Controlled vocabularies allow an accurate and controlled approach in describing physical and digital assets (e.g., data). One of such controlled vocabulary is NICEST-2 Variables. This controlled vocabulary is a result of Metadata 4 Machine (M4M) Workshop 19 (M4M.19 funded by EOSC-Nordic ) provided to the NICEST-2 project. sheet2rdf and OntoStack, developed by FAIR Data Collective, are used to build and serve NICEST-2 Subjects, while PURL will be used to persist identifiers for the vocabulary terms and properties: http://purl.org/m4m19/variables/