CHAIR: Florence Le Cam

Hugo Ljungbäck

Restoration or Distortion? Artificial Intelligence, Early Cinema, and Media History

Over the past three years, popular interest in early film and moving image culture has seen arevival on social media, and YouTube especially, as it has become the home of hundreds of “artificial intelligence restorations”: early actualities, travelogues, and other films have been upscaled to 4K, stabilized, smoothed out at 60 fps, and colorized. Often sound is also added to make these historical images come alive in the present. The results have generated significant interest in these early films, not just from viewers, who frequently post awe-struck comments about the eerie present-ness of the past, but also news outlets, which have helped share and circulate these clips. While archivists originally may have wanted to dismiss these videos as part of a short-lived, contemporary fad, the significant viewership they have developed and the attention they continue to be paid suggest that it might be time to start taking them seriously.
Informed by a recent roundtable discussion and debate I hosted at the 2022 Association of Moving Image Archivists Conference, this presentation is an attempt to grapple with these “restoration projects” as, from the hobbyist restorers’ point-of-view, serious and sincere attempts at restoring or improving the quality of the original films and making them widely available and accessible to contemporary audiences, but, from professional archivists’ point-of-view, as creative practices that distort the media-historical value of the original images and misrepresent the work of professional restorers. My presentation considers three of the most pressing concerns AI brings to the archival field—the distortion of the image as historical index, the muddled meaning of restoration within our own field, and the potential for future collaborations with and use of artificial intelligence tools in professional restoration contexts—to tease out how we can use these clips as potential assets in larger popular conversations about the integrity of media history and the preservation, restoration, and conservation of our moving image culture."

Amanda Egbe

AI, Datasets and Anti-racist Archival Remix Practice

This paper explores the intersection of artificial intelligence (AI), intersectionality, and delayed cinema as an anti-racist strategy within archival and remix practice. Aby Warburg's concept of the pathos formula, which emphasizes the emotional and symbolic power of images, is used to examine the potential of AI as a film analysis tool. The increasing use of AI in the film industry and archives presents both opportunities and challenges. AI can be trained to analyze large amounts of film data, including scripts and visual elements, to develop new ways of understanding and critiquing cinema. However, the critique of AI concerning racial bias, alongside the problem of representation and underrepresentation of race within film archives, presents a significant challenge. The author argues that Yuval-Davis's concept of intersectionality, which considers the interconnectedness of different forms of oppression, can be used to understand the ways in which AI can perpetuate existing biases and stereotypes. Moving image archives can play a crucial role in providing the necessary context for understanding racial representation and can help to mitigate the effects of bias and stereotypes in AI algorithms. The author's interventions with moving image archives and image collections demonstrate how archival and remix practice, informed by Warburg's pathos formula and Yuval- Davis's intersectionality, can be used as an anti-racist strategy in the context of AI.#

Elina Lex

Archival Futures: XR Platforms as Living Archives

As cultural heritage face increasing digitization practices, the role of the digital interface plays an increasing role in how diverse aspects of heritage is being preserved, mediated, accessed, and used. As an increasing amount of cultural heritage material becomes digitized and available online, more radical approaches to preservation, historical visualization, and user engagement have also opened up with emergent XR platforms, providing dynamic entry points for publics to access, use, and navigate records in expanded, multi-layered, and interactive, polyphonic and media ecological environments. This paper thus explores the relationships between immersive media technologies and the living archive through various XR archival projects interrogating these boundaries.
How might VR, AR, XR technologies re-activate the archive from an architecture of storage into an expanded, dynamic, and generative tool for knowledge production built for diverse audiences and communities to preserve, access, use, and interpret cultural memory material? Building on the concept of the living archive (Derrida 1996; Hall 2001) through
immersive media formats aims to open possibilities for challenging pre-existing concepts of what an archive should resemble into a heterogenous space in-becoming. Inviting new kinds of embodied interaction into these spaces, how might these technologies work to overturn archives from “excavation sites into construction sites” that aims to emphasize embodied, collective, and collaborative processes? Through the polyvocal, multi-viewpoint, ecological qualities of emergent XR media, as well as new creative uses of archival artefacts enabled by freely available 3D software, this paper explores how these digital technologies can foster creative forms of archival engagement, furthering access to heritage collections, and engage users with new forms of knowledge creation in heritage spaces.

Andrew Bailey

Landscapes of Digital Art: How Platformized Gameplay is Archived at the Mackenzie Art Gallery

In 2022, Canadian digital media artists and curators Cat Bluemke and Jonathan Carroll concluded a two-year collection of videogame-focused arts programming with the MacKenzie Art Gallery. Consisting of a series of online events that almost exclusively took place across digital platforms such as YouTube, Twitch, and Discord, recordings of these happenings are all now archived on the MacKenzie website on a page called Landscapes of Digital Art (2020-22; LoDA). This paper utilizes LoDA as a multi-part case study to broadly examine the novel ways that play and performance might be archived in the pandemic era through digital platforms such as those listed above. My analysis of this mode of archiving will be conducted through the lens of recent scholarship on the “platformization” of cultural production (Nieborg & amp; Poell, 2018) and pandemic-era approaches to virtual curation (Conner, 2021; Ivanova & amp; Watson, 2021).
Additionally, the latter half of this paper focuses more specifically on an individual aspect of the LoDA programming called “Ender Gallery,” where four residencies and accompanying exhibitions were hosted on Minecraft servers that the public could visit for a limited time. Beyond the live video streams of the openings and talks associated with these exhibitions, Bluemke and Carroll have also archived additional material that each artist produced, such as Google spreadsheets and Minecraft texture packs. This paper will conclude by using Rugill and McAllister’s (2019) theorization of how videogame archives produce meaning to speculate on the ephemeral nature of born-digital material and ultimately attempt to discern what value online videogames (and how they are commonly archived across popular culture) may hold in the future for media arts curators and archivists.