The Closed World Assumption Revisited - Teradata Vantage

Teradata® VantageCloud Lake

Deployment
VantageCloud
Edition
Lake
Product
Teradata Vantage
Published
January 2023
Language
English (United States)
Last Update
2024-04-03
dita:mapPath
phg1621910019905.ditamap
dita:ditavalPath
pny1626732985837.ditaval
dita:id
phg1621910019905

Any tuple in a relation body is assumed to conform to the Closed World Assumption (see The Closed World Assumption). With the addition of nulls to the database relational model, the CWA can no longer be assumed to be true. As an example, consider the following Venn diagram.



In contrast with the Venn diagram in the topic The Closed World Assumption, the sets of true and false propositions overlap, represented by the darker shading. That is, there is potential overlap between those sets. For example, suppose the employee table has an emp_age column that permits nulls. The company hires an employee who does not disclose her age in the application form. All other information for the employee is entered.

Because the age for this employee is missing, the tuple that represents her cannot be evaluated as true or false, and therefore falls into the ambiguous intersection of the Venn diagram.

Later, this employee is discovered to be age 15, which violates a constraint on the emp_age column that requires all employees to be at least 18 years old. The proposition for this employee now evaluates as FALSE, and the tuple that represents her is no longer valid in the database. Nevertheless, that tuple was treated as valid from the time its data was entered until the time the age of the employee was discovered to violate the constraint specified for the emp_age column, and that information was used to produce reports which are now known to be invalid.