Defining the Right Project

I l @ ve RuBoard

The most important question to ask when developing a system is not a methodological question. It is not a technical question. It is a seemingly simple, yet remarkably difficult question: "Is this the right system to make?" Unfortunately, this question is often never asked nor answered . Although misguided methodology or technically tough problems can cause projects to fail, sufficient resources and heroic effort by talented people often can save them. But nothing can save a system that is not needed or that automates the wrong thing.

Before starting a project, there must be an idea for it. The process of coming up with an idea for a system along with a general idea of its requirements and form occurs during the Inception Phase. It finishes the statement: "The system we want does . . ." During this phase of development, a vision for the idea is established, and many assumptions are either validated or rejected. Activities that occur involve the solicitation of ideas, the preliminary identification of risks, the identification of external interfaces, the identification of the major functionality that must be provided by the system, and possibly some "proof of concept" prototypes . Ideas come from many sources: customers, domain experts, other developers, industry experts, feasibility studies, and review of existing systems. It is important to note that any prototyping done during this phase should be considered throw-away code since the code generated is merely to support a list of assumptions and has not been fully analyzed or designed.

The process used during this phase of development can be done formally or informally, but it always involves considering the business needs, the available resources, the possible technology, and the user -community desires along with several ideas for new systems. Brainstorming, research, trade studies, cost-benefit analysis, use case analysis, and prototyping can then be performed to produce the target system's concept along with defined purposes, priorities, and context. Usually, a first-pass cut at resource and schedule planning is also done during this phase. For some projects, the product vision can be sketched on the back of a napkin. For others, the product vision may be a formal phase that is iteratively performed until enough level of detail of the target system has been specified.

An adequate Inception Phase establishes the high-level requirements for a desirable and feasible system, both technologically and sociologically. An inadequate Inception Phase leads to systems so unwanted, expensive, impossible , and ill-defined that they are typically never finished or used.

I l @ ve RuBoard