Publishing Requirements as Linked Data

Our requirements, which are published and referenced on/by gitbook, are lying on github. Several requirements exist in different languages (en, de) sharing the same requirement-id. In order to provide the requirements as linked data, the LD-Server clones the requirements repositories into another repository, located locally on the LD-Server. This data mainly consists of .md files, but it may also contain images, documents, etc. The data then is processed by data-pipes, defined with Talend, creating an ontology. For every requirement, an instance is created. The idea is, that requirements that are available in different languages are represented in a single requirement instance. This instance holds the original URLs for both languages. The Sesame module on the LD-Server then provides an interface to access the ontologies and its instances.