Databases were originally developed using mainframe computers but are now common on personal computers and in networked environments.

Following is a brief description of the concept of databases. In order to understand databases, it is important to begin with the data field.

Data field: A space allocated for a particular item of information. In a database, fields are the smallest units of information you can access.

A data field contains a single piece of information (first name, family name, ministry, employee number, salary and so on). A collection of data fields comprise a record such as, in this example, an employee record.

Database record: A complete set of information in a database; records are composed of fields, each of which contains one item of information.

A collection of records (in this case, employee records) comprises a database. Structured databases typically store data that describes a collection of similar entities. ‘Salaries and benefits’ is an example of an entity; ‘education/training’ is another entity. A n employee database stores data about the employees in an organisation. A medical database stores data for a collection of patients. An inventory database stores data for a collection of items stocked in a warehouse.

A data structure refers to the overall design for records in a database, such as the fields contained in a certain view and the configuration of fields and sub-fields for data entry. The data structure should present a uniform format to facilitate information exchange.

Data structure: A scheme for organising related pieces of information. The basic types of structures include: files, lists, arrays, records, trees, tables. Each of these basic structures has many variations and allows different operations to be performed on the data.

There are three basic database models.

• Hierarchical databases exhibit a branching structure, with information arranged into sets and sub-sets; getting to a particular piece of data may require going through several vertically ordered files. An example is the process involved in finding a distant cousin on a family tree.

• Network databases offer many more direct connections between files, but, similar to hierarchies, the links are predefined and are difficult to change or adjust.

• Object-oriented databases link self contained entities (or objects) together. Objects can be text, a picture, a piece of film or any item that can be individually selected and manipulated. This kind of database is particularly useful for organising large amounts of disparate information, but they are not designed for structured numerical analysis.

The limitations found with these types of databases explain why most organisations have turned to relational databases. Relational databases not only accommodate multiple views but allow new links to be forged as needs arise. Relational databases are powerful because they require few assumptions about how data is related or how it will be extracted from the database. As a result, the same database can be viewed in many different ways.

Relational database: A database that spreads information across different tables while maintaining links between them.

A relational database stores facts in tables called relations. The only requirement is that the information must be capable of being laid out in rows and columns (similar to a list of names, addresses and phone numbers). An example from a university might help to illustrate the concept. In a very simplified view of a university database, each facet of the university’s administration would be represented by a table containing information within the remit of a single department. The admissions office, for instance, keeps track of students by ID number, name and major field of study. The personnel office keeps records of the department, rank, names, and social identity numbers of the teaching staff, and so on. The following figure shows examples of relational tables.

A relational database stores information across different tables and connects them with links in the computer.

Isolated from one another, these tables would be nothing more than a simple computerised filing system. Tied together, however, they become a database, with each table offering access to the information held by all of the others. This versatility stems from deliberate duplication of columns in two or more tables, resulting in a tool called a common key. If, for example, two tables containing information about students each possess a column containing student identification numbers (ID) then this number can be used to find the row in each table that contains information applicable to any student.

Each column in a table represents a single data attribute, or characteristic, of the table’s subject. A column might contain, for instance, the identification numbers of the students or each professor’s department.

Each row, or record, in a table contains all the information about a single entry. In the case of a student, a record might include, in addition to an identification number, the individual’s first and last names and major course of study.

Belonging to both a row and a column, an occurrence is the basic unit of a relational database table. The occurrence contains the value of an attribute of a single record. In some cases, the value is a string of letters making up a word or words, in others a set of numerals, such as a student’s identification number or job title.

Officials in the admissions, personnel, and finance offices have direct access to each other’s information through the ‘Dept’ attribute, common to all the tables. Similarly, tables in the admissions and registrar’s offices are joined by the ‘ID’ attribute for the student identification number. Relations in the registrar’s and finance offices are connected by a ‘section’ attribute that identifies subdivisions of courses.

Database management software is used to create and manipulate databases. Data access software is usually used to search databases. The data access software understands the structure and details of the database which means the user simply has to enter his or her search specifications, using either a menu, a keyword search engine, a query language or a natural language (among others).

Menu: A collection of onscreen choices given to the user to help him or her interact with a computer system.

Database menus are similar to those used in most software. They are typically arranged as a hierarchy so that after the user makes a choice at the first level of the menu, a second series of choices appears.

Keyword searching permits access to databases through the use of keywords.

Keyword search engine: A program that allows a user to search a database by an index entry that identifies a specific record or document.

Keyword search engines are especially popular for searching the many documents stored in a free form database such as the World Wide Web. To use a keyword search engine, the user types in a word and the search engine locates areas in the database where the word or related information can be found.

When information in a database needs to be accessed quickly, it is usually stored as a structured database. However, the structure in structured databases can cause a problem for users who might not know the format for the records in a database. One way to help users search structured databases is by providing a ‘query by example’ user interface based on the use of a query language.