Models
At the heart of Cadmus there are its data models, which are built by composition; and this is true also of its editor.
You are free to create any model with its editing UI, and add them into an existing or new editor; but often you can just reuse models and editors which have been designed to be shared across projects. In fact, it may well happen that a real-world project is completely built by assembling these shared resources, without having to add anything new; while other projects can add their own specific models.
Anyway, as Cadmus is geared towards reuse, it is often the case that project-specific models prove to be general enough to be promoted to project-independent, shared models, primarily designed for cross-project reuse. This approach fosters reuse and even indirect collaboration among scholars, while also providing a foundation for modeling specific knowledge domains by generalizing from specific real-world projects.
Additionally, models are complemented by the usage of taxonomies, named thesauri. These can be simple, “flat” lists, or complex hierarchies with classes and sub-classes. In both cases, they always provide a machine-targeted identifier string and a human-readable label for it.
By convention, usually identifiers are English, while labels are freely defined in the language of choice.
Any UI component can leverage these taxonomies to allow users pick values from a preset (yet editable) list, rather than freely typing them. Next to this “static” data lookup Cadmus also provides a dynamic lookup to link items and parts entered in the editor.