SangyaPundir [CC BY-SA 4.0], from Wikimedia Commons
The four principles (as above) provide guidance for data producers and publishers to enhance the reusability of research data. These principles were developed by the FORCE11 community and focus on the data, rather than the responsibilities and conduct of the data producers.
The FAIR Guiding Principles
Data should be Findable. To be Findable:
F1. (meta)data are assigned a globally unique and persistent identifier
F2. data are described with rich metadata (defined by R1 below)
F3. metadata clearly and explicitly include the identifier of the data it describes
F4. (meta)data are registered or indexed in a searchable resource
Data should be Accessible. To be Accessible:
A1. (meta)data are retrievable by their identifier using a standardised communications protocol
A1.1 the protocol is open, free, and universally implementable
A1.2 the protocol allows for an authentication and authorisation procedure, where necessary
A2. metadata are accessible, even when the data are no longer available
Data should be Interoperable. To be Interoperable:
I1. (meta)data use a formal, accessible, shared, and broadly applicable language for knowledge representation
I2. (meta)data use vocabularies that follow FAIR principles
I3. (meta)data include qualified references to other (meta)data
Data should be Reusable. To be Reusable:
R1. meta(data) are richly described with a plurality of accurate and relevant attributes
R1.1. (meta)data are released with a clear and accessible data usage license
R1.2. (meta)data are associated with detailed provenance
R1.3. (meta)data meet domain-relevant community standards
To be findable, your data or the details of your dataset (if you are restricting sharing in any way) should have a DOI (Digtal Object Identifier) assigned to it. Make sure you describe your data clearly using discipline specific words and rich keywords and descriptions where you can so that your record can be found via text and data mining. Whenever and wherever you can link the data to your output(s).
To be accessible, make sure your DOI works, that you have a data statement in any related publications, that there are no constraints to accessing your record, use a trusted repository and the record of your data is maintained.
To be interoperable, do not share data that can only be read by software that requires an expensive licence, make sure you can reformat it in an accessible language that can be transformed into other easily accessible formats. Use a trusted repository and ensure that indexing used by the repository follows FAIR data principles.
To be reusable, you should give enough details either with the data itself or as an additional readme file so that a future user can understand what you did clearly using standard discipline specific descriptions and keywords.
More information on FAIR data principles;