1.1.1 Identify the context in which a new system is planned

Data gathering

In the initial planning stages there will be a process of investigation into the current system, this will involve the collection of data in a variety of ways (observation, interview, questionnaire, document collection)

Each of these has its advantages and disadvantages:

Requirements specification

Following that, there will be an analysis of the system in terms of how it works. The input process and outputs of the system will be identified and any problems associated with the current system

This allows the analyst to produce the requirements for the new system. It is leading the creation of the requirements specification. This is a tightly controlled process and is set out in the IEEE standard [1].

The requirements specification is a key document in a traditional life cycle as it drives all subsequent work. It is also the basis for project success and is often known as the success criteria.

This stage of the project identifies the context in which the new system is planned and will take into account considerations and justifications of the technologies for the new system and any other requirements in terms of success criteria.

Issues

There may be several issues relating to people using new systems. Clients and managers of the new system may need retraining and there will be a period of readjustment to a new system In its more extreme case this could lead to unemployment.

The context of the system in terms of existing technologies needs consideration. Will the current hardware and software infrastructure be able to deal with the new system in terms of network bandwidth, storage processing power or operating systems?

1.1.4 Compare the implementation of systems using a client’s hardware with hosting systems remotely.

When deciding whether to use local hardware or utilise the cloud several considerations must be made. Four main models of computing can be offered in the cloud.

• Infrastructure as a Service (IaaS)
• Platform as a Service (PaaS)
• Software as a Service (SaaS)
• Network as a Service (Naas)

Each of these will have its pros and cons in terms of hosting locally or in the cloud. Below are some general evaluations of these.

Cloud computing

Local computing

1.1.7 Suggest various types of testing.

Testing is conducted at several stages of the software development. The testing at each stage is dependent on the stage and may be simply debugging code, user acceptance testing or even beta testing when the initial design has been completed but is not yet on general release.

Two common types of testing in this phase are alpha testing and beta testing, but there are others. Here's an outline of these types, along with some additional testing stages that may be involved in releasing software:

Considerations Across Testing Types

Feedback Collection:

Across all stages, collecting feedback and systematically addressing issues is crucial. Tools and methods for gathering feedback can vary, including direct user reports, automated crash logs, and usability surveys.

Iterative Improvement:

Each stage of testing provides insights that should lead to improvements in the software. Multiple rounds of testing might be necessary, especially as issues are identified and fixed.

Environment and Tools:

The testing environment and tools can significantly impact the efficacy of testing. Realistic environments for beta and gamma testing can help identify issues that lab settings might not reveal.
By conducting these different types of testing throughout the software development process, organizations aim to release higher-quality, more reliable products that meet the needs and expectations of their end users.

Test Plans

Detailed test plans will be created and a typical test will contain:

• Details of what is being tested
• The test data to use
• What is expected to happen when the test is performed

When choosing the data we need to consider why we are testing: to see if the system works well and that the system does not break. To do this we can discern three types of test data

Normal

This is data that would be expected to be entered in the system. The system should accept it and process it.

Extreme

Extreme values are still normal but are at the edges of acceptable and are used to see that the system responds correctly to all normal data.

Abnormal

This is data that should not be accepted. The system should be able to handle this data without crashing or breaking. Validation checks and the handling of exceptions are often used to ensure this data is handled well

Example

Here is a program that calculates the grade based on a number input. 

//This psuedocode takes a grade percentage and returns the corresponding letter grade.
output "Enter your grade percentage: "
input gradePercentage
if gradePercentage >= 90 then
    letterGrade = 'A'
else if gradePercentage >= 80 then
    letterGrade = 'B'
else if gradePercentage >= 70 then
    letterGrade = 'C'
else if gradePercentage >= 60 then
    letterGrade = 'D'
else if gradePercentage >= 0 then
    letterGrade = 'F'
else
    letterGrade = 'Invalid grade entered. Please enter a percentage between 0 and 100.'
end if

//Return the letter grade
output "Your letter grade is " , letterGrade

This program defines a function calculate_grade that takes a grade percentage as input and returns the corresponding letter grade. We can use this function to test different types of data:

Normal Data:

• Grade entered: 85

This is a valid percentage within the expected range (0-100). The program will return "B" as the letter grade.

Extreme Data:

• Grade entered: 99

This is a valid but extreme value. The program will still function correctly and return "A" as the letter grade.

Invalid Data:

• Grade entered: -10

This is an invalid percentage outside the expected range. The program should handle this by returning an error message indicating the invalid input.

By testing the program with normal, extreme, and invalid data, we can ensure it:

1. Calculates letter grades correctly for valid percentages.
2. Handles extreme values within the data range (0-100%).
3. Identifies and reports errors for invalid input (negative percentages).