Diverse Data! Diverse Schemata?

One of the key value propositions for knowledge graphs and semantic web technologies is fostering semantic interoperability, i.e., integrating data across different themes and domains. But why do we aim at interoperability in the first place? A common answer to this question is that each individual data source only contains partial information about some phenomenon of interest. Consequently, combining multiple diverse datasets provides a more holistic perspective and enables us to answer more complex questions, e.g., those that span between the physical sciences and the social sciences. Interestingly, while these arguments are well established and go by different names, e.g., variety in the realm of big data, we seem less clear about whether the same arguments apply on the level of schemata. Put differently, we want diverse data, but do we also want diverse schemata or a single one to rule them all?