Software Engineering Assignment

Software Engineering (ISTN 713) Software Estimation (Chapters 25 & 26)

1. Describe the difference between process and project metrics? 2. Why should some software metrics be kept ‘private’? Provide an example of a metric that should be private. Provide an example of a metric that should be public.

3. Why is it important for software developers to make use of measurement to guide their work? 4. Identify and describe four useful indicators of software quality that should have measures defined and monitored by the software project team. 5. What is an indirect measure and why are such measures common in software metrics work? Provide 2 examples of direct and indirect measures. 6. Present an argument against the lines of code as a measure for software productivity.

7. Discuss metrics that are suitable for OO projects? Can use cases be used as a normalization measure? Explain your answer. 8. Discuss 2 methods of performing software sizing using the problem based estimation approach. How does this differ from the process based estimation approach? Chapter 25 Problems: 25.7; 25.10; 25.11; 25.12; 25.13

Chapter 26 Problems:

26.5

Use the COCOMO ll model to estimate the effort required to build software for a simple ATM that produces 12 screens, 10 reports and will require approximately 80 3GL software components. Assume average complexity weights for input and output objects types, difficult complexity weight for 3GL components and average developer/environment maturity. Use the application composition model with object points.

26.12 Re-compute the expected values noted for the decision tree in Figure 26.8 assuming every branch has a 50-50 probability. Would this change your final decision?

1

Additional Problem on COCOMO Model:

You are in charge of developing two 35-KDSI organic mode products. Both are nominal in every respect except that product P1 has extra high complexity, and product P2 has extra low complexity. To develop the products you...