Section: New Results

Open data in the arts and humanities

Participants : Luca Foppiano, Marie Puren, Charles Riondet, Laurent Romary, Dorian Seillier.

The issue of open data has become increasingly important in various scholarly domains for it impacts on the visibility of the corresponding works, the capacity to provide evidence for reported facts and results, but also let other scholars build up new research on existing data sets. This is particularly acute in the humanities where primary sources play an essential role in providing the core material of scholarly results and for which the digital turn has offered a unique perspective of building up a wealth of structure information about human traces at large.

Based upon the experience gained in the definition of the open access policy at Inria [42], [50], [43], we have pursued various activities leading to a better understanding of the technical, editorial and political factors that may improve the wide dissemination of scholarly data sets in the humanities:

• Carry out a large scale questionnaire on data re-use within the partnership of the Iperion projects, which showed the lack of a coherent data management policy across cultural heritage laboratories in Europe from the points of view of documentation, archiving, licencing and re-use [49];

• Design a concept [16], [41] to improve the general fluidity of research results in the humanities based on data quality assessment, data journals and above all the setting of of a data re-use charter between scholars and cultural research institutions in the humanities. This action, carried out in the context of the Parthenos project has started with the organisation of two high level workshops in Berlin and Paris with representatives of major cultural research institutions;

• Coordinate as leader of WP 4 (Standards) in the Parthenos project a major overview of the needs and possible deployment of standards in the humanities based of an in depth survey of possible research scenario and associated practices in the domain of standards (Deliverable 4.1 published in October 2016). This has been accompanied by specific technical developments such as the proposition of an extension to the TEI guidelines for the representation of embedded stand-off annotations [45], [51];

• Develop specific modules for mining digital sources in the humanities, in particular in the domain of named entity recognition as an improvement of the NERD software initially developed in the European Cendari project.