KSE team completes Training on FAIR Principles in Data Management by the University of Tartu

Every research project relies on data: questionnaires, interviews, transcripts, spreadsheets, and databases. How a research team stores, names, and protects these materials determines whether colleagues can verify the results and whether other researchers can reuse the data in their own work. During wartime, the stakes are even higher: a data leak involving people or locations may put their safety at risk.

To learn how to manage research data so that they remain usable for years while protecting sensitive information, the KSE team took part in the course "Training on FAIR Principles in Data Management", organised by colleagues from the University of Tartu and ELIXIR Estonia node.

FAIR is a set of four principles for research data: Findable (they can be found by both humans and machines), Accessible (they can be accessed under clearly defined conditions), Interoperable (they can be integrated with other datasets and systems), and Reusable (they can be reused). Importantly, "Accessible" in FAIR does not mean that data must be openly available to everyone. Data may remain closed, provided that the conditions for access — whether open, restricted, or available upon request — are clearly described in the metadata so that they can be interpreted by both humans and machines.

The course consisted of four sessions held on 26 and 28 May and 2 and 4 June 2026. It was delivered by Diana Pilvar, Data Manager at the University of Tartu and Training Coordinator for ELIXIR Estonia. The course was developed together with Marko Arrak, Principal Analyst at the Institute of Computer Science, University of Tartu.

Participants explored:

• how to organise files and folders and document research data so that they can be understood and used beyond  their original creator
• how to protect the personal data of research participants, including GDPR requirements and the anonymisation of interviews and questionnaires
• how to preserve data over the long term by choosing appropriate file formats and backup strategies — for example, following the 3-2-1 backup rule: three copies, stored on two different types of media, with one copy kept in a separate location
• how to make research data available to other teams so that the data can be verified, compared, and reused in future research

An important part of the course focused on data protection during wartime. Participants discussed the risks associated with geolocation and open-source intelligence (OSINT), restrictions on publishing sensitive information, and the fact that Ukraine is developing its own GDPR-equivalent legislation. As a result, research teams should begin preparing for the new requirements now.

We thank Diana Pilvar, Marko Arrak, and colleagues from the University of Tartu for sharing their knowledge and practical tools.

* In FAIR, "Accessible" does not mean that data must be openly available to everyone. Data may remain closed, provided that the conditions for access — whether open, restricted, or available upon request — are clearly described in the metadata and can be interpreted by both humans and machines.