1 volume 12 issue 23/2023 AUDIOVISUAL HERITAGE AND ITS USES AT THE SWISS PUBLIC BROADCASTER A DIALOGUE ON OPPORTUNIT IES AND CONSTRAINTS FOR ARCHIVES AND ACADEMICS Anne-Katrin Weber University of Basel annekatrin.weber@unibas.ch Simone Comte Radio Télévision Suisse (RTS) Simone.Comte@rts.ch François Vallotton University of Lausanne francois.vallotton@unil.ch Abstract: Conjointly conceived by an archivist at the Radio Télévision Suisse (RTS, the French speaking PSB) and two media historians, this paper aims at discussing the diversity of uses of digitised AV heritage in Switzerland through the lenses of their professional experience. It focuses on the RTS as a particularly productive case study since the RTS has not only been a pioneer in digitising AV heritage and in promoting its holdings to the broader audience, but it also actively develops new tools for internal use, in particular speech to text technologies and more recently AI-based automation for image treatment and analysis, and data- mining tools. Through a discussion of current projects at the RTS, the paper provides insights into the most recent uses and reuses of digitised AV heritage in Switzerland. Keywords: Swiss Public Service, Archival Politics, AVHeritage, Digitisation, Automated Image Analysis, Copyright, Ethics of AI 1 I n t r o d u c t i o n Audiovisual (AV) archivists and media historians develop their professional activities in different spheres. Yet, through the conservation and valorisation of AV archives they are also accustomed to collaborations, which have increased over the past decades. While archivists were traditionally concerned with developing a policy of enhancing the value of AV documents in light of their reuse for television programs, historians demanded facilitated mailto:annekatrin.weber@unibas.ch mailto:Simone.Comte@rts.ch mailto:francois.vallotton@unil.ch Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 2 access to the AV sources for research and teaching, as well as access to additional (paper) resources allowing to contextualise the AV material. In Switzerland, the dialogue between the different professional actors started in the 1990s and gained traction through debates on the digitisation of Swiss AV heritage. Since then, multiple formal and informal collaborations among archival institutions and historians have been established, leading to reuses of AV materials in- and outside the broadcasting enterprises and their traditional outlets. This paper, written by an archivist and two media historians, focuses on the transformations brought about by the digitisation of AV archives regarding access and use of AV material within the Swiss context. To do this,it focuses on the RTS, the (French-Speaking) Radio Télévision Suisse, one of the four regional enterprises which together form the Swiss Public Broadcasting Service SRG SSR (for Schweizerische Rundfunkgesellschaft—Société Suisse de Radio et Télévision). The RTS offers a fecund case study for thinking about reuses of AV material since it has not only been a pioneer in digitising its own collections and in promoting these holdings to the broader audience, but it also actively develops new tools for internal purposes, in particular speech to text technologies and more recently AI-based automation for image treatment and analysis, and data-mining tools. Through a discussion of current projects at and with the RTS, the paper provides insights into the most recent uses and reuses of digitised AV heritage in Switzerland. More precisely, the paper examines first the history of AV heritage’s digitisation in Switzerland by describing institutional frameworks and political decisions which laid the ground for digital access of AV material. We broadly cover the Swiss history of digitisation from the 1970s until today. The aim of this first part is to highlight how, from a historical point of view, the uses of AV sources have shifted in the last decades of the 20th century, from constituting raw material for producing new content to offering privileged access to collective memory and the Swiss society’s past. The paper’s second part focuses on the question of digital heritage at the RTS. It is conceived as a conversation with Simone Comte, content owner at the RTS. Simone Comte is in charge of projects aiming at improving the quality and access to archives, in particular through the creation of metadata with artificial intelligence. She is among the most knowledgeable experts regarding the development of new digital tools for AV material in the Swiss context. Whereas the first part is written as a scholarly introduction, the second part preserves the dialogic format of our exchange with Simone Comte, providing a first-hand account of current projects at the RTS. 2 “ U n - U s e s ” o f AV H e r i t a g e The reuse of AV heritage requires the acknowledgment of the sources’ value for society: their preservation necessitates financial and institutional efforts mainly supported through public funding. Without public, and thus political, support for the preservation of AV material, its reuse and circulation are indeed not possible. In Switzerland, the recognition of this basic fact arrived belatedly. In 1975, the Clottu Report (named after its principal investigator and liberal politician Gaston Clottu) published almost 500 pages on a new federal strategy for a national culture policy: no mention was made of “AV heritage” and television and radio were solely considered in their role as transmitting media, rather than as producers of content worthy to be preserved.1 This relative neglect of AV sources was still tangible 20 years later. At his lecture on the occasion of the Swiss Congress of Historical Sciences in 1994, then director of the Swiss Federal Archives Christoph Graf summarised the many difficulties faced by institutions aiming at preserving AV heritage. Graf emphasised the rapid technological transformation inducing format changes of digital support and highlighted the “extremely poor durability” of any AV support, from nitrate to CD’s. Furthermore, already in the early 1990s the digitisation of archival holdings faced the challenge of abundance: the sheer “mass” of material constituted a central issue for institutions such as the Federal Archives or public regional libraries.2 Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 3 While these preservation issues are shared across national frontiers, the Swiss case has its specificities due to the federal system and its multilingual organisation. In the case of AV heritage, the legal issues were particularly tedious since each regional broadcaster developed its own strategies for digitisation and repeatedly showed “resistance to centralization”.3 Therefore, in the absence of national initiatives to digitise and make available AV heritage, uses of AV material were mainly restricted to internal procedures by the broadcasting companies, which employed their archives as “private” reservoirs of content. Accordingly, until the 1980s academic and historical research on Swiss television was slow, not to say inexistent, and contrary to Anglo-Saxon countries where Television Studies emerged together with Cultural and Feminist Studies, research on and with televisual sources developed only hesitantly. This situation changed progressively in the 1990s. 3 F r o m R e c y c l i n g t o P r e s e r v a t i o n o f AV S o u r c e s The first change modifying the framework for re/using AV sources is introduced in 1991 when a new Federal Act on Radio and Television (RTVA) stipulates that it is henceforth mandatory to preserve broadcasts in case of future litigations. In this context, the AV material is not yet defined as heritage, but as a document or proof for potential lawsuits involving the public broadcasters and its productions: this new status of sources requires their conservation for three months at least. A long-term preservation strategy, however, is not prescribed by the new RTVA. The second, major shift in favour of reuses of AV material involves the creation of Memoriav in 1995, which functions as “Switzerland’s national network for the preservation of the country’s AV cultural heritage”. Figure 1. Memoriav. Memoriav today supports its members with the preservation of film, sound, video, and photography, and curates the open access database MEMOBASE, that aggregates metadata of AV sources from (currently) 67 Swiss institutions. https://memoriav.ch/en/ https://memobase.ch/de/start Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 4 Figure 2. Memobase. Memoriav’s launch in the mid-1990s is the result of the failure of another project, namely the creation of a centralised organisation, which included the creation of a “dépôt legal” and of a national Centre of competence that would provide offers and resources from institutions including the Swiss Federal Archives, the Swiss National Sound Archives, the Cinémathèque Suisse, the SRG SSR, and others. Judged as being too expensive and too ambitious, a Swiss compromise was found in the creation of a “réseau d’information audiovisuel” – a network of information exchange with more limited tasks.4 Since its inception in 1995, Memoriav has become the main player on the national level with regard to AV heritage, its digitisation, and circulation.5 Third, in the early 1990s, the shift to the preservation and reuse of AV material is fostered through new scientific endeavours, first within the framework of the project “History of the SSR”. This vast initiative was initiated in 1993 by then SRG SSR general director Antonio Riva and brought together historians working on the different regional broadcasters and across media. The multilingual team produced three edited volumes published in 2000, 2006, and 2012 respectively, covering the Swiss public broadcaster’s (pre-) history from the 1920s to the early 2010s.6 The scientific and collaborative effort to write this multifaceted history of the Swiss AV landscape remains unchallenged until today.7 4 D i g i t a l H e r i t a g e a t t h e R T S : To w a r d s t h e O p e n A r c h i v e The beginning of the 2000s is marked by a broad reflection on heritage policies, especially in the AV field, as well as by a debate on the establishment of a legal deposit for AV sources.8 However, initiatives for a legal deposit do not reach the stage beyond feasibility studies.9 At the SSR and its different regional broadcasters, a vast digitisation effort in coordination with Memoriav is launched. This effort initially focuses on certain programme areas (mainly regional and national news), but the digitisation gradually extends to the totality of the company’s own television production. At the same time, at the RTS, a public platform is set up to allow the online consultation of clips and programmes. Today, some of the digitized audiovisual material of the Swiss broadcasters are also available on Memobase, which includes metadata concerning the different radio / TV productions and their institutional history. https://www.srgssr.ch/en/who-we-are/the-history-of-srg https://www.srgssr.ch/fileadmin/dam/timeline/PDF/Buecher_Geschichte_der_SRG/Geschichte-der-SRG-Band-1-Teil-1_de.pdf https://www.srgssr.ch/fileadmin/dam/timeline/PDF/Buecher_Geschichte_der_SRG/Geschichte-der-SRG-Band-2_de_fr_it.pdf https://www.srgssr.ch/fileadmin/dam/timeline/PDF/Buecher_Geschichte_der_SRG/Geschichte-der-SRG-Band-3_de_fr_it.pdf https://www.rts.ch/archives/ https://memobase.ch/fr/institution/srf?start=20&context=institutions&position=22 Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 5 Figure 3. RTS archives. This public website is completed by an internal interface, which is much more comprehensive and includes substantial metadata: gradually, researchers have been able to access this professional database via personal access and/or dedicated workspaces set up in public libraries. From 2011 on, the digitisation also covers certain production documents, made available for internal use and for research upon request. Overall, and thanks to the new possibilities introduced by digitisation, the collaboration between scientific actors on the one hand, and archivists at the SRG SSR on the other has intensified in recent years. In particular, we have seen the development of fruitful synergies in terms of shared uses of documentary resources10 as well as innovative narrative formats such as the one developed on the occasion of the 50th anniversary of the RTS feature magazine Temps present.11 Figure 4. 50th anniversary of Temps present. https://bonus.rts.ch/temps-present/50-ans/index.html Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 6 Simultaneously, television scholars continue to face several challenges regarding the utilisation of AV material for scientific research: from the perspective of historical research, the AV material constitutes a central, but only one possible source. As important as the actual televisual content, for instance, are publications by the broadcaster, such as the weekly program journal Radio TV Je vois tout. Thanks to a collaboration including the editor of the journal, the RTS, and the BCU Lausanne (the University Library of Lausanne), the latter has been made available online recently on the platform Scriptorium. Furthermore, due to the federalist organisation of the SRG SSR and its regional enterprises, the national AV landscape is divided into different professional (that is internal) platforms hosted by the regional broadcasters of the SRG SSR. This parcelization, together with the overall limited public access to AV sources, represent major obstacles for research and teaching. In addition, media and television history are a very small field in Swiss academia with only a handful of scholars engaging in research. Historical scholarship is accordingly slow to develop. 5 R e / U s e s o f AV C o n t e n t a t t h e R T S : C o n v e r s a t i o n w i t h S i m o n e C o m t e Our conversation with archivists at the RTS is based on regular exchanges and common projects that have made it possible to move beyond either the commercial relationship (for the RTS the archives may be a source of monetization) or the narrowly utilitarian one (researchers sometimes find it difficult to think beyond their targeted needs). This paper represents an opportunity to record some of our conversations with Simone Comte, a senior official (“Product Owner”) at the RTS archives. Simone Comte is in charge of projects aiming at improving the quality and access to archives, and oversees the creation of metadata with artificial intelligence, the uses of archival material by multiple stakeholders, and the evolution of the database of the RTS AV collections. She also collaborates with various institutions to enhance the value of archives for the broader public. Anne-Katrin Weber: As we have mentioned in the introduction, the patrimonial dimension of Swiss AV archives has been recognized only recently. At the RTS, archivists traditionally served the production team and were responsible for preparing and preserving content for internal reuse. The journalists and radio/TV-makers needed to access rapidly “old” information as well as “old” content to be integrated into “new” features and news coverage – what William Uricchio has termed the “recombinatory practice”12 at the core of television’s economy and aesthetics. Simone Comte: Indeed, historically, the archive was primarily intended for internal reuse: journalists consult the material on a daily basis to prepare new programmes, find images for illustrations, etc. Today, the archive continues to serve this function; however, it increasingly assists researchers who wish to use its resources: we collaborate with academics on specific projects. For instance, the archive was a partner on a recent public history initiative focusing on the Lausanne industrial fair Comptoir suisse, which is part of the collective memory of the entire region and served as a starting point for an exhibition and science communication activities in schools. The RTS Archives Department and its collaborators helped the historians gather AV sources for activities with school-classes; the RTS was also directly engaged in the public event that preceded the exhibition. These collaborations are important since the archives are at the service of the lay public: the link with the broader public is crucial. Some of our archival ‘treasures’ are very successful on social networks; they allow us not only to promote our collections, but also to highlight the added value of the archive to the company itself. This again allows us to obtain additional budgets and to emphasise the importance of our work overall. Within the RTS, where archives and the work of archivists are not always valued, we need to be successful to have credibility. François Vallotton: Communication scholar Olga Kolokytha and others have recently argued that archives are “essential elements in the democratic function of PSBs”13: the question of access to AV heritage is crucial for political and cultural reasons. As we outline in the introduction, in Switzerland, the initiative to create a https://scriptorium.bcu-lausanne.ch/page/home http://www.unsiecledecomptoir.ch/ Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 7 mandatory repository for AV sources was not successful due to financial reasons as well as the federalist specificities of the Swiss archival landscape. Today, we observe a multiplication of websites and online databases that distribute historical AV content. Platforms such as notrehistoire.ch or Memobase are taken in charge by libraries and associations. (For an overview of websites and online databases that provide access to Swiss AV heritage, see https://wp.unil.ch/tvelargie/cartographie-des-archives/.) How do you evaluate the question of access to AV heritage in Switzerland as an archivist at the Swiss PSB? Simone Comte: For a long time, access—and the difficulty of it—was determined by technical and economic issues. More recently, the legal framing of AV heritage has become increasingly important. Since the RTVA ordinance of 2007 and its article 33 the RTS is obliged to archive its productions and make them available to the public. This ordinance has obviously had a major impact on us in the archive service. It has given us a more solid basis for internal and external negotiations. However, there are areas where legal issues remain complex. The question of a general(ized) access to our holdings, for instance, poses a number of problems, in particular with regard to copyright issues. The transition from the national broadcasting of content to a distribution via the world wide web obviously poses some challenges, and we have to navigate some grey areas. A good illustration of the issues at hand is the example of a source put on Facebook, namely a 1980s clip of a 13-year-old Punk walking down the street. On Facebook, followers recognised the person and posted his name and address - probably with no malicious intent! What to do in this situation? The identification came from the public, in a benevolent way. But the protection of the privacy of the filmed person was violated and the person immediately intervened to have the video removed. This is an ethically problematic case. Even if the appropriations by the audience are not mischievous, the content’s circulation beyond the controlled archive space can cause major problems. On this question of ethical decisions, we regularly collaborate with partner institutions. For the moment, in Switzerland, the legal framework allows us to use the archival holdings for experiments with facial recognition. Figure 5. Facial recognition software. The sole requirement so far is that only so-called public figures are treated (which poses the difficulty in defining what criteria delimit a public figure!). Here we have an advantage compared to other PSB institutions. Also, a lot of efforts are made to collaborate on artificial intelligence (AI) on a European level, but again, the complexity of the different http://notrehistoire.ch http://Memobase.ch/ https://wp.unil.ch/tvelargie/cartographie-des-archives/ Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 8 legal frameworks, or the lack of a legal framework, is a major obstacle. Overall, these ethical issues are very new to us: we are moving forward cautiously, in exchange with other institutions such as the EBU. Anne-Katrin Weber: You mention that the ethical issues of face recognition, and more broadly of AI in the archive, lead to new international collaborations. This topic of professional practices and their transformation is important for our own reflections, as it allows us to think about the relation of technologies, institutions, tacit and formal knowledge. Therefore, if we think more broadly of digital A/V heritage as a practice: in your experience, how does the digitisation of large holdings affect professional practices of those who work with AV heritage? Simone Comte: I was hired at the beginning of the digitisation process. The indexing of sources was done manually. The archivists retrieved data from the TV productions, decided on keywords and other metadata, etc. Therefore, the progress of manual indexing of a large set of sources mainly depended on the available human resources: the more archivists were working on the task, the quicker the holdings would be indexed. Today, part of the data is created automatically by algorithms that we do not entirely control. Indeed, algorithms are a bit like black boxes—but maybe we can come back to this question later on. Regarding the transformation of professional practices, it is evident that the digitisation has allowed us to introduce new tools that we apply to vast corpora of AV material. At the Archive department of the RTS, we work mainly with facial and voice recognition, and with speech to text. This should soon allow us to produce thematic tags that are of a higher coherence than manual indexing (but probably less subtle). For us, these tools are very valuable, because they allow journalists to quickly find content by having a maximum of descriptors, and of course, to help document the 80% of our collection that would not be found without these tools. In addition, the RTS is experimenting with a geographical and thematic descriptor, which is still in the prototype stage. We are also doing a lot of research on automated image analysis. Here, the difference with the work before digital repositories is clearly visible: manual indexing focuses on the broadcast’s main topic, the invited guests, etc. but rarely on the general “environment” such as backgrounds, weather, landscapes, or other details. Algorithms do not make such distinctions and do not prioritise content in the same manner a human eye would. Figure 6. Algorithmic image analysis. Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 9 Finally, we use automatic gender detection, which is new. In the days of manual indexing, no one specified the gender of guests, hosts, and interviewees. Digital tools now allow us to easily make statistics and thus monitor our best practices. Indeed, the request for statistics on the gender of guests’ speaking time on air came from the management, who needs this type of information to report to national bodies and adapt institutional strategies. We were able to produce data for the 19.30 newscast. However, as with all automated tools, it is not possible to guarantee total accuracy or completeness. But this is also impossible with human labour… François Vallotton: For television scholars, Digital History, and more broadly Digital Humanities, has provided new ways of working with, but also thinking about digitisation and its tools. Yet, as the editors of a recent View issue have noted, the field “is still dominated by a focus on textual studies”14 and AV sources remain comparatively marginal. For us, it is thus particularly stimulating to know what projects and digital tools an institution such as the Swiss PSB is developing. Simone Comte: We are currently working on the implementation of speaker recognition of audio and video archives, i.e., the processing of media by algorithms, and the creation of voice profiles for those public persona we wish to authenticate automatically. Figure 7. Voice profiles. This serves different purposes. First, the automated documentation facilitates the treatment of vast audio and video collections. Without automation, our archives would remain 80% silent as their content would not be easily indexed and remain hidden. Secondly, perhaps paradoxically, the automated documentation helps retain some of the https://viewjournal.eu/18/volume/7/issue/14 Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 10 knowledge about “older” archives. Indeed, recognising a voice requires having heard it before. This is easy for living personalities, much less for the past. At the RTS, the archivists who have sufficient knowledge of the archives to be able to recognise speakers in older collections have retired, or are about to do so. Building a database with voice profiles is a partial response to this loss of knowledge. Finally, automatically identifying speakers and their gender enables us to provide precise statistics on the evolution of gender parity, the speaking time of guests, etc. These statistics are requested from the SSR by the national authorities. The tools we use are mainly developed in collaboration with other institutions. Take the algorithms: the algorithms we use are open-source products. Indeed, most of the companies cannot afford to develop their own algorithms, and it would probably not be the right place to put in the effort. For example, the facial recognition tools used for processing AV sources is open source, tested by many actors who collaborate at a European level. Assessing the quality of an algorithm requires a huge amount of data to be tested, which is one of the big difficulties, and a strong argument for collaboration with other institutions. However, it is often impossible to be certain why an algorithm works, and especially why it does not work in rare but existent cases: why does it no longer recognise a celebrity in one document, when it has spotted the same person in another document? No one has the answer to this question, possibly not even the designers of the algorithms in question. Anne-Katrin Weber: This observation points back to your earlier remark on algorithms as a “black box”. You also emphasise once more the importance of collaborations. Simone Comte: Indeed, we collaborate a lot with peers, both internationally and nationally. We also regularly exchange or work with the technical university EPFL on specific projects. For example, we currently collaborate with the EPFL on a project to showcase the archives of Prix de Lausanne. This international dance competition celebrates its 50th anniversary in 2023, and we cooperate to present related AV material in an innovative way, using AI and other tools. Among other things, the project is interested in new ways of image analysis: Is it possible to detect the same sequence of movements across the 50 years of recordings, and put them together? For the history of the Lausanne competition—and the history of dance more broadly—, such a feature would be valuable and allow for instance to compare the styles and techniques over time. In this case, the RTS has digitised and sequenced the AV material, whereas the EPFL develops the analytical tools. The collaborations allow us to keep abreast of the latest technical developments, even if in practice the developments themselves are not directly used by our services. Obviously, and unlike an academic institution, we do not have the means to pay for fundamental research: exploration for exploration’s sake does not concern us, and our projects are always linked to proven needs. Any technological development must serve pre-defined objectives. François Vallotton: For us as television scholars, it is valuable to learn about such projects, since they give us new ideas for our own research and for collaborations with you: from our perspective the ongoing exchange is a fundamental way not only to get a glimpse at the “behind the scenes” of your archive, but also to receive input for our own work. For the broader public, however, these projects and their elaboration remain often hidden: the most visible signs of the digitisation of AV heritage have been your frequent activity on social networks such as Facebook and Twitter. Simone Comte: Indeed, with the emergence of digital culture, new services within the RTS have been opened that deal with the promotion of content on digital platforms, and in particular on social networks. Today, the RTS is working on solutions for the automated creation of content that can be published across platforms. For example, starting with content produced by colleagues from the news service including images and text, it is now possible to automatically create content for other platforms, for other formats. The question then arises as to whether this automatically created content constitutes a new piece of content: should all versions be archived? In the same way, we had to think about archiving the Twitter feed of RTS and its divisions: should Twitter be archived by us? In the end, it’s a question of capacity: we can’t store all this data. Indeed, for a long time, we did not even Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 11 archive RTS websites since they were not considered as content produced by the institution at first. Similarly, archive content edited for broadcast on Facebook was not archived for a long time: we have just changed that. However, we only archive the “capsule”, i.e. the content prepared by us, and not the comments and discussions of users and followers. Finally, it can be said that the archive has been transformed into data: what was previously considered a “document” is now data, integrated into digital networks. This transformation also generates new uses, such as quantitative research. For example, we are starting to get requests from researchers for raw data - in xml - which is new. Anne-Katrin Weber: The shift from “document” to “data” is central to understand the digital transformations with regard to archival practices. As data, AV material is quantified, calculated, and measured in new ways. Yet, with all these new tools and possibilities, is there not a danger that the AV archive as such is disappearing? Simone Comte: Yes, there are new challenges: not that the archive is lost, because it is always possible to go back to the sources and reconstruct their history as data. What has changed profoundly is the possibility for multiple reuses of these sources. Nowadays, data is extracted from the mass of sources: this data can potentially serve different editorial points of view, and therefore orient opinions and analysis. We obviously try not to betray the original context, but neutral description is not necessarily the priority in social networking publications. For researchers, these new practices require an even more meticulous attention to the original source. 6 C o n c l u s i o n The conversation led with Simone Comte provides a snapshot of ongoing projects at the RTS in Summer 2021 when the interview was conducted. It takes its roots in a continuous exchange with the RTS archives and its members that started two decades ago and that has proven essential for our work as media scholars and television historians. Collaborations on teaching modules, exhibitions, internships for students, and other more or less informal projects have allowed us to better understand the differences in work rhythm, goals, and work culture between the academic universe and the archival institution. Yet, as the conversation here shows, a regular exchange with the PSB archives is necessary if we wish to keep track of the many internal—and thus little publicised—projects, which may provide new avenues and inputs for our own research. The transcript does obviously not offer a comprehensive inventory of all the practices of reuse of AV material by a PSB it is necessarily a limited and situated overview. Yet, it highlights the manifold purposes of an AV collection within such an institution. As Simone Comte underlines, the RTS itself multiplies the usefulness of AV archives: it employs AV clips for public relations on social networks, as raw material for new content, or as a data set for the testing of new digital tools. The material’s reuse in broadcasts or on social media does not fundamentally alter the original documents’ function as mass media, and it constitutes the most visible “recycling” of AV sources, well known to the broader audience. Its reconfiguration from documents to data, however, does imply an epistemological shift that is important to reflect upon. As the conversation with Simone Comte reminds us, whereas early efforts for digitising AV content turned an analogue source into a digital file, more recent developments build upon the digitised collection to test tools that include facial recognition or deep learning algorithms. Here, we may wonder to which degree the archive still functions as a repository of cultural heritage—or has it turned to a laboratory or an experimental space, in which institutional and non-institutional actors converge around common goals regarding the development, testing, and improvement of algorithms? In this context, we may ask whether Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 12 a data approach to AV archives does not risk producing historically problematic narratives as sources are extracted from their original context of production and circulation? This risk of de-contextualization is already present with the reuse of AV material on social media and other platforms that promote short clips with a mainly nostalgic or humoristic approach; the “algorithmization” of AV archives potentially increases this problem and renders it more acute. Another question open to debate is whether the AV source potentially could serve objectives removed from the initial goals of archival preservation or curation. As Simone Comte emphasises, the development of software of facial recognition, for instance, does raise ethical questions that are debated internationally, but for which no final or global answer exists. What if the tools developed for AV curation are transferred into domains that employ large data sets for surveillance and policing? Is it even possible to control the reuse not only of AV material but also of the tools developed with and for it? Finally, an equally difficult question emerges in relation with the climate crisis: to what degree the experimentation with digital tools is sustainable (or not)? Do the benefits justify its costs in terms of the resources used and their carbon footprint? And more broadly, to what degree do AV archives contribute to the climate crisis, and what are the options for becoming part of the solution rather than the problem? In light of these questions, the exchange between AV archivists and media scholars seems all the more important. Only through an ongoing conversation that takes into account the expertise and experience of archivists and media scholars it seems possible to assure the best possible future for what constitutes literally the primary (re)source of our shared interest and labour. N o t e s 1. See Olivier Pradervand and François Vallotton, ‘Le patrimoine audiovisuel en Suisse : genèse, ressources, reconfigurations’, Sociétés & Représentations 35, no. 1 (2013), 27–39, https://doi.org/10.3917/sr.035.0027; For a recent and more detailed assessment of access to AV archives, see also Felix Rauh, Françios Vallotton, “Der Zugang zu den audiovisuellen Archiven oder die Kehrseite des Schlaraffenlandes” taverse. Traverse. Zeitschrift für Geschichte – Revue d’histoire 23, no.1, 2023 2. Christoph Graf, ‘Die Archivierung audiovisueller Quellen: Referat am Schweizerischen Historikertag 21. Oktober 1994’, Revue Suisse d’histoire 45, no. 5 (1995), 514-520, https://doi.org/10.5169/SEALS-81145. 3. Pradervand and Vallotton, “Le patrimoine audiovisuel,” 27. 4. Graf, “Die Archivierung audiovisueller Quellen”. 5. Felix Rauh, ‘Audiovisuelle Mediengeschichte: archivarische und methodische Herausforderungen’, Revue Suisse d’histoire 60, no 1 (2010), 23–32. 6. Markus T. Drack, ed. La Radio et la télévision en Suisse: Histoire de la Société de radiodiffusion SSR jusqu’en 1958 (Baden: hier + jetzt, 2000); Theo Mäusli and Andreas Steigmeier, eds. Radio und Fernsehen in der Schweiz: Geschichte der Schweizerischen Radio- und Fernsehgesellschaft SRG 1958–1983 (Baden: hier + jetzt, 2006); Theo Mäusli, Andreas Steigmeier, and François Vallotton, eds. Radio und Fernsehen in der Schweiz: Geschichte der Schweizerischen Radio- und Fernsehgesellschaft SRG 1983–2011 (Baden: hier + jetzt, 2012). The PDF files of these publications are available on the SRG SSR website: https://www.srgssr.ch/en/who-we-are/the-history-of-srg 7. Theo Mäusli, Andreas Steigmeier, ‘Service-Public Medien und kollektive Erinnerung’, Revue Suisse d’histoire 60, no. 1 (2010), 44–55. 8. Peter Knoepfel, Mirta Olgiati, Politique de la mémoire nationale, Etude de base de l’IDHEAP, 2005. 9. See Pradervand and Vallotton, “Le patrimoine audiovisuel”. 10. See for instance Histoire audiovisuelle du contemporain. 11. See also les “grands formats” produced by the RTS in collaboration with historians, for instance La guerre d’Algérie et les accords d’Evian sur les ondes de la Radio romande - rts.ch - Grands formats. 12. William Uricchio, ‘The Recurrent, the Recombinatory and the Ephemeral’, in P. Grainge, ed., Ephemeral Media: Transitory Screen Culture from Television to YouTube (London: British Film Institute/Palgrave MacMillan, 2011), 23–36. 13. Olga Kolokytha, Izabela Korbiel, Krisztina Rozgonyi, and Katharine Sarikakis, ‘Who’s Afraid of the Past: The Role of Archives in Shaping the Future of PSBs’, VIEW Journal of European Television History and Culture 8, no. 16 (2019), 5–17, https://doi. org/10.18146/2213-0969.2019.jethc171. 14. Andreas Fickers, Pelle Snickars, and Mark Williams, ‘Editorial Special Issue AVData in Digital Humanities’, VIEW Journal of European Television History and Culture 7, no. 14 (2018), 1–4, https://doi.org/10.18146/2213-0969.2018.jethc149. https://doi.org/10.3917/sr.035.0027 https://doi.org/10.5169/SEALS-81145 https://www.srgssr.ch/en/who-we-are/the-history-of-srg https://www.unil.ch/hist/fr/home/menuinst/recherche/poles-de-recherche/histoire-audiovisuelle-du-co.html https://www.rts.ch/archives/grands-formats/12904416-la-guerre-dalgerie-et-les-accords-devian-sur-les-ondes-de-la-radio-romande.html https://www.rts.ch/archives/grands-formats/12904416-la-guerre-dalgerie-et-les-accords-devian-sur-les-ondes-de-la-radio-romande.html https://doi.org/10.18146/2213-0969.2019.jethc171 https://doi.org/10.18146/2213-0969.2019.jethc171 https://doi.org/10.18146/2213-0969.2018.jethc149 Anne-Katrin Weber et al., Audiovisual Heritage and its Uses at the Swiss Public Broadcaster 13 B i o g r a p h i e s Dr. Anne-Katrin Weber is a television historian and a researcher at the University of Basel. She is the author of Television before TV. New Media and Exhibition Culture in Europe and the USA, 1928-1939 (AUP 2022), and has edited several volumes and journal issues, among others La télévision du téléphonoscope à Youtube: pour une archéologie de l’audiovision (with Mireille Berton, Antipodes, 2009), and an issue of View: Journal of European Television History and Culture (‘Archaeologies of Tele-Visions and –Realities,’ with Andreas Fickers, 2015). Her research has been published in Necsus. European Journal of Media Studies, Grey Room, among others; she was co-responsible for the project “Au-delà du service public: pour une histoire élargie de la télévision en Suisse, 1960 à 2000”. Simone Comte is a Product Owner at RTS where she is in charge of projects aiming at improving the quality and access to archives, in particular through the creation of metadata with artificial intelligence, their use, and the evolution of the database of the RTS audiovisual collections. She also collaborates with various institutions to enhance the value of archives for the broader public. Prof. François Vallotton is full professor of contemporary history at the University of Lausanne, where he teaches media history. Author of numerous contributions in the field of cultural and intellectual history, he is one of the moderators of the website “Histoire audiovisuelle du contemporain” and was co-responsible for the project “Au-delà du service public: pour une histoire élargie de la télévision en Suisse, 1960 à 2000”. VIEW Journal of European Television History and Culture Vol. 12, 23, 2023 URL: https://doi.org/10.18146/view.291 Publisher: Netherlands Institute for Sound and Vision in collaboration with Utrecht University, University of Luxembourg and Royal Holloway University of London. Copyright: The text of this article has been published under a Creative Commons Attribution-ShareAlike 4.0 International License. This license does not apply to the media referenced in the article, which is subject to the individual rights owner’s terms. https://wp.unil.ch/tvelargie/ https://wp.unil.ch/tvelargie/ https://wp.unil.ch/tvelargie/ https://wp.unil.ch/tvelargie/ https://doi.org/10.18146/view.291 https://creativecommons.org/licenses/by-sa/4.0/