Abstract
The aim of this article is to introduce and to elaborate on computational aspects of TshwaneLex, a new South African software application for dictionary compilation. A brief introduction to the lexicographic needs for such a computer program in South Africa will be given, followed by a description of the most salient features of the software package. Four key features of TshwaneLex will then be analysed, viz. (a) the data presentation in a tree structure, (b) the extendible input/output architecture, (c) database loading issues, and (d) the cross-reference system. For each of these the challenges encountered and/or the design decisions that have been made will be examined.