Question:
(a)
Consider that every software failure can be classified as either
• Syntactic,
• Semantic,
• System crash.
Suppose additionally that every system crash failure is more critical than every semantic failure, which in turn is more critical than every syntactic failure.
(i) Describe two different measures of the attribute criticality of software failures.
(ii) What is the scale type of each of the measures? Define the admissible transformation between the measures to justify your answer.
(iii) Validate theoretically your measures.
(iv) Consider you are evaluating two set of failures categorized according to their critically as follows:
Set 1: system crash, syntactic, system crash, syntactic, semantic, system crash, system crash
Set 2: semantic, system crash, syntactic, semantic, syntactic, semantic
You would like to know which of the two sets has the higher central tendency of criticality. Measure the criticality for the entities in both sets, collect the resulting measurement data and apply meaningful statistics. Validate your results using the notion of meaningfulness.
(i) Explain briefly the idea behind Albrecht's function points measure.
(ii) List two main applications of function points and compare function points with the lines-of-code measure.
(c) Consider the following specification of a simple system for collecting student course marks or grades.
The course-marks system enables lecturers to enter student marks for a pre-defined set of courses and students on those courses. Thus, marks can be updated but the lecturers cannot change the basic course information, as the course lists are the responsibility of the system administrator. The system is menu-driven with the lecturer selecting from a choice of courses and then a choice of operations. The operations include:
- enter coursework marks
- enter exam marks
- compute averages
- produce letter grades
- display information (to screen or printer)
- The information displayed is always a list of the students together with all the known marks, grades and averages.
Compute the number of function points in this system stating carefully the assumptions you are making.