Collection replication for enhanced re-usability and preservation planning

Presented by Paul Trilsbeek, Kavon Hooshiar, Antonina Werthmann and Andreas Witt at the Language Documentation and Archiving Conference, Berlin & Online, 4-6 Sept, 2024 The Language Archive (TLA) is one of the first Language Archives created after Himmelmann (1998), housing the DOBES collections of endangered language data. The development of the metadata schema and access restrictions for TLA represent innovations that have now been adopted and streamlined by newer DELAMAN archives. In this paper, we present an ongoing collaboration between TLA and the Leibniz Institute for the German Language (IDS) to improve the security, accessibility, and discoverability of DOBES collections. TLA will replicate collections to IDS to create a novel layer of institutional redundancy to the data, which goes beyond redundancy of location and media. We review details and challenges of planning this infrastructure, including: maintaining the integrity of the data, metadata, and access restrictions of the original collections; renegotiating relationships with depositors; legal constraints; and automating digital infrastructure across institutions. We aim to improve the accessibility and discoverability of the data by: a) Asking depositors to review access restrictions, and to provide: secondary contacts; criteria for who should have access; and criteria for who should be allowed to moderate access requests in their absence. a) Enhancing and standardising documentation for each collection, encompassing annotation formats and conventions. c) Converting structured text transcriptions into the TEI-based ISO standard format (2462:2016), enabling easier automatic migration to diverse formats and usability across different disciplines. d) Integrating the data into Germany’s National Research Data Infrastructure (NFDI) via Text+. In this context, the project focuses on questions like: What happens to restricted data once depositors are no longer available to moderate requests? How can we insulate the data from unforeseen institutional changes? This project seeks to plan for the future of these data by improving their accessibility, discoverability, and security, thereby benefiting scholars across various disciplines and contributing to the long-term preservation of global linguistic diversity.