On capturing information requirements in process specifications

Process and information modelling Business process modelling has been in vogue for nearly 20 years, and its origins are much older than that. The growth in the last 20 years can be attributed primarily to inexpensive and powerful computer graphics technologies, which enabled rapid display and modification of complex diagrams. Software process modelling goes back to the flowchart era of the 1950s and 1960s, and reached its high point in the Structured Analysis and Design era (1970-85). Structured analysis taught that one defined information systems by modelling the to-be business process at a high level and refining each high level activity to its component activities, flows and decisions, and repeating this process until the "activities" could be reduced to a few machine instructions. But information requirements were described only at a very high level. In a language of the time, like IDEF0 [1], the basic idea of "information" is essentially a "document" idea. A body of information is characterised by a name and an overall description – the individual information units are not given at all. And this view of integrating information into process specifications is still very much alive in proprietary