Humanities computing: essential problems, experimental practice

The application of computing to the disciplines of the humanities has two principal outcomes: useful results for the field of application and failures completely to demonstrate what is known. These failures, an inevitable feature of modelling, point to the key question for humanities computing, how we know what we know, and so to the beginning of its own scholarly enquiry. This, I argue, proceeds along three branches, the algorithmic, the metatextual, and the representational. Examining the first of these I argue for research toward an open-ended, interoperable set of primitives based on previous work in the field and designed for the emerging digital library environment. To set the stage for their further development I argue that the field as a whole does not wait on a theoretical formulation of what humanists do, rather should look to the tradition of experimental knowledge-making as this has been illuminated in recent years by historians, philosophers, and sociologists of science. (100 References).