X Close

DIS Research Seminars



Archive for February, 2024

Research Talk by Rebecca Roach

By Antonios Bikakis, on 22 February 2024

Conversation Machines, Missing Secretaries, Bad Readers

The talk was delivered on 21 February 2024 by Dr. Rebecca Roach, Associate Professor of Contemporary Literature and Principle Investigator (Digital) of The Stuart Hall Archive Project at the University of Birmingham., as part of the DIS research seminars series.

‘Giant electronic brains’: it was an early model for understanding computers, one that has been enormously generative, spurring advances in cybernetics and Artificial Intelligence across the decades. In this talk I want to tell a different story. There is another metaphor about computers that is just as prevalent in our culture, but much less commented upon: computers as talking machines. Conceived as interactive, as ‘conversational’, computing technologies start to look very different and the relations that they posit across disciplines (the inherent value of literary studies in particular) very different too. Taking as my case study the ‘ur-bot’ ELIZA (1965), I will pull out some of the methodological and conceptual implications of conceiving of computers as conversational. Call it a back-history of ChatGPT if you will.

Presentation of the Sloane Lab project

By Antonios Bikakis, on 9 February 2024

Sloane Lab: Looking back to build future shared collections

The talk was delivered on 7 February 2024 by Dr. Andreas Vlachidis, Dr. Marco Humbel and Dr. Alda Teracciano, members of the team working on the Sloane Lab project, as part of the DIS research seminars series.

Funded by the ‘Towards a National Collection’ (TaNC) Program of the Arts and Humanities Research Council (AHRC), UK, the ‘Sloane Lab: Looking back to build future shared collections’ devise automated and augmented ways that mend the broken links between the past and present of the UK’s founding collection in the catalogues of the British Museum, the Natural History Museum and the British Library. The project develops new technologies, including the use of Artificial Intelligence (AI), to open up the contents of museums and collections in ways that are more intuitive and relevant to the way the public and academics want to discover and use them. The task of integrating these disparate records and facilitating interoperable access to them poses significant challenges. Sloane’s historical catalogues are especially difficult to represent digitally, because the descriptions of the objects are often incomplete or inaccurate. Therefore, it is crucial to adopt a critical digital heritage approach since the semantic representation of Sloane’s historical catalogues may produce datasets that contain uncertainty and biases. The project addresses such biases and absences, allowing for certain worldviews and answerings to this challenge of ‘multivocality’ by adopting a data modelling approach that focuses on the record than on the object, viewing records as different perspectives over the object. Moreover, the project employs a participatory research design methodology that unpacks the latent challenges in international collection data infrastructure development. The participatory design delves into research questions that address experiences of heritage organisations of participating in national and international digital infrastructure projects, explore factors that enable and impede heritage organisations in unifying siloed collections and investigate how these factors differ between institutions and countries.