Indexing language data between libraries and archives

Presented by Heike Renner-Westermann, Ivana Vrdoljak and Tobias Weber at the Language Documentation and Archiving Conference, Berlin & Online, 4-6 Sept, 2024.

A deposit of language data has several purposes and meanings: it documents and represents language use for researchers and language communities alike, captures a snapshot of the global linguistic diversity, offers an empirical basis for linguistic work, and serves as a documentation of (academic) activity (e.g. Himmelmann 2006; Holton 2012; Chelliah 2021). For these reasons, depositors aim for their data to be as findable, accessible, interoperable, and reusable as possible (Wilkinson et al. 2016), while also paying attention to (indigenous) community needs (Carroll et al. 2020), with the ultimate goal of supporting discovery, citation, and reuse of their data sets. This goal links archiving to practices in publication and academic discourse, which prominently takes place in book and journal publications where data sets are cited and analysed. Linguistic descriptions and underlying data are often kept in different infrastructures – archives and repositories for data and libraries for publications. Even if a deposit has elaborate technical and descriptive metadata, a librarian can only establish a link to a publication in a catalogue if they are able to locate and assess the data. This paper presents challenges, possible solutions, and opportunities from the perspective of a specialised library service for indexing linguistic publications and linking them with their data sets. To our knowledge, no automatic approaches to linking data citations exist, links are mostly established and checked by a human. This is a work-intensive task, partially due to diverging metadata schemas and institutional requirements. At the same time, authority control can add benefits for depositors and users, which also covers controlled language nametags and classifications. These steps shall lead to increased searchability through semantic web technologies across library catalogues and repositories, as a goal of the current project.