Margot Postema

Including practical software evolution in software engineering education

Software engineering typically requires more effort in maintenance than in development time. As such, software engineering education needs to actively include software evolution. Teaching software evolution to undergraduate students usually includes the theoretical aspects, but does not normally include the actual implementation. This paper describes the practice of teaching software evolution to undergraduate computing students at Monash University. It demonstrates how the four maintenance activities: corrective, adaptive, perfective and preventative, can be included into the practical component of a software engineering course, providing students with a much more realistic view of software engineering.

Tool Support for Teaching the Personal Software Process

Software engineering education can be viewed as a challenging task. Computer science students tend to focus on the programming aspects of a project, and take a “hacking approach” to completing a project, rather than viewing the process. Our course material includes teaching the Personal Software Process (PSP). Students are required to produce defect and effort metrics, as well as project summary reports. Tools to assist information recording and production of reports are difficult for students to access. The cost in terms of financial and learning time is usually too high at the educational level to justify using commercial tools that are available for the professional software engineer. We have developed a tool—Personal Assistant for Software Engineers—to aid students in the learning of the PSP. The tool has been successfully used in four subjects for a semester. Following feedback and evaluation, it has been redesigned, and is currently in stage two. The tool has been useful for students, and has automated the previous manual process of producing time and defect reports, along with project summaries, thus allowing them to focus on the meaning of the information.

Teaching tools for software engineering education

This paper examines a set of teaching tools in the Software Engineering Practice course at Monash University. Analyzing various aspects of the course and student survey response to their value provides evaluation of the overall success of the tools. The paper demonstrates that a successful teaching program needs to combine a range of teaching tools to achieve its aims.

Improving student performance in software engineering practice

This paper looks at the improvements that can be achieved in the learning of software engineering practices when a proactive process of measurement and intervention based around adult learning principles is utilised. It also examines the many obstacles that can be placed in the way of student performance, both external and internal, in the current university learning environment. The paper is based upon the results of two and a half years of teaching software engineering practice to over 800 second year students in the School of Computer Science and Software Engineering. It also draws upon surveys of the students over that period. A key conclusion of the paper is that a course can be significantly affected by any one of a number of aspects and successful courses must ensure that all aspects of a course are satisfactory.

Teaching software evolution practice (poster session)

we store hash values that represent chunkg. Not only do we expect false positives because l~h functions can produce the same value for different chunks but also because the number of possible cbnnk.~ in Interuet-docomants onmumber the available mmaber of hash values. There are two ways to reduce index-space. We can either reduce the number ofchunk~ to be kept, which increases the chance of false negatives, or we can reduce the size of the hash value we calculate on each chunk, which increases the chance of false positives. False negatives are harder to handle because we have already missed potential documents. We propose a method that is able to ellm/nate false positives from a given set of documents. The comparison is completed in two phases. In the :first phase we define candiclate documents using the aforementioned methods and the second stage ellrn/nates false positives. Our algorithm for eliminating false positives uses a suffix tree built on the suspicious document to compare candidate documents and elim/nate accidental matches. Comparison of the chnnking methods and our algorithm are presented in this poster. Maintenance can be defined as the single most expensive activity in large software engineering projects, requiring 65% to 75% of total effort. Hence software engineering can be termed software evolution. The subject Software Engineering Practice (CSE2201) taught in the School of Computer Science and Software Engineering at Monash University is a second year core subject in an undergraduate degree program and comprises about 250 students per year. CSE2201 introduces software engineering concepts to students and expects students to view software engineering as an evolutionary process. Students are additionally introduced to and expected to implement the practical aspects of the Personal Software Process (PSP). We demonstrate the practice of teaching software evolution to undergraduate computing students in the Bachelor of Computing degree, and show how the four maintenance activities of corrective, adaptive, perfective and preventative can be included into the practical component of a software engineering course, providing students with a much more realistic view of software engineering. Expla-ations are needed to give feedback to students, as part of an Intelligent Tutoring System (ITS). A student submits an SQL query to the ITS as a solution to a question. When the query is incorrect the student receives an explanation, from the ITS, of how they can improve it so that it satisfies the task at hand. An expert enters an SQL …