithmic architectures of everyday media consumption.
This paper explores how streaming platforms mediate access to national television archives, and in doing so, contribute to the construction of shared cultural memory. Starting from earlier work on RaiPlay, it expands the scope to a comparative analysis of two additional European case studies: RTVE Play (Spain) and INA Madelen (France). Although institutionally dissimilar, these platforms share a common ambition: to preserve, reframe, and circulate audiovisual heritage within owned OTT platforms.
The analysis raises several guiding questions: How is archival content selected, categorized, and made visible within platform architectures? What roles do interface design and algorithmic curation play in shaping historical access and affective engagement? Can new curatorial taxonomies emerge beyond classical archival classifications? And what temporal logics—linear, fragmented, or personalised—govern the user’s encounter with mediated memory?
Bringing together insights from archive and memory studies (Assman 2008; Baron 2008; van Dijck 2009; Jacobsen and Beer 2021) and platform studies (Lotz and Hesmondhalgh 2020; Nieborg and Poell 2021) this article investigates the ambivalent interplay between cultural preservation and algorithmic mediation, between historical continuity and platform-led transformation. In doing so, it seeks to understand how the function of the public archive is being redefined in an era where memory circulates through interface, metric and infrastructure.
2 The Archive: Some Attempts at Definition1
In common understanding, archives are often conceived as “repositories” of cultural artifacts whose existence must be preserved from oblivion. The archive is thus seen as a mechanism that neutralizes the risk of losing the past. Reducing the concept of the archive to a mere container implies stasis and immutability, but above all reflects an anachronistic view. In the digital era, the archive should instead be understood as an ever-expanding space, shaped not only by the centrifugal forces that govern the formation of its holdings, but also by the technological and media infrastructures that enable its existence, accessibility, and circulation (Neiger 2011; Van Dijck 2007). Archive embodies a constant tension between the past it aims to preserve (and represent) and the future-making practices that produce “memories of the future”. As archeologist Harrison (2021: 38) puts it: “specific arrangements of materials might include not only the ‘historic’ fabric of a heritage site itself, […] but also the various technologies and display by which it is exhibited and made ‘visitable’”.
To formulate a more comprehensive definition that accounts for the archive’s internal tensions and epistemic blind spots, it is useful to revisit key 20th-century theoretical reflections on archiving. Following Foucault’s idea of hétérotopie (1984), archives symbolize:
la volonté d’enfermer dans un lieu tous les temps, toutes les époques, toutes les formes, tous les goûts, l’idée de constituer un lieu de tous les temps qui soit lui-même hors du temps, […] une sorte d’accumulation perpétuelle et indéfinie du temps dans un lieu qui ne bougerait pas.
It is possible to argue that this wide approach may suffer from an overemphasis on the infinite power of archive. In every archive, objects are collected and preserved because they are regarded as significant, valuable, and relevant for a given culture. No archive is possible outside the framework of a community, or better, of a society that give value to the specific act of remembering. Derrida (1994: 89) shifts the focus to the idea that archiving always presupposes a principle of selection and reaffirmation: “an inheritance is never gathered together […] Its presumed unity, if there is one, can consist only in the injunction to reaffirm by choosing. ‘One must’ means one must filter, sift, criticize, sort out several different possibilities that inhabit the same injunction”.
According to the French philosopher, the archive is marked by radical and necessary heterogeneity: it cannot be simply gathered or unified, but always entails an act of choice or selection. That choice involves the exercise of power, what Derrida in Mal d’archive (1995: 74) refers to as “pouvoir de consignation”, drawing on the ancient “archontic” authority: a power of unification, identification, and classification. “L’archonte de l’archive institue l’archive comme elle doit l’être, c’est à dire non seulement en exhibant le document, mais en l’établissant”. The power of consignation tends to order the unrelated into a single corpus, creating a system in which all elements revolve around an ideal unity, exerting what Derrida calls “la violence de l’unification”.
If archiving necessarily involves processes of classification, historicization, and interpretation, then the archive emerges as a spatialized distribution of power—determining not only access but also the selection of what is to be preserved, and by implication, the attribution of value. Following this perspective, Sturken (1997: 74) defines cultural memory as a “memory that is shared outside the avenues of formal historical discourse yet is entangled with cultural products and imbued with cultural meaning”. Cultural memory does not stand in opposition to the discourse of official history, but rather “entangled” with it, drawing attention to the active, continuous, and unstable processes of remembering—and therefore forgetting—that occur in socio-cultural contexts.
3 Tele-Archives and OTT platforms: Ephemerality vs Algorithmic Memory
Studies in media, memory, and digital technologies have drawn attention to the increasingly central role played by media platforms in constructing, preserving, and producing memory (Garde-Hansen 2009). The platforms environment, thus, has emerged as a crucial space for shaping both individual and collective memory—operating alongside traditional institutions such as libraries, national archives, and private collections.
Nieborg and Poell (2021) have pointed out how cultural production—and, by extension, cultural memory—has become increasingly contingent, highlighting its dependence on platforms that are “malleable and informed by datafied user feedback, open to constant revision and recirculation”. In other words, the past is no longer merely recorded and stored by media platforms; rather, the ways it is accessed, retrieved, and represented are deeply entangled with the form, function, and governance of contemporary media platforms.
Documentary theorist Baron coined the term “Archive Effect” to describe the effect produced when archival content is placed within new contexts, thereby shaping the viewer’s experience of—and relationship to—the past they reenact (2014: 18). This framework invites a reconsideration of Tele-Archives in terms of its reception and the spectatorial experiences it generates. According to Baron, the archive thus functions less as a stable repository, but as a space of affective reception, that evokes a past perceived as authentic precisely through its fragmentary, opaque, and incomplete nature. What emerges is a form of “experienced memory” a modulation of human temporality shaped by the architectures and rhythms of platform interfaces, namely a sort of a “non-human time” created by technical media systems.
Existing researches recognize the controversial role played by contemporary digital practices in shaping the archive: Van Dijck, for instance, stated that memory practices in the digital age might be conceptualized as “amalgamations” of complex “interactions between brain, embodiment, culture, and emerging technologies such as social media platforms” (2009: 158). A debated question turns around the hypothesis if these “amalgamations” could be conceptualized as “repackaged memories” or “resurfaced memories” (2009: 171). A repackaged memory is the result of an activity of selection and reframing, in the past made by professional worker who, like curators, used to operate a knowledgeable “selection” of audiovisual content to be proposed to the viewers. In the current streaming platforms landscape, this task is less and less entrusted to human curators—who operate based on qualitative criteria—and instead performed by algorithms, which select highly specific categories of memories according to users’ preferences and quantitative data, as occurs in apps such as Timehop or Apple Memories.
Jacobsen and Beer introduced the notion of “quantified nostalgia” (2021: 8) that suggests the abstraction of the memory into metric form: “what is being measured is not just the memory, there is also an attempt to use metrics to expand and reconfigure the attachments to those memories”.
These theoretical premises are essential to any critical engagement with contemporary television archives as they circulate within streaming and OTT platforms. This paper specifically investigates the tensions and transformations that occur when a legacy medium, like television, historically defined by ephemerality, is rearticulated within the logics of algorithmic memory. These tensions are central to understanding how Tele-Archives are reorganized, accessed, and shared on digital streaming platforms such as RaiPlay, RTVE Play or INA.
3.1 Archive Television as a Form of “Deep Mediatisation”
Television’s ephemerality stems not only from its original ways of liveness or serial broadcasting, but also from the instability of its material supports and the absence of systematic preservation strategies throughout its recent history: “television remains highly ephemeral—perhaps even more so than before’ requiring innovative forms of analysis and preservation to avoid entering a ‘scholarly dark age’” (Kelly 2020: 6).
Williams, in his foundational theorization of television as “flow”, captured this fleeting nature through the metaphor of uninterrupted programming sequences that challenge stable textual boundaries:
A crime in San Francisco […] began to operate in an extraordinary counterpoint not only with the deodorant and cereal commercials but with a romance in Paris and the eruption of a prehistoric monster who laid waste to New York (1974: 91–92).
Building on this, Grainge (2011) considered ephemerality as operating along two intertwined regimes: one of transmission (circulation, storage, accessibility) and one of temporality (duration, immediacy, seriality). Television’s interstitial texts like trailers, promos, or idents emerge as paradigmatic ephemeral forms: short-lived, rarely archived, yet essential to the construction of televisual experience.
The transition to streaming platform environments has significantly reconfigured this traditional model: audiovisual content once destined to disappear is now preserved in platform infrastructures, reshaped and recontextualized through algorithmic processing: “algorithmic television functions as an archive in its own right, where automated classification and recommendation algorithms shape how television content is presented and received” (Shapiro 2020: 660). “Algorithmic memory” refers to the processes through which large-scale data infrastructures and content management systems automate the selection, retrieval, and display of audiovisual archives (Taurino and Aitaki 2024). Unlike traditional archives, which rely on human curation, algorithmic memory is governed by quantitative metrics, metadata, and real-time feedback loops, organizing the past not as a linear or contextual sequence, but as a modular and responsive dataset, designed to optimize engagement across users personalised interfaces.
This transformation carries profound implications for the construction and consumption of Tele-Archives. Some scholars have proposed the term “Archive Television” to describe the new ecosystem in which archival practices are embedded within streaming logic (Taurino and Aitaki 2024); others have emphasized that algorithms, databases, and search engines should be understood as expressions of “deep mediatisation”, whose “ideology of ordering, archiving, filtering, and searching contributes to a potential change in our ordinary activities when we appropriate them for our own purposes or when we are interpelled by them in our everyday life” (Andersen 2018: 1136).
This process of “deep mediatisation” increasingly blurs the line between archival curation and user personalisation. For public broadcasters such as Rai, RTVE, and INA, television memory is reframed as “quantified nostalgia”: a datafied, emotionally charged asset designed for circulation whereby television heritage—once marked by ephemerality, tied to live transmission and limited preservation—becomes a marketable product.
3.2 Public Service Algorithm: Facing the Dilemma between Universality and Personalisation
The tension between Tele-Archives and algorithmic memory becomes particularly complex when examined through the lens of public service media. RaiPlay, RTVE Play, and INA Madelen exemplify the dilemma faced by contemporary broadcasters: how to uphold their institutional mandate (Lotz 2007) while operating within an algorithmically mediated platform economy.
As Van den Bulck and Moe (2017) argue, public service media are increasingly caught between the foundational value of universality—serving broad and diverse audiences—and the necessity of adopting personalisation technologies to remain relevant. Algorithmic technologies challenge this principle by fragmenting audiences and favoring highly individualized content (e.g. “filter bubbles”). Although some organizations embrace personalisation as a tool for outreach and modernization (NRK, YLE, BBC), others perceive it as a threat to social cohesion and to the democratic function of public broadcasting (SVT, VRT). The result is what Van den Bulck and Moe describe as “a challenge to the democratic function of broadcasting” where “the idea of public service is weakened by the logic of quantification and feedback” (2017: 11).
Bonini and Mazzoli (2022) identify this dynamic as part of a broader “crisis of imagination” (Van Es 2017) affecting many European public broadcasters, who have struggled to differentiate their streaming services from commercial platforms like Netflix or Amazon Prime Video. While RaiPlay, RTVE Play, and INA Madelen all operate under public mandates, their infrastructures and interfaces increasingly mirror the design logic of SVOD recommendation systems. In response, Bonini and Mazzoli propose two alternative frameworks: the first is grounded in agonism (Mouffe 1999), suggesting that public platforms should coexist conflictually with commercial and cooperative platforms within a pluralistic, non-hegemonic ecosystem; the second is based on “conviviality” (Illich 1973), calling for public platforms to function as convivial tools—that is, transparent, modifiable, and participatory technologies, opposed to the extractive logic of Big Tech. Public platforms should therefore cultivate a distinctive digital ethos: on the one hand, allowing for conflictual coexistence with commercial actors (agonism), and on the other, empowering users through transparent and ethically governed systems (conviviality). From this perspective, the challenge for platforms like INA Madelen or RaiPlay is not merely to algorithmically organize heritage, but to design systems of access that balance personalisation with democratic openness: engaging users as citizens rather than consumers.
In this context, the Public Service Algorithm faces a “paradoxical mandate”: to harness the affordances of personalisation while preserving core public values such as universality, pluralism, and independence. Fieiras, Vaz, and Túñez (2023), in their comparative analysis of 14 European Public Service platforms reveal divergent strategies: some platforms, like NPO or YLE, experiment with encoding normative values into algorithmic design, while others (such as BBC or DR) prioritize editorial diversity over strict personalisation. In the case of RaiPlay and RTVE Play, personalisation engines are often outsourced and configured around user engagement metrics, raising concerns around transparency and editorial control. As Fieiras et al. note, in these contexts, “what is being measured is not just memory, but a platform’s capacity to reconfigure audience attachment through metrics” (2023: 8). For platforms like RaiPlay, RTVE Play, and INA Madelen the tension between archival curation and algorithmic modulation becomes a defining challenge.
4 RaiPlay, RTVE Play, and INA: a Comparative Analysis
Television archiving has traditionally pursued two main objectives: to preserve cultural heritage and to ensure public accessibility for purposes of entertainment, education, and academic research. This tension becomes especially pronounced in the context of public service media, which must reconcile their mission of access with the need to generate sufficient revenue to ensure their long-term sustainability. Streaming platforms like RaiPlay and RTVE Play represent a hybrid opportunity: they allow television archives to circulate but at the same time they also capitalize on archival content to fill their digital catalogues, enhancing cultural capital while reducing production costs.
Yet, the integration of audiovisual archival content into platform infrastructures is far from neutral: the archive becomes a site of negotiation between institutional preservation and algorithmic circulation.
This hybridization raises new questions about how archival content is made searchable and visible: what criteria govern the classification and visibility of archival content? To what extent can new taxonomies emerge alongside traditional archival categories? How do interface design reshape visibility, accessibility, and interpretation of television heritage? And more broadly, what kinds of temporality are activated when archives are accessed through personalised, data-driven environments?
This paper addresses these questions through a comparative analysis of three European models: the public streaming platforms RaiPlay (Italy) and RTVE Play (Spain), and the archival streaming platform INA Madelen, developed by the Institut National de l’Audiovisuel (France).
These cases were selected for their distinct national contexts and shared challenges in balancing public service mandates with platform-based distribution models. The decision to compare these three public service platforms stems from multiple reasons. First and foremost, all three platforms that host archival content acquired their algorithmic recommender systems from third-party providers. Unlike other national public service broadcasters—such as YLE, VRT, or SVT—that have developed their own conceptual frameworks and technologies for algorithmic recommendation, RaiPlay did not pursue an in-house build strategy, as it required immediate access to the technology (Fieiras et al. 2022). RTVE opted instead for outsourcing via public tenders, such as the one launched in early 2020 with the EFE agency and the company Narrativa. INA also employs external tools like Whisper, TextRazor, and InaSpeechSegmenter to support metadata extraction and automatic cataloguing.
A second reason concerns access policies: all three platforms require personalised login to access full functionalities. A third rationale is quantitative in nature, relating to the breadth and diversity of the archival content available. Compared to other public service platforms—most notably BBC iPlayer—RaiPlay, RTVE Play, and INA Madelen host more extensive and varied catalogues
The comparative analysis will focus on three key analytical dimensions: (1) platform access policies and content availability; (2) interface design, including screen layout and labeling systems; and (3) curatorial strategies.
This analysis focuses primarily on how archival content is shared and accessed—what is made available, how it is presented, and through which curatorial and navigational frameworks.
The selection process behind available materials, while equally relevant, remains opaque: platforms rarely disclose the criteria guiding these decisions. Such choices likely involve a mix of editorial, legal, and rights-related factors, particularly given the complexity of reconstructing the rights chain for content dating back to the early decades of broadcasting. Investigating this would require direct interviews and represents a valuable path for future research.
4.1 Platform Access Policies and Content Availability
Both RaiPlay and RTVE Play operate as OTT services managed by their respective national public broadcasters—Rai in Italy and RTVE in Spain. Access to these platforms is free, ad-supported, and requires only a one-time user registration. Available via web browsers, mobile applications, and Smart TVs, both platforms offer live streaming of national television channels (14 in the case of RaiPlay, 5 for RTVE Play), along with an extensive on-demand catalogue comprising documentaries, films, television series, music programming, and children’s content.
The first Rai on demand platform dates back to the early 2000s, under the name of Rai Click, changed into Rai.tv on 1 January 2007. During the press conference that launched the service, the then CEO of Rai Net, Alberto Contri, declared “the BBC has announced that by mid-January it will put around 1,000 hours of video content online. We already have them, at a rate of 40 new hours a week, we’ll always be ahead of the BBC”. In 2016, Rai.tv was fully redesigned and changed its name to RaiPlay.
RaiPlay platform serves as a public national web-based archive, offering public access to a curated selection from Rai Teche, the Rai department responsible for the conservation, documentation and enhancement of the audiovisual heritage produced by the national broadcaster since its inception, in 1954. The archival section is reachable via a dropdown menu on the homepage labeled Teche Rai, and all its contents are accessible free of charge and without time limits.
Launched on 22 June 2021, RTVE Play replaced the former RTVE “a la carta” service (in operation since 2008) and integrated content from Playz, a platform aimed at younger audiences. Access to RTVE’s archival content is granted through a dedicated “Archivo” section, which can be located via the main navigation banner or search interface. While the majority of content included in this section remains freely available, a limited selection - mainly older original TV series of historical value—are only accessible through a paid subscription to the premium service RTVE Play+. Originally launched in 2020 for international audiences across the Americas, RTVE Play+ was later rebranded to distinguish it from the domestic version. It is a subscription-based platform, currently priced at €5.99 per month, and is gradually expanding into new markets, into Asia and the Pacific, including Japan, India, and Australia.
By contrast, the INA Madelen platform exemplifies a markedly different model, both in terms of its institutional framework and its access strategy. The Institut National de l’Audiovisuel (INA), established in 1975, serves as France’s national repository for all radio and television archives. In 1995, the institution expanded its mission with the launch of the Inathèque, specifically designed to provide researchers and students with controlled access to its extensive archival holdings. Unlike Rai and RTVE, archival television content in France is not hosted on the national broadcaster’s OTT platform (France.tv) but rather made available through INA’s dedicated streaming platform, INA Madelen, a subscription-based platform available at €2.99/month, accessible via a dedicated website and mobile app. It replaces the earlier INA Premium service, launched in 2015, and can be accessed directly through a dedicated window on INA’s official website (http://www.ina.fr). The name chosen “Madelen” and its logline “la platform à remonter le temps” suggest a nostalgic relationship to the past, and more specifically to television archive. It explicitly references the famous madeleine episode in Marcel Proust’s A la recherche du temps perdu (1913–1927), as noted by French newspapers at the time of its launch (Challand 2020). The Proustian reference encapsulates the idea of memory as something accessed through “morceaux” and fragments: brief, sensory epiphanies capable of restoring the emotional resonance of a long-forgotten moment. This metaphor also captures INA’s editorial strategy: from its mission to its naming, the platform is not merely intended to display historical content, but rather to activate “intergenerational nostalgia” and affective resonance through carefully curated archival fragments. As will be explored in the following sections, this curatorial strategy closely aligns with the “archive effect” as theorized by Baron.
4.1.1 Screen Interface and Labelling
A further dimension of comparison concerns the visual and navigational experience of the interface—that is, how archival television content is categorized and made visible within the screen architecture of each platform.
According to Lotz and Hesmondhalgh, home screen interfaces on smart TVs, set-top boxes, and gaming consoles as—well as those of video streaming platforms—have become “new sites of media circulation power” (2020: 386). These interfaces extend earlier mechanisms used to direct viewer attention, starting with printed television listings, later developed into electronic program guides (EPGs), and now further evolved into platform-specific browsing environments.
Recent media scholarship has increasingly emphasized the role of streaming platform interfaces—as opposed to physical devices—as curatorial and affective environments. Sanson and Steirer (2019) show how Hulu reproduces aspects of the linear television schedule within its navigation logic, preserving the temporal structure once typical of electronic program guides (EPGs). Lobato (2019) identifies in Netflix’s evolving visual layout a clear departure from video-store aesthetics—characterized by vertical cover boxes—toward a more cinematic visual grammar, with horizontal framing reminiscent of filmstrips. McKelvey and Hunt (2019), meanwhile, introduce the metaphor of “rabbit holes” to capture the immersive, non-linear trajectories that streaming interfaces enable through personalised pathways and recursive recommendation algorithms.
In all three case studies considered here—RaiPlay, RTVE Play, and INA Madelen—the interface aligns with what Lotz and Hesmondhalgh define as the category of “services” (2020: 390): platforms—like Netflix or Disney+—that curate not only content but also the visual structure, layout, and sequencing of the interface itself. These differ both from “device interfaces” (2020: 393), which are linked to the operating system of smart TVs or set-top boxes, and from “aggregator interfaces” (2020: 406), which synthesize multiple services into a single curated environment.
The first mechanism to consider is layout, that is the visual organization of content within a platform’s user interface. Layout is never neutral: the position of archival content—whether it appears at the top or bottom of the screen—significantly influences the archival content that it will be resurfaced, viewed, or selected. On RaiPlay, the “Teche Rai” section opens with two prominent layout categories: “Scelti per te” (“Chosen for you”) and “Da non perdere” (“Unmissable”). These labels compile a heterogeneous mix of TV archival content spanning different genres, formats, and periods—from 1970s drama serials to sports broadcasts and religious programming. Immediately following is a visually distinctive banner granting direct access to 70 x 70, a curated program created by television professionals to celebrate the 70th anniversary of Rai in 2024. Subsequent categories reaggregate archival content using intersecting macro-labels, which combine various logics of classification. These include chronological groupings (“Anni ’90 e 2000”), genre-based taxonomy (“Distretto Giallo”, “Grandi varietà”), and thematic indices (“La nostra musica”, “Biblioteche”). Some of these macro-categories, however, appear hybrid and overlapping, or even as what Blanco (2000) would define as “sets of sets”. For example, there is a significant degree of duplication between the sections “Sceneggiati” and “Giallo e Mistero”: paranormal-themed dramas such as Dov’è Anna (1976) and Ritratto di donna velata (1975) are listed in both. A similar overlap occurs between I grandi varietà and Intrattenimento e risate, with titles surfacing across multiple categories without clear criteria of differentiation.
Turning to the other two European platforms—RTVE Play and INA Madelen—one immediately perceives a greater degree of taxonomic rigour in the organisation and presentation of TV archival content. This structure reflects, albeit not systematically, the classic archival categories based on genre and chronology.
In particular, INA Madelen presents from its homepage a user-friendly visual mosaic just beneath the main content slider, where colour-coded labels allow users to explore five primary macro-genres: “series”, “films”, “émissions”, “documentaires” and “théâtre”. This taxonomy is reiterated at the bottom of the page, where more specific subcategories appear under each macro-genre—for instance, the “films” section is subdivided into tags such as “policier”, “saga familiale”, and “syfy” etc.
A distinctive feature of INA Madelen lies in its use of labels such as “Top of the Docs”, “Top of the Top”, and “Séries cherries”, which explicitly signal popularity metrics by highlighting the most-viewed content on the platform. These strategies, though not overtly quantitative, function as “discursive markers of value”, contributing to a shared perception of success and cultural relevance. These forms of preference signaling act as symbolic instruments of cultural power, reinforcing network effects and cumulative visibility. By amplifying content that is already privileged by the platform’s logic, they contribute to the construction of what Jacobsen and Beer define as “quantified nostalgia”: a nostalgic engagement not anchored in personal memory, but rather driven by the pervasiveness of the algorithmic infrastructure.
A similar, albeit less explicit, mechanism is visible in RaiPlay’s recommendation labels such as “Scelti per te” (“Selected for You”) and “Da non perdere” (“Unmissable”). These categories exemplify what we might term “opaque algorithmic curation”, as they operate through black-box systems of content prioritization. As Gillespie notes, these recommendation and search algorithms “exercise circulation power on the basis of decisions that are extremely opaque to both users and content producers” (2014: 168). The issue is not whether these outputs are “biased”—as no output can ever be entirely free of bias—but that they are profoundly influenced by institutional frameworks, human decision-making, and the platform’s interpretations of user behaviour.
Labelling strategies based on “popularity metrics” and “recommendations” risk detaching TV archival fragments from historical context. Revisiting Van Dijck’s notion of “amalgamation” (2009), one of the main risks of such labelling is to flat memory into a single, ahistorical axis of relevance. Yet, drawing on Didi-Huberman’s theorization of “anachronism” (2000), these distortions of time can also be generative, productive ruptures that, when archival fragments are reassembled like mosaics, open up unexpected constellations of meaning and renewed perspectives on the past.
Among the platforms examined, RTVE Play arguably offers the most taxonomically explicit and philologically consistent interface: although its opening category, “Programas míticos”, is relatively hybrid—bringing together quiz shows (e.g. 1,2,3, El precio justo), talk shows (La clave), and serialized fiction (Historias para no dormir)—the subsequent layout is structured around four clearly defined curatorial strategies. The first one might be called “collective memory”, organizing content around anniversaries of public figures’ deaths. This takes the form of dedicated sections—such as those for Pau Donés, Lola Flores, and Mario Vargas Llosa—gathering together interviews, performances, concert clips, producing a monumentalization of the audiovisual past. As Garde-Hansen (2009) observes, these “content mash-up” or “memorial formats”, are typical of digital platforms, which often reassemble archives around anniversaries, official remembrance days, or posthumous retrospectives—what Italian television used to refer to as “coccodrilli” (obituaries).
The second strategy is chronological, producing archive sections that mirror traditional archival categories (“Años ’60”, “Años ’70”, “Años ’80”, “Años ’90”, “Años 2000”). The third is genre-based, organizing content by traditional televisual genres and formats. The fourth constitutes a meta-televisual category: the section “Programas sobre el archive” includes all content in which television reflects on its own history and production. These curatorial programs engage with the way television has shaped public opinion, and how, in turn, cultural contexts and historical moments have influenced the emergence of specific genres and formats.
4.1.2 Curatorial Strategy
While the organization of content into categories and labels constitutes a first level of curation—a necessary grid that orients the user’s search and selection process, like traditional TV schedule—a deeper level emerges when we consider curation as a form of care (Re, 2024). This is not only a semantic proximity between the terms “curate” and “care”, but also an ethical alignment. If “cure” suggests healing as resolution, “care” refers to a set of actions driven by an ethic of attentiveness and responsibility.
While RTVE Play, through its “Programas sobre el archive”, offers meta-reflective content such as Días de Tele (2023) and Ochéntame otra vez (2014-2021)—a nostalgic look back at 1980s Spanish TV that echoes the long-running series Cuéntame cómo pasó (2001-2023)—INA Madelen undertakes a more institutionally approach through Rembob’INA (2018-ongoing). Produced exclusively for the platform, the program revisits the landmark moments of French television, with commentary from actors, eyewitnesses, and INA archivists.
The most notable curatorial effort is found in RaiPlay: the original format 70 x 70, produced for the 70th anniversary of public broadcasting in Italy (1954–2024), consists of 70 short episodes celebrating specific dates, facts, and figures in Rai’s institutional history. The project reflects a vision of television as a “repertoire and practice of memory” (Hagedoorn 2024): a dynamic form of active and passive remembering and forgetting, shaped by the same selective mechanisms that structure broader processes of “cultural memory”.
Yet this same curatorial impulse is not exempt from risk. As with any mnemonic process, errors and anachronisms can occur—sometimes resulting in the construction of false memories. One such example appears in RaiPlay’s commemorative programme Rai: I primi 70 anni, which includes a segment titled Divorzio – 1974. Despite its historical reference to the year of the Italian divorce referendum, the program is not an original archival recording but rather a curated anthology of Rai Teche, compiled between 1969 and 1973. As such, Divorzio represents not only a “repackaged memory”, but also a “resurfaced memory”, a mnemonic construction that, once presented to the user, risks being perceived as authentic archival footage when it is a mediated and false memory (Reyna, 2016).
5 Conclusion
As Hagedoorn (2020) reminds us, television—as a form of cultural memory—draws attention to the mediated nature of memory texts and to the politics of remembering and forgetting. The narrative reconstruction of the past, through the televisual medium, is experienced both individually and collectively, shaped by specific cultural and technological frameworks. These experiences are not only determined by user engagement, but also by the ways in which institutions make archival content accessible.
This paper has shown that streaming platforms such as RaiPlay, RTVE Play, and INA Madelen have become new gateways to public television archives or Tele-Archives. However, they also introduce risks and ambiguities. First, interface design and labelling strategies tend to replace the linear, structured logic of traditional archives with personalised and ambient programming: metrics of popularity, even when non-numerical, convey implicit hierarchies of value, reinforcing what Lotz and Hesmondhalgh call “opaque algorithmic curation” (2020). These systems generate new asymmetries of power, where archival visibility is shaped by infrastructural decisions hidden from users, displacing Derrida’s archontic power of archival authority with a more subtle and invisible “power of infrastructure”.
Second, the growing reliance on automation raises questions about the future of professional curation. As Re (2024) suggests, platforms are shifting the focus from content to the consumer, enabling “algorithmic caregiving” that risks standardization and soft control. At the same time, this shift calls for new professional figures—archivists and curators trained to navigate algorithmic infrastructures and AI tools, as seen in the audiovisual production sector with the rise of the archival producer, expertise in metadata, rights management, and editorial curation.
Ultimately, the relationship between Tele-Archives and public service streaming platforms emerges as a dynamic site of tension—one that demands ongoing scrutiny in light of the structural opacity of algorithmic memory. In this context, public broadcasters have a responsibility to ensure greater transparency regarding the AI and recommendation systems they employ, particularly when these rely on user profiling. Without safeguards, there is a growing risk that “historical television memory” will be reduced to a “marketable asset”, rather than treated as a plural, independent, and critically mediated cultural heritage. Future research should not only expand comparative analysis across additional European models, but also address the need for regulatory frameworks governing the use of algorithmic curation and AI in public service broadcasters.
References
Andersen, Jack (2018). “Archiving, Ordering, and Searching: Search Engines, Algorithms, Databases, and Deep Mediatization.” Media, Culture & Society 40(8): 1135–1150. https://doi.org/10.1177/0163443718754652
Anderson, Benedict (1983). Imagined Communities: Reflections on the Origin and Spread of Nationalism. Londra-New York: Verso.
Assmann, Aleida (2008). “Canon and Archive.” In Cultural Memory Studies: An International and Interdisciplinary Handbook, edited by Astrid Erll, and Ansgar Nünning. Berlin: Walter de Gruyter, 98.
Baron, Jaime (2013). The Archive Effect: Found Footage and the Audiovisual Experience of History. London: Routledge. https://doi.org/10.4324/9780203066935
Barra, Luca and Cecilia Penati (2013). “‘Catch-up’ con gli archivi. Digitale terreste e patrimonio Rai.” In Storie e Culture della Televisione Italiana. Milano: Mondadori.
Blanco, Matte (2000). L’inconscio come insiemi infiniti. Torino: Einaudi.
Bonini, Tiziano and Elena Maria Mazzoli (2022). “A Convivial-Agonistic Framework to Theorise Public Service Media Platforms and Their Governing Systems.” New Media & Society 24 (4): 922–941. https://doi.org/10.1177/14614448221079036
Challand, Romain (2020). “L’INA lance Madelen, son nouveau service de streaming illimité.” In Les Numeriques. https://www.lesnumeriques.com/vie-du-net/l-ina-lance-madelen-son-nouveau-service-de-streaming-illimite-n148235.html (last accessed 20-05-2025).
Derrida, Jacques (1995). Mal d’Archive: une impression freudienne. Paris: Editions Galilée.
Derrida, Jacques (1994). Specters of Marx: The state of the debt, the work of mourning, and the New international. London: Routledge.
Didi-Huberman, George (2000). Devant le temps: Histoire de l’art et anachronisme des images. Paris: Les Editions de Minuit.
Doane, Marie Anne (2002). The Emergence of Cinematic Time: Modernity, Contingency, the Archive. Cambridge, MA: Harvard University Press. https://doi.org/10.4159/9780674263024
Fieiras-Ceide, César, Vaz-Alvarez, Martìn, Túñez-López, Miguel (2023). “Designing personalisation of European public service media (PSM): trends on algorithms and artificial intelligence for content distribution.” Profesional de la información 32(3). https://doi.org/10.3145/epi.2023.may.11
Foucault, Michel (1984). “Des espaces autres. Conférence au Cercle d’études architecturales, 14 mars 1967.” Architecture, Mouvement, Continuité 5: 46–49.
Garde-Hansen, Joanne, Hoskins, Andrews and Anna Reading (2009) (eds). Save As… Digital Memories. London: Palgrave Macmillan. https://doi.org/10.1057/9780230239418
Gillespie, Tarleton (2014). “The relevance of algorithms.” In Media technologies: Essays on communication, materiality and society, edited by Tarleton Gillespie, Pablo Boczowski and Kirsten Foot, 185–193. Cambridge MA: MIT Press.
Grasso, Aldo (2020). La storia pubblica. Memoria, fonti audiovisive e archivi digitali. Milano: Vita e Pensiero.
Hagedoorn, Berber (2017). “Collective Cultural Memory as a TV Guide: ‘Living’ History and Nostalgia on the Digital Television Platform.” Acta Universitatis Sapientiae, Film and Media Studies 14: 71–94. https://doi.org/10.1515/ausfm-2017-0004
Harrison, Rodney (2021). “Heritage Practices as Future Making Practices.” In Cultural Heritage and the Future, edited by Cornelius Holtorf e Anders Högberg, 29–45. Abingdon and New York: Routledge. https://doi.org/10.4324/9781315644615-2
Hesmondhalgh, David and Lotz Amanda D. (2020). “Video screen interfaces as new sites of media circulation power.” International Journal of Communication 14: 386–409.
Hoskins, Andrew (2018). Digital Memory Studies: Media Pasts in Transition. New York: Routledge. https://doi.org/10.4324/9781315637235
Jacobsen, Benjamin and Beer David Gareth (2021). “Quantified Nostalgia: Social Media, Metrics, and Memory.” Social Media + Society 7(2). https://doi.org/10.1177/20563051211008822
Jameson, Fredric (1991). Postmodernism, Or the Cultural Logic of Late Capitalism. Durham: Duke University Press. https://doi.org/10.1215/9780822378419
Kelly, John Paul (2021). “‘This Title Is No Longer Available’: Preserving Television in the Streaming Age.” Television & New Media 23(1): 3–21. https://doi.org/10.1177/1527476420928480
Lotz, Amanda D. (2014). The Television Will Be Revolutionized. New York: NYU Press.
Mouffe, Chantal (1999). “Deliberative democracy or agonistic pluralism?” Social Research 66(3): 745–758.
Neiger, Motti; Meyers, Oren and Zandberg Eyal (2011). On Media Memory. London: Palgrave. https://doi.org/10.1057/9780230307070
Nieborg, David B., and Thomas Poell (2018). “The platformization of cultural production: Theorizing the contingent cultural commodity.” New Media & Society 20(11): 4275–4292. https://doi.org/10.1177/1461444818769694
Nieborg, David B., Poell, Thomas and Brooke Erin Duffy (2021). Platforms and Cultural Production. Cambridge: Polity.
Re, Valentina (2024). “Curation, Algorithmic ‘Caregiving’ and Collective Archival Practices. Rethinking the Archival Work of Culture in Streaming Media.” VIEW Journal of European Television History and Culture 13(26): 12–30. https://doi.org/10.18146/view.330
Reyna, Valerie F. et al. (2016). “How Fuzzy-Trace Theory Predicts True and False Memories for Words, Sentences, and Narratives.” J Appl Res Mem Cogn 1;5(1):1–9. https://doi.org/10.1016/j.jarmac.2015.12.003
Robertson, Craig (2011). “Introduction: Thinking about Archives, Writing about History.” In Media History and the Archive, edited by Craig Robertson. London: Routledge.
Roberts, Graham; Taylor, Philip M and Pronay, Nicholas (2001). The Historian, Television and Television History: A Collection. Luton: University of Luton Press.
Scaglioni, Massimo (2006). “L’immagine come fonte, come evento, come memoria. Questioni e problemi nel rapporto fra televisione e storia.” In Fare storia con la televisione, edited by Aldo Grasso. Milano: Vita e pensiero.
Shapiro, Stephen (2020). “Algorithmic Television in the Age of Large-scale Customization.” Television & New Media 21(6): 658–663. https://doi.org/10.1177/1527476420919691
Sturken, Marita (1997). Tangles Memories: The Vietnam War, the AIDS Epidemic and the Politics of Remebering. Berkeley: University of California Press. https://doi.org/10.1525/9780520918122
Taurino, Giulia and Georgia Aitaki (2024). “Television in and After the Archive: Catalogues, Databases, Interfaces and Other Ways to Organize Audiovisual Records.” VIEW Journal of European Television History and Culture 13(26): 1–11. https://doi.org/10.18146/view.353
Van den Bulck, Hilde and Hallvard Moe (2018). “Public Service Media, Universality and Personalisation through Algorithms: Mapping Strategies and Exploring Dilemmas.” Media, Culture & Society 40(6): 875–892. https://doi.org/10.1177/0163443717734407
Van Dijck, José (2009). “Mediated memories as amalgamations of mind, matter and culture.” In The body within: Art, medicine and visualization edited by Reené van de Vall and Robert Zwijnenberg, 157–172. Leiden: Brill Academy Publications. https://doi.org/10.1163/ej.9789004176218.i-228.61
This section elaborates on and extends some reflections contained in A.F. Casiraghi, Archivi televisivi e piattaforme OTT tra memoria nazionale e nostalgia intergenerazionale: RaiPlay e Rai Teche, Mimesis (forthcoming).↩︎