Due to increasingly specialised, diverse and also new competencies and competency profiles it becomes progressively more difficult to interpret educational achievements and to match requirement profiles with the competencies of individual persons or groups. The computational support with regards to keeping the information up to date, communication, search and analysis is limited if the competencies are described in natural language only. Thus, it seems advantageous to model competencies in a formal and machine-readable specification language. The following article suggests the notion of a generic formal syntax for learning outcomes. We outline how this would allow expressing intricate learning outcomes in a machine-readable ontology and their further processing with the Linked Data- and Semantic Web-approaches.