Relational databases are the backbone of many humanities projects that engage with and produce scholarship. The relational model—storing information in key-based table structures—is a literacy in three dimensions, providing authors an opportunity to create webs of knowledge that drive digital production. Yet some properties of relational databases that have helped make them popular—they are fast and comfortably similar across table schemas—are a result of normalization derived from their origins assisting bank transactions and military interpretations. When faced with scholarly nuances that have trouble conforming to relational model rigidity, technologists working in the humanities circumvent normalization with creative and sometimes bizarre schemas and data structures.
Innovative software architectures emerge when traditional relational models can't capture a project's nuances. Incorporating semantic web principles, many humanities projects store data along with its meaning, merge knowledge ontologies, and facilitate sharing via open networks. This talk will discuss scholarly software challenges and innovative solutions by traversing four case-studies of systems in development—with specific code from each. Featuring a range of relational and semantic approaches: the Mukurtu Archive content manager based on Aboriginal cultural protocols; Metaserver, an art archive registry produced by UMaine's Still Water lab; Scalar, a media studies publishing platform by USC's Vectors Journal; and Magic, a system for annotating code with cultural material and vice verse.