The MASi repository service — Comprehensive, metadata-driven and multi-community research data management

Nowadays, the daily work of many research communities is characterized by an increasing amount and complexity of data. This makes it increasingly difficult to manage, access and utilize the data to ultimately gain scientific insights based on it. At the same time, domain scientists want to focus on their science instead of IT. The solution is research data management to store data in a structured way enabling easy discovery for future reference and usage. An integral part is the use of metadata. With it, data becomes accessible by its content and context instead of its name and location only. The use of metadata shall be as automatic and seamless as possible in order to foster a high usability. Here, we present the architecture and developments of the Metadata Management for Applied Sciences (MASi) project that is currently building a comprehensive research data management service. MASi extends the existing KIT Data Manager framework by a generic metadata programming interface and a generic graphical web interface. Furthermore, MASi is OAI compliant and supports the OAI-PMH protocol while providing support for provenance information using ProvONE, a well-established and accepted provenance model. To illustrate the practical applicability of the MASi service, we present the adoption of initial use cases within geography, chemistry and digital humanities. The MASi research data management service is currently being prepared to go into production to satisfy the complex and varying requirements in an efficient, useable and sustainable way.