Approximating an information repository for document production

We address the problem of constructing an electronic information repository to be used in a document production environment for technical writers, i.e. a workbench for technical authors. This information repository shall be the core source of different information types for supporting the variety of tasks related to technical writing; it is facilitated through knowledge representation technology. A purpose-driven information repository can only be approached asymptotically, thus the method outlined involves a stepwise folding in of one information cluster at a time, (i.e. domain/world knowledge, text/document knowledge, knowledge about a document's structure, knowledge about tasks, and knowledge about linguistic realisations) and can thus be seen as an iterative dynamic knowledge acquisition and formalisation process. Our methodology is based on (1) empirical and practical concerns, namely what data, information and knowledge the document production processes require, and (2) a thoroughly analysis of the writer's tasks and their potential formalisation and automatisation by computational means. After the initial knowledge acquisition and formalisation process, the resulting knowledge structures contain, declaratively and explicitly represented, those distinctions required to control and support the document production process in various ways and the acquisition of new information in a guided and tailorable way. The information repository forms the knowledge kernel of an Intelligent Assistant System for Technical Authors (IASTA). The paper is based on work-in-progress within the IASTA project.<
>