Digital humanists use databases to store and locate information related to their field of study. Because databases are complex systems, problems arise and make research difficult for various types of users. For this reason, digital humanists have looked to a relational database management system (RDBMS) model to “remove some of the complexities and idiosyncrasies” that exist with “database design.” Since making information accessible is of the most importance, digital humanists concern themselves with designing and successfully implementing relational model databases, which is not a simple process. However, the steps in creating such a database are just as important to the “scholarly endeavor” as the finished “product.”

The relational model was first introduced by E.F. Codd to use space more efficiently while avoiding redundancies in formation, which was a problem with earlier database systems. Codd’s idea was meant to store information (in a domain) and to allow users to navigate easily through that information while finding all relevant matches to inserted key phrases. Relational model databases are more efficient in finding matches because information is matched through the use of attributes and entity relationships. This means that certain key phrases are matched with more similar key phrases that the designer has linked together through records and fields. For example, the database will not only “connect authors to works in some generic way” but will also connect a single author with multiple texts, or a singular publisher with multiple authors or locations. Primary keys, the references used to help users find more specific and important data, are unique to relational model databases. They are set up through tables of columns and rows. This allows for primary key values in “one table” to “be referenced from multiple tables.”

There are five steps in the process of eliminating redundancy, which are also used to evaluate the “soundness” of a design. These five steps move from the “first normal form” (where a designer checks for any repeated phrases in the tables) to the “fifth normal form” (where all of the redundancy has been eliminated). To make the design useful, it must be translated by using a database schema, which allows the “machine” to understand it. Structured Query Language (SQL) is used by most digital humanists to make the design readable, and it works much like a style sheet would to make the design look attrative. SQL uses datatype declarations (commands) to limit the characters that can be entered into a search, and this helps users find desired information faster than earlier search systems.

Although there may be many users of a database, administrators of these RDBMSs support limiting access to most users so that “administrative data stored in the database” is protected. Still though, users have access to most of the database’s information, but usability of the data becomes fragmented when RDBMSs are too often updated with “insertions, deletions, and updates.” Since these complications do occur, and because many users of databases are not experts in design, it becomes neccessary to create a database that protects users from problems with SQL commands. There are several forms of software that make using databases in the humanities much easier for non-experts. They include middleware systems, DataBase independent systems (used to work between systems and translate “SQL syntax” for other databases), Java Database Connectivity systems, and PHP Hypertext Processors (a programming language used in HTML pages and translated by other systems). Research continues for an even more efficient system, although it looks as if the relational model “will continue to be the dominant one for many years to come.” Digital humanists are now looking for a way to increase “collaborative database creation” so that knowledge becomes more widespread.