According to the FAIR Guiding Principles for Scientific Data Management, researchers are encouraged to make their data Findable, Accessible, Interoperable, and Reusable.
The FAIR Principles are not rigid rules or a technical standard, but rather a framework supported by the European Commission to help researchers manage, share, and preserve data so that it can be effectively used by both humans and machines. Applying these principles ensures that data are easier to locate, understand, exchange, and reuse, while also supporting long-term preservation.
Importantly, FAIR does not require data to be openly available to everyone.
- Data can be FAIR but not open, for instance, when access is restricted for privacy, ethical, or legal reasons. However, metadata and conditions for reuse need to be clearly described.
- Conversely, open data may not be FAIR if they lack sufficient documentation, metadata, or licensing information to enable proper understanding and reuse.
The FAIR Principles
Findable
Data should be easy to discover and identify by people and machines. Each dataset should have a persistent, unique identifier (such as a DOI) and be described by metadata that can be located through disciplinary, institutional, or global portals. Even if the data themselves are restricted, the metadata should remain available and searchable.
Accessible
Data and metadata should be retrievable using standardized protocols. A clear and transparent license or access policy must describe the terms of use. FAIR does not mean that all data must be open. Sensitive or confidential data can remain closed, provided that access conditions are documented and discoverable.
Interoperable
Data should be stored and described in formats and vocabularies that are widely recognized within the research community. Metadata should indicate relationships between datasets and use consistent identifiers so that data can be integrated, exchanged, and reused between disciplines, institutions, and countries.
Reusable
To enable long-term reuse, data should include comprehensive documentation, provenance information, and clear licensing. They should retain their full informational value, not just a subset used for publication. Using community standards and detailed metadata ensures that others can interpret and reuse the data appropriately.
FAIR in practice
Researchers should consider FAIR principles from the beginning of their project, ensuring that:
- Datasets are assigned persistent identifiers (e.g., DOIs)
- Metadata is openly available even if data access is limited
- Open, interoperable formats and recognized vocabularies are used
- Licenses clearly state who can use the data and under what conditions
- Provenance and contextual information are recorded to support future reuse
Help and further information
Checklist: How FAIR are your data?
DATICE – The Icelandic Research Data Service
GOFAIR – The Global Open FAIR guides for people and organisations on solutions for making data Findable, Accessible, Interoperable, Reusable for people and machines
OpenAIRE – Open Access Infrastructure for Research in Europe
Wilkinson, Dumontier, Aalbersberg et al. 2016. “The FAIR Guiding Principles for scientific data management and stewardship”
Guidelines on FAIR Data Management in Horizon 2020.