Avoiding Pitfalls in Data Modeling - High-Performance Web Databases: Design, Development, and Deployment - page 297

Exhibit 18-20. Chart of pitfalls and remedies. ( continued )

Modeling
Areas

Possible Pitfalls Plausible Causes

Cardinality Suspect cardinality • Model too generalized

• Model too specific and restrictive
• Current requirements emphasis only,
did not allow for future cases

Associative
Entities

1. Restricting the use of

associative entity by
limiting it to resolving
many-to-many
relationships

• Lack of understanding of purpose of
associative entities

2. Syntax error in

cardinality around rule
associative entities

• Lack of understanding of syntax

Recursive
Relationship

Recursive relationship with
an impossible case

• Lack of validation of model

Normalization Data Modeling practice

without normalization
consideration

• Not able to identify entities quickly
upon seeing their attributes
• Attributes are placed in wrong entities
• Did not identify key early enough

Reference
(minor)
Entities

Stored minor entities in
attributes of major
entities instead

• Did not believe in the value of validation
table

Attributes One or more clusters of

attributes not separated
out as an entity when a
major business entity can
be created

• Did not model with normalization
concept
• Did not believe the current application
justifies the creation of new entity
(place holder)

Data Model
Knowledge
Base

Incomplete documentation • Did not decide beforehand what to

document
• Information not available

Modeling
Areas

Considerations Remedies

Entities • Completeness • Conduct brainstorming sessions to

generate brand new ideas
• Use other data gathering techniques
(e.q., JAD); interview can also be used
as remedy or precaution

• Accuracy
• Semantics
• Accuracy
• Semantics

• Identify entities that own attributes on
paper document
• Rigorous definition of an entity
• Close examination of fitness of
attributes to an entity

Relationships • Completeness • Validate business queries that must be

answered against the relationships

• Accuracy
• Semantics

• Examine the data view that supports
the information on a document