OLAC Usage Data

I was recently reading (Citation: Wenige et al., 2018) Wenige, L. & Ruhland, J. (2018). Retrieval by recommendation: using LOD technologies to improve digital library search. International Journal on Digital Libraries, 19(2-3). 253–269. https://doi.org/10.1007/s00799-017-0224-8 on recommender systems within digital libraries with an mind on OLAC and how to help OLAC website visitors to find relevant resources.

The current OLAC XSLT based interface assumes that the end-user person starts with the geographical entity as the known entity and then drills down to find a language. Then finally they can discover the known resources about the language. It uses geography as a proxy for language collection. The faceted search interface at search.language-archives.org makes a different assumption. It allows the user to start their search via unconstrained text input and then filter by something like 18 different attributes potentially found within the record. The XSLT method uses information not contained directly within the record whereas the faceted search method is limited to record context.

In (Citation: Bird et al., 2021) Bird, S. & Simons, G. (2021). Towards an Agenda for Open Language Archiving. University of North Texas. https://doi.org/10.12794/langarc1851171 (Citation: Bird et al., 2022) Bird, S. & Simons, G. (2022). The Open Language Archives Community: A 20-year update. The Electronic Library, 40(5). 507–524. https://doi.org/10.1108/EL-08-2022-0192 it is mentioned that several thousand click-throughs occur on the OLAC website per month. I would like to get a hold of this data:

1. to verify the claim in the literature.
2. to determine in which types of resources people are most interested, but also the breadth of interest.
3. to determine which pathways through the website people are following. Where are they starting, where are they stopping. How much back and forth-do-they do? What are the popular facets?
4. to determine at which data providers people are ending up.

Having this data would be a first step in an investigation related to relevance negotiation behavior of end-users as they perform search tasks. This sort of knowledge should be a first order task as we think about user-interface and user-interaction design for the future of OLAC and language resource discovery.

References