WDCC data organization

Data in WDCC is organized in multiple layers with a hierarchical structure. These layers are named: Projects, Experiments, Dataset groups and Datasets. The relationships between them can be described with a tree structure as in figure 1.

For the top-level elements only single-parent (mono-hierarchical) relationships are supported. That means that only one parent is allowed for each Experiment and Dataset group (both dataset collections), but a parent can have multiple children-elements. Datasets and Additional Infos (which are a bit of an exception and will be described later in this section) need to have at least one parent, but it is possible that they have more than one.

The Project, Experiment and Dataset groups contain the overall description of the data, i.e. the metadata. Metadata is the description of the data itself. Detailed metadata is essential for later reuse of the data and therefore for long-term archiving. These elements (Project, Experiment and Dataset groups) can therefore be used to structure the data. The data itself is located in the datasets (which also contain description of the data).

Each Dataset is part of a hierarchical tree with an Experiment and a Project, while Dataset groups are optional elements that are only used for more complex structures. The most minimalistic data to be archived needs to contain at least one Project, one Experiment and one Dataset in the hierarchical structure.

The only hierarchical elements that can get a DataCite DOI are Experiments and Dataset groups. A DOI for Projects or Datasets is not possible.