New Presses for Old Grapes: I: Multiple Classification Analysis

Searching, gathering, organizing, and retrieving data are basic tasks for historians. As long as historians work by themselves, decisions concerning data format, data exchange, computer platform, and the like remain secondary. Teams of historians often use relational databases for centralized data storage. However, fundamental risks are implicit when one uses databases. Among technical considerations, the process of transformation between the source and the database is a deciding factor. For those who gather data, the effective use of such possibilities as data exchange, compatibility, and simplicity of survey and the reuse of the data in other contexts and platforms becomes increasingly important. In contrast, the user's needs include the possibility of data verification and use of the data for more than one question. Many relational databases have considerable shortcomings because the stored data lose the visual characteristics, the syntax, and the semantics of the original source. The EuroClimHist database environment is a part of the NCCR Climate project. It uses a data tool, written in Java for the gathering of documentary data, which generates extensible markup language (XML) files for data exchange.