Adopting Transkribus in the National Library of Scotland: Understanding how Handwritten Text Recognition Will Change Management and Use of Digitised M

点击翻译按钮获取中文摘要

Recently, artificial intelligence (AI) has become a core tool in making digital archival material more accessible. Nevertheless, in these studies users and their behaviours are often neglected. Working in close collaboration with the National Library of Scotland (NLS), this studentship will deduce how one AI platform, the text recognition software Transkribus, can be best deployed to benefit user communities. An understanding of how implementing such systems might disrupt, extend and change traditional workflows at leading libraries and archives will also be reached. HTR technology can be seen as part of the wider trend of digitisation in archives - a trend which is also behind the newfound focus on 'digital humanities'. Archives are currently looking to employ such digital software precisely, undergoing a reimagining of how they act as custodians of historical material. Nowadays, internet access has become so ubiquitous that discussions of a 'post-custodial world', where archives are mere 'access hubs' or 'virtual archives without walls', has emerged. This change is only seen to increase and, as such, the emphasis on systems like Transkribus will only grow. Such questions will be asked as; how can libraries embed themselves within digitisation workflows effectively? How might HTR usage disrupt or enhance existing archival practices? What institutional, ethical and infrastructural issues need to be tackled to take up HTR more widely? Taking place on the ground with the NLS, especially its Digital Archives and Manuscript Divisions, as well as with the developers of Transkribus at the University of Innsbruck, key access to relevant areas of interest will be enabled. Such an opportunity will allow an analysis of the software in context and the application of ethnographic approaches in understanding how HTR relates to tradition palaeographic practice. Occurring in year two of the post: a core component of the methodology of this project will be running user experience testing with members of the public at the NLS, extending their existing outreach activities, using material from the John Murray Archive, evaluating barriers and strengthens in using Transkribus. Other areas of user analysis will be made available, including pop-up exhibitions tied to the NLS's annual events, as well as showcases through the Scottish Graduate School for Arts and Humanities (SGSAH). Through the supervision of Prof Terras, opportunities to speak about research findings at the annual Transkribus conference will be taken up also. Much research into AI is currently amorphous future-casting without this grassroot analysis of users. By reinserting users back into the analysis of such software, a more critical view of how to deploy digital tools like Transkribus will be formed through this research project.