Bibliographic data science. A model for collaboration between researchers and the library

Bibliographic data science is an interdisciplinary method that applies data science technologies on bibliographic data in order to reveal patterns that are relevant to Humanities or social science domains such as history, literature, translation studies, sociology of culture, library and information science. Big data might shed light to phenomena that were previously unknown or supported only by anecdotal evidence. The bibliographic data science’s research questions include examples like how did the format and language of books change over time in different regions? what are the patterns of translations from a given language, how have they changed, and which languages were super-central, central, and peripheral in a given era? how many historical works or publications could have been destroyed without any trace? In order to answer these questions the bibliographic data should be not only accessible but also suitable for computational use. As cultural heritage organisations, and in the first place libraries are who are the main collectors and creators of bibliographic data, it is their role to ensure accessibility. This new role involves new knowledge and skills, and organisational changes. Libraries should also prepare for the “data roundtrip”, to handle improvements and enhancements made by the researchers, and build them into the catalogue.