Complex objects
The digital world offers far more options for linking components of research and publications. Such coupled components are referred to as a “complex object”. One example is an enhanced publication, which consists of the publication itself supplemented by text, research data, software, and other objects.
In a complex object, it is important that the relationships between the components should be properly indicated and that the complex object should be properly described. The description can be stored in the metadata or in an external metastructure, one that is set up in MPEG21 DIDL format.
An international consortium is working on a new future-proof metastructure for describing such complex objects that takes the semantic web, atomic relationships and domain ontologies into account. For example: a dataset of geologists and a dataset of astronomers will be represented in different ways. Representations of interrelationships and structures are closely related to metadata.
The issues involved are not only technical but also practical in nature: what is a complex object anyway, what are its components, and which components must or must not have their own Persistent Identifiers in order to replicate the entire contents? With the participating institutions, SURFshare is attempting to find answers to these questions.