FAIR and Open Data

While Open Data asks for an unrestricted opening of the data, FAIR principles pursue a more pragmatic approach to data and allow leveled access criteria that meet scientific requirements. Data can be FAIR, even if they are not 'Open'.


The FAIR principles aim at a scientific registering of the data so that they can be addressed, quoted and verified as persistently as possible on the fast-moving internet. The SNSF focuses its research funding on FAIR data. The catchy abbreviation FAIR stands for Findable, Accessible, Interoperable and Re-usable.


To make data findable, it must have a unique identifier, a PID (persistent identifier). In addition, the metadata should be machine-readable and also easily understood by humans.


Data should be freely accessible, i.e. online and without a paywall. In practice, however, any data that is subject to personal rights, copyrights or special contracts is excluded. In any case, it should be defined which data can be shared in which way, and which data must be protected.


(Meta)Data should be compatible with different computer systems. Open formats, and if possible no protected formats, should be used. In addition, the metadata should reflect the already established standards of the respective research discipline.


The metadata should clearly indicate the conditions under which the data may be used.

Open Data

Making data available without restriction (public domain) is the universal ideal of Open Data. Data must be machine-readable and have open file formats. Applied licenses must not exclude individual groups from using the data. Nevertheless, Open Data is subject to data protection and must therefore not contain any sensitive data. The practical implementation of Open Data is often difficult.