Presentation Open Access
The engineering sciences are characterized by a high degree of heterogeneity and
close links with other scientific disciplines. So far, no metadata standards specific for engineering have achieved a sufficiently wide community acceptance. However, there are current developments of engineering-focused metadata standards among the participants of NFDI4Ing, e.g. EngMeta or D-SI, that can be used as building blocks. In addition, there are several controlled vocabularies relevant to engineering (e.g. QUDT, VIM3 international vocabulary of metrology, Allotrope Taxonomies) as well as commercial terminologies (e.g. Thesaurus Technology and Management, DKF thesaurus for automotive technology, or DIN, ISO and other norms), that need to be considered.
NFDI4Ing will implement a hierarchical metadata model, where specific metadata objects (terminology-classes or schemata) are derived as “children” of more general ones. In this model, like in object-oriented programming, each child inherits all elements contained in its parent, and extends these with new elements. The design will be a multi-level hierarchy, in which the scope of the derived children gradually becomes more narrow and precise. The resulting hierarchical tree of related objects maximises their interoperability, since the objects are always downwards-compatible up to the level of their most specific common parent. To further increase interoperability, NFDI4Ing will make use of existing terminology and standards and reuse their terms as building blocks, assuring compatibility to widely used existing standards like, e.g., the DataCite MDS.
Building on this approach, we will also implement a modular metadata design in which different realms of metadata (e.g. description of method, object of research and tool) are separated, resulting in metadata consisting of multiple smaller datasets governed by different standards. This drastically increases the standards’ applicability and maximises their reusability across disciplines.
NFDI4Ing will provide tools and technical infrastructure to facilitate and guide development and usage of metadata build upon the concepts outlined above. This includes, among others, a generator for application-specific metadata schemata, a repository indexing and hosting the developed schemata and allowing their reuse and adaptation, and a terminology service.