From Cataloging to Transcoding: The Role of LLMs in the Remediation of Electronic Literature
Michael Hurtado, a mathematician graduated from the National University of Engineering, is also a technologist, new media artist, and poet. He earned a Master's degree in Applied Mathematics at the Pontifical Catholic University of Peru and a diploma in Digital Fabrication at the MIT Center for Bits and Atoms. He works as a professor and researcher in the Department of Architecture at the Peruvian University of Applied Sciences and is co-director of Masmédulab, a laboratory of poetry and new media. Through Masmédulab, he led significant projects such as the "Genealogy of Peruvian Electronic Poetry" and the "Rescue and Remediation of Latin American Electronic Poetry" project.
He is a fellow of the FabLearn program at Columbia University, a member of the Electronic Literature Organization and the FabLab network, co-founder of LimaMakers, and editor of volume 2 of the Lit(e)lat anthology. In 2014, he was awarded the VIDA16 prize by the Telefónica Foundation for artificial life projects. In 2018, he gave a TED talk on the relationship between art, science, and technology. His electronic textile works were exhibited at the Rome Fashion Week in 2019, and in 2020, he was honored with the first edition of the Hub Musical Chile award for immersive art projects. His poems have been published in the anthology "Nós da Poesia Vol 08: verbal and visual poems", in the "I Mostra Virtual de Poesia Visual", in the "Revista Bufo" of experimental poetry, in the "TAPER" magazine of electronic literature, and on the illitera website. His electronic poems are part of collections such as the "Electronic Literature Collection vol 4" and the "Cartography of Latin American Digital Literature" by Diego Portales University in Chile.
Abstract
In the post-digital context, the field of electronic literature in Latin America has garnered significant academic interest focused on the cataloging and promotion of works by authors from the region. Key initiatives such as the Lit(e)lat Anthology, the Digital Cartography of Electronic Literature by Universidad Diego Portales (UDP), and the Brazilian Digital Literature Atlas have set a precedent in documenting this corpus between 2020 and 2021. However, there is a tendency to overlook the original experiential dimension and the specific technological framework in which these works were conceived. Therefore, it is imperative that the processes of restoration and recovery of electronic literature go beyond the mere preservation of visual and kinetic aspects, equally addressing the restoration of works through source code, which stands as a fundamental component of textuality, challenging technological obsolescence.
This study proposes the use of Large Language Models (LLMs) in the remediation task, identifying them as crucial instruments for the transcoding of digital works to contemporary formats and platforms. An innovative perspective on the remediation of electronic literature is presented, emphasizing the critical role of LLMs in the adaptation and preservation of texts for their revaluation within the current Latin American cultural panorama. In this sense, not only is the relevance of electronic literature reaffirmed, but the need for transdisciplinary strategies for its preservation and study within the framework of contemporary digital culture is also highlighted.
From Cataloging to Transcoding: The Role of LLMs in the Remediation of Electronic Literature
Algorithms & Imaginaries
In the post-digital context, the field of electronic literature in Latin America has garnered significant academic interest focused on the cataloging and promotion of works by authors from the region. Key initiatives such as the Lit(e)lat Anthology, the Digital Cartography of Electronic Literature by Universidad Diego Portales (UDP), and the Brazilian Digital Literature Atlas have set a precedent in documenting this corpus between 2020 and 2021. However, there is a tendency to overlook the original experiential dimension and the specific technological framework in which these works were conceived. Therefore, it is imperative that the processes of restoration and recovery of electronic literature go beyond the mere preservation of visual and kinetic aspects, equally addressing the restoration of works through source code, which stands as a fundamental component of textuality, challenging technological obsolescence.
This study proposes the use of Large Language Models (LLMs) in the remediation task, identifying them as crucial instruments for the transcoding of digital works to contemporary formats and platforms. An innovative perspective on the remediation of electronic literature is presented, emphasizing the critical role of LLMs in the adaptation and preservation of texts for their revaluation within the current Latin American cultural panorama. In this sense, not only is the relevance of electronic literature reaffirmed, but the need for transdisciplinary strategies for its preservation and study within the framework of contemporary digital culture is also highlighted.