Beginning Database Design Solutions, 2nd Edition

12Building a Data Model

The previous chapter described requirements gathering for the Pampered Pet database project. It took the basic requirements and used them to build the fundamental entities that will take part in the database's operations.

This chapter builds more formal data models describing those entities. Semantic object models emphasize the entities' fields, and entity-relationship diagrams emphasize the relationships among them.

In this chapter, you'll see examples of:

• Converting requirements entities into semantic objects
• Splitting off repeated data into new objects
• Converting requirements entities and semantic objects into entity-relationship diagrams
• Converting semantic object models and entity-relationship diagrams into relational models

SEMANTIC OBJECT MODELING

Semantic object models have the advantage that they are relatively close in structure to the kinds of entity definitions that you typically get out of requirements gathering. They focus on the attributes that objects have. That is the same type of information that you get by studying the customer's needs and user interface mockups, and then figuring out where those mockups will get their data.

Building an Initial Semantic Object Model

To build a semantic object model, review the tables showing data integrity needs that were presented in the section “Determining Data Integrity Needs” in Chapter 11, “Defining User Needs and Requirements.” The chapter's text showed the data needed by the Order and ...