According to the Guidelines for Safeguaring Good Research Practice, project processes and results must be documented in a comprehensible manner. Like many other German universities, LMU has established a set of guidelines for this purpose – the “Ordnung der Ludwig-Maximilians-Universität München zur Sicherung guter wissenschaftlicher Praxis” (in German only). LMU members are asked to follow these guidelines when writing and publishing articles or collecting research data.

The criteria created in the planning phase are used to decide which of the raw data collected are considered relevant. These are then archived and the remaining data deleted. To support data exchange and data integration it is important to convert the data into a sustainable data format suitable for the planned research work, and to select an appropriate collaboration platform. In collaborative projects, the individual specifications and procedures of a collaborative research environment must be taken into account. If, for example, research software is also used or independently (further) developed, it is advisable to describe and include these components in the data management process.

Transparency regarding the data considered relevant is ensured by providing the most detailed documentation possible. It is useful to enrich the data with a minimum quantity of metadata*, including documentation of the place and date of creation, and documentation of the information about the device used to generate the data. Persistent identifiers** should also be assigned to enable unique identification and allocation of the data. The use of subject-specific standards should also factored into this process. The choice of secure data storage and the use of back-up mechanisms, possibly by involving the use of local infrastructures, are also essential. Another aspect to be considered in the preparation and documentation phase is data visualisation and data back-up.

* Metadata are data about data. They include structured information about research processes and results. Examples of metadata are: bibliographic metadata (e.g. title, author), content information (e.g. keywords) and administrative metadata (e.g. data type, licence, access rights)

** A Persistent Identifier is a unique reference to a resource through the assignment of a code. Widely used examples of persistent identifiers are Digital Object Identifiers (DOI) or URN.