Considering the application and domain point of views, we consider that the use of the LOD provides many benefits. First, it enables us to address a large audience with the description, in different languages, of drugs sold in different countries. Characteristics of LOD also support the discovery of new data sets and valuable information by following dereferenceable URIs. For instance, it enables us to integrate data in an efficient and fluid way that could not be supported with non-RDF formats. In terms of other usage of semantic technologies, reasoning over our knowledge repository is quite useful but is mainly supporting back end operations, e.g., data quality enhancement. Finally, we appreciate the ability to query in a federation kind of way over several LOD data sets, i.e., executing a SPARQL query over several RDF graphs.