Diagram Conventions   «Prev 

Naming Strategy

Naming fields

When creating a medium-to-large database with a dozen or more tables and dozens of fields, it is especially important to have some sort of naming strategy for fields (attributes) that you apply consistently across the database. Often, you will implement the naming strategy when you verbalize the ER diagram.
One very common practice among database designers is to use the first two, three, or four letters of a table’s name to distinguish between similar fields across tables. You have already seen this strategy applied throughout this course to the Stories on CD case study.
The ID (identification) fields for Customers and Distributors are similar in the case study; these were named CustID and DistID respectively. My own preference is to use capital letters to separate the table portion of the field’s name (Cust) from the descriptive or identification portion (ID). Other designers prefer to use the underscore character (_) to separate the two portions (for example, cust_ID).
The majority of database designers agree on one thing: Do not use a space to separate the two portions (Cust ID, for example, is not acceptable), even if the RDBMS you are using supports multiple words to identify fields. Not all RDBMSs support multiple words, and this fact alone makes it an unwise practice.
Finally, be careful about using abbreviations to name fields. Do not use abbreviations that are not intuitive. In our case study, for example, first and last names of customers are specified as CustFirst and CustLast, as I feel this abbreviated format is intuitive. Other designers are not comfortable using abbreviations at all, and would opt for a format such as CustFirstName and CustLastName (or, cust_first_name and cust_last_name).
Here, as elsewhere, consistency combined with standard database design practices will serve you well.