FAIR data
- They are discoverable by anyone.
- They use standard formats with appropriate metadata and identifiers, making them reusable and citable.
- Access to the data itself may still be conditional (e.g., signing an agreement or complying with contractual restrictions).
- Their use is limited by licenses that define rules for handling the data. Making data available according to FAIR principles does not automatically mean giving everyone access.
- The goal is to follow the principle: “as open as possible, as closed as necessary,” i.e., to make data “as open as possible, as restricted as necessary.”
FAIR is an acronym representing the initial letters of these attributes:
Findable
The first step toward (re)using data is to make it findable for interested parties.
Data should be:
- Annotated with metadata
- Assigned globally unique and persistent identifiers
- Indexed in appropriate search systems
Accessible
Once a user has discovered the data, they must know under what conditions they can access it.
Access should meet the following criteria:
- Authorized/authenticated
- Restricted, e.g., based on certain conditions (such as signing a data access agreement)
- Metadata should be accessible via a persistent unique identifier using standard open protocols
- Metadata should remain available even if the actual data are no longer accessible
Interoperable
To make data (re)usable, they must be made available in a way that allows integration and processing by other applications, systems, and users.
(Meta) data should meet the following requirements:
- Use standard, widely recognized representations and data formats in the given domain
- Use controlled vocabularies and ontologies that follow FAIR principles
- Be linked to other (meta)data through formalized references
Reusable
Reusability is a core goal of the FAIR principles – to enable data, once created, to be reused by their authors and other interested parties.
Essential conditions for reusability include:
- Availability of well-described data
- Accompanied by appropriate metadata dissemination under a clear license
- Clear documentation of the provenance of the (meta)data
- Compliance with state-of-the-art standards and practices in the relevant research domain (community of researchers in the field)
How FAIR is your research data?
Checklist for evaluating the FAIR attributes of your data