Requirements for, and extensible model of, transparency: A goal of DeKaloG is to be able to handle the transparency of statements in a KG, but also of a knowledge graph in its whole, including the specific case of the transparency of an index-KG. A first step consists in drawing a whole picture of the needs. To do this, we need to consider not only the results and tools in the linked data, semantic web, and workflows domains but also achievements in other scientific domains (e.g. physics, medicine, biology…). A second step will focus on modeling and querying transparency. We are convinced that different levels of transparency will have to be defined, from providing usual contextual data such as authorship, sources, date of production, etc., up to enabling reproducibility (e.g. fact verification, KG reconstruction). As far as possible, we will seek for domain-independent models with the possibility of extension for domain specificities.