Information Systems Development Methodologies - High Level Models and Methodologies for Information Systems - page 53

Fig. 3.4 Stages of JSD

can be seen as a new approach to reality-driven methodologies that center on
designs associated to reality, and information-driven methodologies that have a
propensity to focus on data and processes.
Consequently, the main purpose of JSD is to guarantee that the

final system is an

accurate re s perceptions of the

real world, by considering the current state of the real world and describing it
(Rohde 1995 ). This methodology, much like soft systems methodology and similar
approaches, also attempts to bridge the gap between technical issues and real-world
contexts. It is rooted on established criteria for modeling real-world entities, and the
conceptual separation of system speci

ection of not only the developer ' s but also the user '

cation and system implementation (Savage

and Mingers 1996 ).
For Cameron ( 1986 ) and Jackson ( 1992 ), there are three stages to a JSD
methodology, as outlined in Fig. 3.4 . The starting point is the premise that the real
world is comprised by sequential components ordered in time, called events or
actions, and the process will begin by acknowledging which of these components
are of interest to the issue at hand. So, the first stage (modeling) is essentially

concerned with the real world, not with the system that is being developed. This
event list is the

first mechanism by which the scope and purpose of the system is

de cation of relevant entities and common actions. All

of these components are organized into event and data models (Jackson 1992 ).
The network stage essentially entails the analysis of the processes and connec-
tions that

ned, followed by an identi

flow between the various components of the system, described in the
model stage. These connections are organized into a system speci cation diagram,

with different shapes representing different forms of process communication
(Jackson 1992 ). Thus, a systemic description of inputs and outputs of the system is
achieved, allowing for the development of the new system cations.

Finally, the implementation phase is mainly based on two issues: scheduling of
the process speci

' s speci

cations, and organization/management of the data. The tools and
techniques used in this stage are essentially the tools of JSP (Jackson 1992 ).
The JSD methodology has been used to develop systems of all sizes. However,
the characteristic rigor of this methodology can make it considerably complex to
use.