The Maury Correspondence Database
One of the things I most wanted to learn at the beginning of the semester was to build a relational database from scratch. I know how to use some of the programs out there (FileMaker Pro, mostly), but felt that I needed to know how to create one from scratch to be a real digital historian. I knew I needed some sort of relational database because there are 8 boxes of papers from Consul Maury's family at the University of Virginia, and countless other boxes of Maury family material elsewhere (although which Maury family I've yet to determine).
I have organized the database around three primary tables: document, person, and place. Places and persons in this database refer only to those from and to which letters were sent, not those mentioned in the various letters. The document table has the largest number of fields, as I intend to add controlled keywords and transcriptions for most documents. As the graphic below shows, the places and people are linked to documents through a join table, one for author and one for receiver. I may eventually add a secondary table linked from document for repositories, so that I can keep better track of what I have and have not yet looked at.
My main goal with building the database is to have somewhere to put data in an organized way so that I can easily access it later. I may simply use the records as reference for writing, or I might submit queries to run some topic modeling.
The version I have posted here is read-only. I was mostly successful creating full CRUD (create, read, update, delete) using php, but no matter how hard I tired, the edit document page would break at some point. I intend to explore alternative tools for interfacing withMySQL, particularly Ruby on Rails given Jean Bauer's success with Project Quincy.