Extended subject access to hypertext online documentation. Parts I and II: The search-support and maintenance problems

Abstract

The DFT (DOCUMENT, FIND, THESEUS) online documentation system resembles other hypertext software in managing a full-text database with reference links (pointers) between passages (nodes). But DFT's prime role as an end-user reference service at a computer center where text updates are frequent required extra access methods often neglected in hypertext systems. To solve the problem of rapid, high-recall search for specific answer passages we combined syntactical interface features (substring and fuzzy matching of search terms) with a semantic expansion of the system's entry vocabulary. Extensive aliasing increased the number of descriptions under which any passage could be found. Conversion of the NBS GAMS classification scheme for mathematical software into descriptors allowed us to impose a virtual organization on our subroutine documentation that supports easy, task-oriented retrieval without tedious path walking. To solve the problem of reliable database maintenance amid frequent passage revisions we developed software tools for flexible input of text and entry-term changes, as well as for thorough and versatile status reporting. These, together with special features for testing and privately simulating changes and for coordinating the sequence of text and entry-term updates for maximum efficiency, yielded a robust and practical answer-delivery system.

Ease of use and the richness of documentation adequacy

“…In the special case of online documentation, the access aids may be external as well--search mechanisms built into the software by which readers find and display text passages (Bates, 1988;Girill, Griffin, and Jones, 1991). Either way, the ease-of-use approach highlights the extra, access-support features that readers need in that subset of technical prose intended to answer questions.…”