The Data Model Resource Book, Volume 3: Universal Patterns for Data Modeling

Silverston c05.tex V2 - 11/21/2008 3:04am Page 197

Level 2 Classification Pattern 197

Synopsis

In this section you saw that the Level 1 Classification Pattern provides a very

specific way to model classifications. This pattern is significant because of its

simplicity. This pattern uses attributes to capture which instances of an entity

are classified into which ‘‘types.’’ The level 1 pattern is useful as part of a

statement of scope and visual representation of the different classifications for

a particular entity. This pattern can help you start your data modeling effort

in a very straightforward fashion, exposing the terminology a company uses

to categorize its types. It may even be useful as basis for a prototype.

This pattern is not meant to be used as the basis for implementing databases

for almost all circumstances because there are usually serious pitfalls regard-

ing data redundancy and data integrity issues in this pattern. Also, if new

categorizations need to be added or if the nature of the classification changes,

new attributes need to be added, and thus, this pattern does not provide much

flexibility for future needs.

Level 2 Classification Pattern

Many enterprises need a more flexible and normalized approach when model-

ing classification data. In the previous pattern, each of the entities was classified

into types that were captured as attributes. In this pattern the types are not

captured as