Naming and Defining Entities and Attributes › Definitions and Business Rules

Definitions and Business Rules

Business rules are an integral part of the data model. These rules take the form of relationships, rolenames, candidate keys, defaults, and other modeling structures, including generalization categories, referential integrity, and cardinality. Business rules are also captured in entity and attribute definitions and validation rules.

For example, a CURRENCY entity defined either as the set of all valid currencies recognized anywhere in the world, or could be defined as the subset of these which our company has decided to use in its day to day business operations. This is a subtle, but important distinction. In the latter case, there is a business rule, or policy statement, involved.

This rule manifests itself in the validation rules for “currency-code.” It restricts the valid values for “currency-code” to those that are used by the business. Maintenance of the business rule becomes a task of maintaining the table of valid values for CURRENCY. To permit or prohibit trading of CURRENCYs, you simply create or delete instances in the table of valid values.

The attributes “bought-currency-code” and “sold-currency-code” are similarly restricted. Both are further restricted by a validation rule that says “bought-currency-code” and “sold-currency-code” cannot be equal. Therefore, each is dependent on the value of the other in its actual use. Validation rules can be addressed in the definitions of attributes, and can also be defined explicitly using validation rules, default values, and valid value lists.