Business Rules: Capturing the Most Elusive Information Asset - High-Performance Web Databases: Design, Development, and Deployment - page 112

stated followed by a nonaction verb (e.g., “is”) and a meaningful descrip-
tive noun. It is important to make clear distinctions (e.g., “agreements” ver-
sus “requests,” “legal” versus “not legally binding,” and “written” versus
“verbal” characteristics). The person reading a business rule should not be
confused — the rule should not be characterized by any ambiguity.
Similarly, each relationship definition business rule should be expressed
in terms of a meaningful action verb or set of verbs along with minimum
and maximum cardinalities. Generic verbs (e.g., “has,” “exists,” and “is”)
fail to fully describe a relationship. Instead, more specific action verbs
should be used to describe the relationship (e.g., “generates,” “groups,”
and “manufactures”) because they convey more relevant information. To
make a relationship definition rule complete, the relationship should be
defined in both directions.
Explaining the purpose of a business rule is sometimes important in con-
structing a complete business rule. For example, a business rule describing
a purpose could be “Although current skills are specified on an application,
the agency maintains a complete list of a caregiver's skills for historical
purposes.” This rule explains the reason for associating skills with the car-
egiver entity rather than with the application entity. In many cases, includ-
ing the purpose in the business rule may make the difference between an
acceptable business rule and a quality one. Purpose gives the reader a
frame of reference that allows for greater understanding of the business.
Consistency
Each business rule must be consistent with all other business rules in the
enterprise and the repository of rules. Contradictory business rules create
confusion for the end user querying the database as well as for developers
and the programmer trying to use them as a guide for program logic.
Because consistent business rules lay the foundation for a shared data envi-
ronment, any inconsistencies must be flagged, investigated, and resolved.
Unfortunately, such consistency checks are typically done manually.
Independent
Probably the most difficult task of constructing a quality business rule is
to ensure it is unbiased in attitude towards any one information system or
business area. Because a shared business rule must be valid for the entire
enterprise, any business rules that pertain to a specific business group
should be identified in the rule itself. Even so, these rules may be of interest
to other parts of the enterprise.
Business
Every effort should be made to express each business rule in meaningful
business terms. Although business rules should be in the language of the