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

Silverston c04.tex V2 - 11/21/2008 3:02am Page 151

Level 2 Recursive Pattern 151

After starting the effort, XYZ realized that they needed to further break

down tasks into subtasks. The last two rows in Table 4-2 illustrate the

flexibility of this pattern to accommodate new requirements such as this

one, without changing the data model. Thus a new WORK EFFORT TYPE

name of ‘‘Subtask’’ may be added and the last two rows show that the task

‘‘Discover Target Requirements’’ is made up of ‘‘Conduct Interviews’’ and

‘‘Model Target Requirements.’’

Some work efforts may depend on others being completed before they can

start. Figure 4-7 shows that ‘‘Each WORK EFFORT may be dependent on

one and only one WORK EFFORT. Table 4-3 illustrates an example of this

and shows that the ‘‘Create Mapping’’ instance of TASK is dependent on the

‘‘Create Mapping Template’’ being completed. This is an example of using this

pattern not for a hierarchy or aggregation, but for a peer-to-peer relationship.

Table 4-3 Example of Level 2 Recursive Pattern, DEPENDENCY

WORK WORK WORK WORK EFFORT. WORK WORK

EFFORT. EFFORT. EFFORT WORK EFFORT EFFORT. EFFORT TYPE.

WORK NAME TYPE. ID (FOR NAME (FOR NAME (FOR

EFFORT ID NAME PREREQUISITE) PREREQUISITE) PREREQUISITE)

9010 Create

Mapping

Task 9005 Create Mapping

Template

Task

In Figure 4-7 there are two recursive relationships, one stating that ‘‘each

WORK EFFORT may be made up of one or more WORK ...