There are total 7 principles of Testing.
Testing Shows the Presence of Bugs: Testing is the process to find the defect but after 100% testing we can't ensure that our application is error free. We can't sure if tester is unable to find any error in the application then that application is error free.
Exhaustive Testing is impossible: For any application, we can test the application with limited inputs and verify the limited output. It is not possible to test any application with all possible inputs so that we can priorities our test data to cover maximum possibility of error. Suppose for any example you are checking any text box which accept any number only so it is not possible as a tester to check that text box with each possible number so tester can check this with some rule to divide all possible data in different group like tester can check with -ve values, 0, + ve values and any big data like 5653636555565 to check the database limit.
Early Testing: Testing should start as early as possible. it's good if testing started in starting phase of product development which is Requirement Document or Design Document. If defects are found in starting phase then it is low cost to fix them. If we start testing in later phase then the cost to fix them is high due to defect multiplication (Defects would show in different part of Application).
Defect Clustering: It means that the most of the defects found in the small modules of Application. It means that 80% of bugs found in 20% of the modules. It means distribution of bugs are not across the application rather then it is centralized in limited section of the Application.
Pesticide Paradox: In this case if user use the same type of Test Cases again and again to find the defects then it may be possible at some points, user are unable to find any defects because old defects are fixes now and due to same test cases, no new defects found. Due to this complexity, User need to updates the Test Cases regularly to find more defects.
Testing is Context Dependent: Context dependent means that every sites have different natures and the way to test the site is also be different. The testing process that you follow in eCommerce site will be different the way you follow the testing of Social Sites.
Absence of Error Fallacy: Suppose you are testing a Software an d now you are unable to find any new bugs. If you are able to find any new bugs we can't say that now we completed our software. It may be possible that your designed Test Cases are not enough to test the software. it may also be possible that we complete the software but is it same as per client Requirement Document.
Defect Clustering: It means that the most of the defects found in the small modules of Application. It means that 80% of bugs found in 20% of the modules. It means distribution of bugs are not across the application rather then it is centralized in limited section of the Application.
Pesticide Paradox: In this case if user use the same type of Test Cases again and again to find the defects then it may be possible at some points, user are unable to find any defects because old defects are fixes now and due to same test cases, no new defects found. Due to this complexity, User need to updates the Test Cases regularly to find more defects.
Testing is Context Dependent: Context dependent means that every sites have different natures and the way to test the site is also be different. The testing process that you follow in eCommerce site will be different the way you follow the testing of Social Sites.
Absence of Error Fallacy: Suppose you are testing a Software an d now you are unable to find any new bugs. If you are able to find any new bugs we can't say that now we completed our software. It may be possible that your designed Test Cases are not enough to test the software. it may also be possible that we complete the software but is it same as per client Requirement Document.
 |
| Image source Google |
Comments
Post a Comment