WELLINGTON (12/04/2003) - The biggest risk in IT development is the lack of a clear link through which business knowledge can be translated into technical IT knowledge, says software architect Nitish Verma.
An enterprise architecture provides a common vocabulary for IT developers and the organization's businesspeople to talk about the elements of the business. It establishes a base for the good governance of the associated processes and becomes a significant and measurable business asset.
Although cash flow and balance sheet are key measures of a company's success, he says, running a company just on the financials "is like flying an aircraft by looking only at the fuel gauge".
Verma was addressing the New Zealand chapter of the Worldwide Institute of Software Architects at its bimonthly meeting in Wellington last week.
Verma, who has been a consultant for a number of organizations such as ACC and the Ministry of Health, advocates a systems approach to business decisions. This involves figuring out the relationships between aspects of the enterprise and the governance of these relationships over time and maintaining a big-picture view across the whole of enterprise, to identify and manage the most effective "value drivers".
He displayed a table outlining a process of progressive refinement of the business -- its objects, processes, locations, organization events and goals -- through enterprise and system models to the technological model.
The other axis of the table displayed the essential parameters on which to gather information, familiar to journalists as the "six Ws" -- what, how, where, who, when and why.
At the system model level, for example, the elements for consideration are: logical data model, application architecture, distributed systems architecture, human interface architecture, processing structure and business rules model.
He gave practical examples. For a health insurance company he was asked to assess the risks of contemplated new processes and to "build traction between the teams, based on a common understanding".
He evolved a "central architecture document" with "views" which made sense to the various stakeholders, then got them all into one room to discuss their various perspectives on each change.
Verma acknowledges enterprise architecture is no "silver bullet" capable of solving all problems by application of a set of rigid methodologies.
"You have to live and breathe (each) organization" to judge what parts of the methodologies work best in its particular environment.