0. Metadata

Metadata is data about data. It comes in two flavors:

Structural metadata: how the data is stored; e.g. file format, column data types, data model.
Descriptive metadata: data about specific data entries/content. (This is usually what Digital Humanists talk about)

A metadata schema is a (usually formal) description/vocabulary for metadata. It describes what properties and types can exist in the metadata. Examples include:

Dublin Core– Popular for document repositories;
Metadata Object Description Schema– From the US Library of Congress;
Friend of a Friend (FOAF)– Describes people and their relationships.

Resource Description Framework (RDF) is a data model for metadata. It describes how to structure metadata. The core concept of RDF is the triple: a three-part relation comprised of a subject, a predicate, and an object.

These triples can be combined into an RDF document, which takes the form of a graph.

Graphs and Networks

For the purpose of this course, we will distinguish between graphs and networks.

A graph is a data structure comprised of nodes (vertices, points) and edges (links). Sometimes you will see this formalized (usually unnecessarily) as: G = { N, E }.

A network is a system of interconnected things. This includes telecommunications networks, social networks, anthills, websites, and lots of other things. We often use graphs to represent/describe networks. You could also say that a graph is a network, since it is comprised of nodes and edges.

One way to think about the difference between a graph and a network is in epistemic terms: a network is usually a phenomenon or system that we want to study, whereas a graph is a model or representation of that phenomenon.

We can serialize RDF (write it down) in a variety of formats.

When you export a collection from Zotero as "Zotero RDF", a file with an .rdf extension is generated. This is an RDF/XML document.