Abstract:

Software engineering (SWE) has rapidly evolved into a dominant industry. Yet, the education has often neglected the importance of managerial and organizational aspects, and failed to incorporate them together with technology into a coherent and relevant laboratory project. Over time much has been written on how to organize the SWE laboratory, with arguments flowing for and against additional complexity.

We believe that SWE is a capstone course in any computer science curriculum. Unfortunately, this may still not be widely recognized. The modern course on SWE requires that the student is familiar with many aspects of computing (e.g., networking, Internet technologies, programming languages, databases, HCI, computer graphics, and project management). Depending on the year of academic study these expectations could be unrealistic, especially if that is a 2nd year course.

The theoretical component of any SWE course should be structured and presented such that the student eventually gets comprehensive understanding of the discipline, with technical and nontechnical topics logically and causally interwoven. The practical component (i.e., the laboratory project) should refrain from unnatural simplifications of roles and work products. The student should experience that, at times, solutions are to be found, not prescribed by the book.

The uneven distribution of knowledge or even lack of it must not become an obstacle to the student. However, we should not scale down the curriculum but incorporate, if and when necessary, additional elements that will allow all students to make a uniform progress. SWE is a complex discipline that primarily aims at answering the question of how to build SW systems. But, it must also be decided what tools to use, learned how to select tools and technologies, and their character and impact must be understood.

This problem is far-reaching, and should be dealt with accordingly. In this paper, we elaborate on our ideas to address these issues.

Keywords:

software engineering, student project, object oriented.