Software Testing Essay

The Principles of Software Testing

Delivering a quality product is not solely down to how good the developer is, even though that helps, it is down to the quality of the original analysis and understanding of the requirement. Software testing is carried out to provide developers with a report on the quality of the program or application under testing. It allows the business to get an objective and informative view of how the product works. The purpose of testing is to resolve as many issues as possible before it is released. Each of the following testing principles help play a part but collectively they look to provide a working solution.

Exhaustive testing is not possible; program testing can only show the presence of defects, never their

If no defects are found, you cannot assume that the program is completely correct. It is important to understand how the function or application will be used as this will make you prioritise areas where you think the number of defects is high.

Testing applies through the software life cycle and is not an end-of-cycle activity. The sooner testing begins, you have more time to iron out defects and you will save money. When defects are found they are much easier to fix as you can fix an incorrect requirement much quicker than you can change functionality in a large application or program that is not working as requested. Also, when you test each stage of the development cycle, you can test each stage as it becomes available, thus preventing testing being squeezed in at the end of the lifecycle.

Understand the reason behind the test. In test-driven development you first think about the how you want the program to work and then write a test for it. In some cases it will be obvious what the program should do. You then write a test that makes sure that the assumption worked. The design of the test should be well understood and test exactly what it's supposed

In order to ensure an effective test it is important that requirements of the system are understood as this will lead to effective testing. User Requirements should be well known before test case design. Testing should be performed against those requirements. The test case scenarios should be documented and scripted before testing commences. If the user requirements are not understood then test cases cannot be designed correctly if you do not understand the user needs it is likely that test cases will be incorrect and more software faults will occur when it is released. A test case should include a description of the input data to the program and a precise description to the correct output of the program for that set of input data. A necessary part of test documentation is the specification of expected results, even if providing such results is impractical. In testing the test cases you we will minimise the amount of errors in the delivery of the application and it will show how the system will be used.

Defects occur in convoys or clusters, and testing should focus on these convoys. There is no equal distribution of errors within the product. Where there is one error, there is a high probability that there will be more. The testing process must be able to respond to this behaviour quickly. A small number of modules will contain most of the defects for operational