Building a Data Model - High-Performance Web Databases: Design, Development, and Deployment - page 213

Exhibit 14-2. RENTER sample value listing.

REENTER

MAX-MONTHLY
-RENT-AMT

NAME (UIX)

ADDRESS PHONE-NUMBER

Harry Smith 12 Oak Lane, Hopetown, NY 01111 212-984-3158 400

Josephine Morgan 5 Central Ave, Dallas, TX 75080 214-232-7990 650

in the sample value listing (sample entity occurrences with their respective
attribute values) in Exhibit 2.

Typically, all of the attribute values are not needed to identify a particu-
lar entity occurrence. For example, a particular RENTER can be identified
by an assigned RENTER-ID. This identifying attribute or set of attributes is
known as an entity identifier. The entity identifier for PROPERTY is the
combination of two attributes, STREET-ADDRESS and TOWN-STATE-
ADDRESS. Both attributes are needed to identify a particular property,
because multiple properties may have the same value for STREET-
ADDRESS (in different towns) or the same value for TOWN-STATE-ADDRESS
(with different street addresses). RENTAL-AGREEMENT has a primary key
consisting of three attributes: PROPERTY* STREET-ADDRESS and PROP-
ERTY*TOWN-STATE-ADDRESS (identifying the property) and BEGIN-DATE
(identifying the rental period for that property). Entity identifiers are writ-
ten above a horizontal line in the logical data model diagram (as in Exhibit 1)
or on the far left above a primary key designation in a sample value listing
(as in Exhibit 2).

Some assumptions about definitions of the attributes are made when
choosing the entity identifiers. These definitions should be explicitly
understood and recorded within the design documentation or data dictio-
nary. A data dictionary is a manual or automated repository of information
about information systems, databases, data models, users, and access
authorizations. Assumptions about the model are based on naming con-
ventions used in the data model diagram without referring to the data dic-
tionary. All entity, relationship, and attribute names are composed of
English words, are frequently and consistently abbreviated, and are usually
connected by hyphens. Moreover, each attribute name includes one word
(the class word) that indicates the nature of the data represented by the
attribute (e.g., NAME, ADDRESS, NUMBER, AMOUNT, DATE, FLAG).

Some of the attribute names include the name of another entity as a prefix,
followed by an asterisk-for example, RENTER*NAME, PROPERTY*STREET-
ADDRESS, PROPERTY*TOWN-STATE-ADDRESS. These attributes are part of
a foreign key-an attribute or set of attributes that completes a relationship by
identifying the associated entity. The term

foreign conveys the idea that the