In recent years the size and complexity
of software applications has grown markedly. Commensurate with the
growth of software applications, the development cycle of these software
applications has also grown. As a result of this there are usually a
large number of people involved in the development of any one software
application. In some cases a customer or end user may have particularly
expert knowledge with respect to one or more processes or areas, and
wish to translate this expert knowledge into a software application for
the automation or implementation of this process or area.
In a typical software development process, the customer or end user may
communicate this expert knowledge with respect to the behavior,
operation or interface of the software application; architects,
designers and systems analysts may be responsible for translating these
desires, specifications or knowledge into an over-arching architecture
for the software application and its interface while developers or
programmers may be responsible for actually implementing these designs
to create a functional software application and user interface. A
feedback cycle may then ensue, with the application being presented to a
customer or group of end users for evaluation. This evaluation process
may result in desired changes, which in turn, may result in another
cycle of the development process for the software application.
Consequently, these software development processes are prone to
inaccuracies and delays as a result of a myriad number of factors such
as: problems in translating the customer’s desires, knowledge or
requirements into a system design or architecture, trouble developing a
system which meets the desires or requirements of the customers, or
human factors as simple as the customer, end user or developer changing
his or her mind about the way the interface to the system should look,
feel or operate.
To address these delays a variety of automated approaches have been
developed to assist in the life cycle of the software development
process. Typically, these automated approaches allow a software system
or application to be modeled, for example with a modeling language such
as universal modeling language (UML). From these models a set of class
definitions in an object oriented language such as C++ or Java may be
created. These class definitions may be the classes needed to implement
the software application and an interface for the software
application.
The software classes generated by these automated approaches are,
however, typically stub functions only. The actual coding of the
software application must still be accomplished. Additionally, a user
interface must be developed for the software application which will
allow operation of the software application. Thus, while these
automated approaches provide a framework for the development of a
software application these automated approaches do not actually generate
a complete and functional software application and user interface. As
these methodologies still require a significant amount of human
intervention in order to generate a functioning software application and
user interface the problems discussed above with respect to human
factors cannot be completely ameliorated.
Thus, it is desired to be able to capture expert knowledge and utilize
this expert knowledge to generate a complete, functional software
application and user interface in an automated fashion.