Page tree
Skip to end of metadata
Go to start of metadata
TitleRepository as a continuous graph of related objects
Primary Actorrepository managers
Scoperesearch community at the Smithsonian, administrators

The researchers repository at the Smithsonian is designed to be a continuous graph, building out from a root node that describes the repository as a whole. Every object that is added to the repository should have at least one relationship path that leads back to that root node. As we give accounts to researchers, their root node in their account is a
"person" object that describes them as a researcher at the Smithsonian. They build out their research data as project objects that are child objects of their person object. Above their account, each researcher's person object is a child of unit object that describes a research center or museum at SI, which is a child of the root object.

Many more graphs of relationships can be overlaid, but the one described above is the basis.


  1. This is describing a specific type of a graph, namely a tree, rather than the more general case that sometimes gets thrown around when we talk about Fedora as a graph.


  2. Actually, it would be more accurate to say that a specific tree will be created and maintain within the graph. It is completely possible for the graph to grow in non-tree ways, but that specific tree is a useful construct that we will maintain within the graph.

  3. This is ready for acceptance testing as of 4.0-Alpha-4.