Programming Fundamentals/Program Design

From Wikibooks, open books for an open world
Jump to navigation Jump to search

Overview[edit | edit source]

Program design consists of the steps a programmer should take before they start coding a program. These steps when properly documented will make the completed program easier for other programmers to maintain in the future. There are three broad areas of activity:

  • Understanding the Program
  • Using Design Tools to Create a Model
  • Developing Test Data

Understanding the Program[edit | edit source]

If you are working on a project as one of many programmers, the system analyst may have created a variety of documentation items that will help you understand what the program does. These could include screen layouts, narrative descriptions, documentation showing the processing steps, etc. If you are not on a project and are only creating a simple program, you will likely have only a brief description of the program’s purpose. Understanding a program's purpose usually involves understanding its:

  • Inputs
  • Processing
  • Outputs

This IPO approach works well for beginner programmers. It might help to visualize the program running on a computer: You can imagine what the monitor will look like, what the user must enter with the keyboard, and what processing or changes will be made.

Algorithm[edit | edit source]

An algorithm is a series of specific and finite instructions that produce a result (output), Algorithms are everywhere. For example, in a recipe, directions of a GPS, how to tie a tie, etc. Flowcharts and pseudocode are very useful tools to organize and design algorithms. However, in order to develop a useful algorithm, it is necessary to:

  1. Understand the problem
  2. Define an input
  3. Process the input data
  4. Expect output
  5. Test and analyze data

Algorithms are the basis of any computer program. Before writing a single line of code it is necessary to design an algorithm that solves the problem. Therefore, a good programmer must be a good problem solver and be knowledgeable of their own inputs.

Using Design Tools to Create a Model[edit | edit source]

At first, you will not need a hierarchy chart because your first programs will not be complex. But as they grow and become more complex, you will divide your programs into several modules (or functions).

The first modeling tool you usually learn is pseudocode. You will document the logic or algorithm of each function in your program. At first, you will have only one function, and thus your pseudocode will follow closely the IPO approach above.

There are several methods or tools for planning the logic of a program. They include: flowcharting, hierarchy or structure charts, pseudocode, HIPO, Nassi-Schneiderman charts, Warnier-Orr diagrams, etc. Programmers are expected to understand and create flowcharts and pseudocode. These methods of developing a program's model are usually taught in computer courses. Several standards exist for flowcharting and pseudocode and most are very similar. However, most companies have their own documentation standards and styles. Programmers are expected to quickly adapt to any flowcharting or pseudocode standards for the company at which they work. The other methods that are less universal require some training which is generally provided by the employer.

Later in your programming career, you will learn about using application software that helps create an information system and/or programs. This type of software is called Computer-Aided Software Engineering (CASE).

Understanding the logic and planning the algorithm on paper, before you start to code, is a very important concept. Many students develop poor habits and skipping this step is one of them.

Develop Test Data[edit | edit source]

Test data consists of the programmer providing some input values and predicting the outputs. This can be quite easy for a simple program and the test data can be used to check the model to see if it produces the correct results.

Key Terms[edit | edit source]

algorithm
Series of specific and finite instructions that produce a result.
Computer-Aided Software Engineering (CASE)
Application software that helps create an information system and/or programs.
IPO
Inputs – Processing – Outputs
pseudocode
Written statements used to convey the steps of an algorithm or function, not actual code.
test data
Providing input values and predicting the outputs.

References[edit | edit source]