As noted on the “design principles” page, Hydra supports two forms of object in a repository
- a compound object, where an object may contain one or more “items” of content (in the special case of a single item of content we often refer to it as a “simple” object rather than compound), and
- atomistic (complex) objects where there is a parent object linked to content held in one or more child objects
We offer notional guidelines as to which approach might be used in particular situations and a Hydra repository can support both approaches simultaneously. Whichever form you use, in a mixed content repository it is likely that you will want to separate your content into different “types” (e.g., ETD’s, archival images, articles, …). Each item of content is likely to comprise a number of elements (in Fedora, contained in datastreams): there will be the binary content itself (one or more files), a statement of rights, probably some descriptive metadata, and so on. Hydra makes available examples of models that can be used for various forms of content which can be used or adapted; of course, the community is continually developing new ones which we encourage them to share.
Modelling your content carefully is an essential first stage to creating a Hydra repository. Much more information is given on the wiki.