This article offers a commentary on Richardson, Ormerod, and Shepherd (in p
ress) while building on the previous discussion in this journal of the rela
tive merits of task analysis and systems analysis in human-computer interfa
ce design [1,2,7]. The SGT scheme described by Richardson et al. represents
a valuable contribution to the work analyst's toolkit. However, it is limi
ted in the extent to which it can identify the information requirements ass
ociated with unanticipated events. The abstraction hierarchy [23] is an eve
nt-independent work domain analysis technique that can be used to overcome
this limitation while still satisfying the criteria of device-independence
and psychological relevance. Future research should integrate the complemen
tary advantages of SGT and the abstraction hierarchy into a single, unified
framework for work analysis. (C) 1999 Elsevier Science B.V. All rights res
erved.