O This work brings the search results that aimed at the development of a visual language representing parameters and metadata of learning objects in a synthetic and imagery way to facilitate their assessment and selection process by educators. The search was based on principles of NURBS theory and Manovich databank logic (2001, 2010) in addition to the concepts of learning object, metadata and recommender systems (Wiley, 2000; CAZELLA et al., 2009, 2010). The research has shown fruitful the possibility of transforming subjective culture into data through the proposition of categorisations that have taken into account qualitative variables in OAS analysis from learning theories applied to instructional design and standard features used in its organisation and cataloguing in digital repositories.