Collaboration is a crucial part of joint research. Especially in projects, where multiple academic domains and disciplines are involved, this is challenging. Different research foci and individual perspectives are often mismatching. Heterogeneous data, diverging working habits, and differing standards are common. Research itself is agile, dynamic, and evolving. Changes and revisions of data structures, requirements, and actual data values arise continuously. And so on…

Adaptations due to new circumstances and evolution driven changes are normal for such projects and need to be considered explicitly by its data management strategy. The same applies for heterogeneities in general. No project wide working standards, models, or structures can be expected to be continuous throughout the complete run time of a project. Thus any static approach for managing data in research projects will fail to provide sufficient support. In effect also the collaboration cannot be supported well.

We propose a new approach basing on the separation of data storage and data usage. The actual research is performed in local and individual working environments. These can be set up to meet the individual research requirements of the project members best. Therefore the storage does not need to support research related computation activities or reflecting the working models of project members. Instead it uses a universal model to store data decomposed into values and structures. Thus the individual perspectives can be modelled as compositions of these decomposed elements and allow the creation of customisable interfaces. In effect we get a flexible and modern approach for handling data in research projects and for supporting interdisciplinary research.