(Oriental Institute, University of Chicago, USA)
The creation of a virtual archive for Diyala records required the design of an elaborate relational database (using Oracle 9i) that allowed systematic searches but also allowed adjustments to the realities of paper records made on excavations some 70 years ago. This paper presents the conceptual layers of information needed to create an archive and a finds publication from a data architect’s point of view. It discusses the adjustments needed to handle unsystematic record keeping and on-the-fly adjustments made in the paper records as novel situations emerged during an era when excavation recording was becoming grid-oriented rather than architecture-oriented. Some novel “denormalizations” will also be explained, which make the database tables flexible enough to be a pattern for future archival projects elsewhere.