The Problem

Over the years, organizational management has become increasingly disenchanted with their information processing support.  In many cases they have begun to seriously question whether the information technology group truly helps their organization or merely depletes business resources while providing superficially efficient services.

Information technology departments are often accused of being:

Too slow in responding to problems.

Too expensive in implementing solutions.

Too reliant on jargon to justify expensive hardware and software.

There is another side to this problem, of course.  For one thing, organizations rarely articulate their goals in a coherently organized manner.  Historically, they haven't really had to do so.  Furthermore, organizations often relegate their information technology departments to a technical supporting role.  (This is often true even in companies whose product is packaged software.)  As a consequence, the information technology department is often the last to find out about changes in corporate policy that impact software specifications.  As a result, they are often under great pressure to patch systems together with unreasonably short notice.

In any case, as information technology professionals we ought to assume part of the responsibility for helping to turn this situation around.  We must learn how to analyze and understand the organization's problems, goals and objectives… learn how to build effective integrated systems that help rather than hinder the organization… and learn how to make communication a two-way process.

The Solution

The key to resolving most of these issues is to establish a clear requirements specification before beginning to design and code solutions to perceived business problems.  We recommend beginning with a model of business information requirements that can be used as a baseline for any number  of event-based models of application requirements. 

The underlying premise behind an information oriented approach to systems development is that the information requirements of any system remain fairly constant regardless of how political responsibility for each piece of the system is divided up.  Likewise, the list of responses that a system must make to external events remains relatively stable.

This makes the planning job possible.  It means that we don't have to fully document all the various processes and procedures and management structures and screens and forms in order to come up with a reasonable plan for implementing systems.  We look instead at essential processes and activities and define their requirements in informational terms rather than in political or procedural ones.  The resulting definitions don't change much.  They comprise the basis of a set of integrated requirements definitions that can be specified to as detailed a level as is necessary without becoming concerned with specific implementation constraints.

Datamorphosis presents two core seminars that lay out a fundamental approach to the requirements specification process.  The first is the Information Modeling Workshop that describes methods for creating graphic models and specifications of an organization's information requirements using entity relationship diagrams. The second seminar is the Application Modeling Workshop that describes methods for creating graphic models and specifications for a set of business applications using essential systems analysis.  It includes graphic models at the event and sometimes the state transition level and links to the definitions in the information model. 

In addition to the two core requirements specification courses, we have several additional seminars that focus on various kinds of analysis and design approaches.