What is a database?
A database is a collection of data that is organized and stored according to some purpose.
The database is organized into tables (which look like HTML tables or spreadsheets). Each table stores data about some real-world entity.
Each table is organized into rows and columns.
Early Databases in the Humanities
- [any] Library catalog [ever]
- Father Busa’s database of the works of Saint Thomas Aquinas (print volumes in the 1970s, CD-ROM in 1989)
- Historical Geodatabases (Great Britain Historical GIS, China Historical GIS, etc.)
Thinking about databases
In the last twenty years, humanities scholars have written about databases as a genre worthy of study.
- “Database as Symbolic Form” (Manovich)
Questions to keep in mind
Are databases only digital?
To what extent can we apply the term “database” to historical forms of knowledge organization?
What is new about digital databases if humanists use them to mimic index cards?
What is the relationship between databases and the other forms of knowledge we use and create?
What is the relationship between database and narrative?
Where do we find them in humanities research now?
- secondary sources (for ex. Searchworks, JSTOR)
- primary sources (EBBO, Internet Archive, NETFLIX)
- Hathitrust research environment
- Reference/bibliographic management (of research or teaching materials) [and organization of notes]
Originally designed to help researchers document references for materials cited in research outputs. Many of these applications also included notes fields.
If the index card note-taking system is formally related to the card catalog era, these applications were designed to help researchers manage newly digitized library records.
In time, applications adapted to help researchers organize information and ideas from other kinds of digital records.
- Archive or fieldwork data management
bird’s eye view of materials
information collated from many sources of different kinds
digitization of individual or sets of similar sources (and annotation)
digitization of an entire archival collection
As scholars move away from using pen and paper to take notes, many have adopted software that help organize materials that do not have records designed by institutions using common metadata standards.
- Excel, Word, text files, etc.
- Filemaker pro
- Zotero (again)
- Tropy (images + metadata + notes)
Finally, we come full circle to databases that are at the foundation of research outputs in humanities projects.
Why would you build a database?
- to generate knowledge through the process of creation/representation
- to remember/capture
- to organize/retrieve
- to relate/associate
What kinds of databases are there?
Flat databases contain only one table (or sheet). They are helpful when you need to represent one-to-one relationships.
One book has one author. One person has one address. One object exists in one museum.
When is one spreadsheet not enough?
Often, we encounter information that is not neatly organized into one-to-one relationships.
Relational databases contain multiple tables that are linked through keys (or IDs).
French Book Trade
A graph database is based on edges, nodes, and properties. Graph databases directly store the relationship between records. Relational databases link information through keys.
Examples and Discussions
- Panama papers
- Build your own graph of text annotations with Recogito
- Deb Verhoeven and Toby Burrows, “Even Sweeter: What happens when the humanities gets graphic”
- Elijah Meeks, “Visualizing Databases”