HS23 – Software Evolution and Developer Productivity Seminar

Overview

Software requires constant evolution due to changing customer needs, bugs that have to be fixed, or changes in the environment. Lehman’s first law of software evolution states that a software system must be continuously adapted, or it becomes less and less useful. The constant change poses many challenges, and both researchers and practitioners have recognized the importance of studying and supporting software evolution and the humans involved in the process. In this seminar, we will cover some of the most relevant studies, approaches, and techniques that researchers have looked at in this context.

In this seminar, students will work on a specific research topic in software evolution and developer productivity. By the end of the course, students will be asked to deliver a research proposal. Furthermore, the students will take part in a simulated peer-reviewing process for the proposals. Finally, they will share as a presentation the results of their work and discuss them with the class.

For this seminar, you will be asked to read, analyze and discuss research papers, as well as work on and write a scientific report about a proposed topic in software engineering.

Grading Scheme

This course will be a project and paper-based course. The research proposal is the primary artifact of the course; the outcome of all projects will be a research proposal. To pass this course, you must pass the research proposal (obtain at least 50%) and must obtain a 50% overall mark that is based on the following:

• 75%: Proposal, including continuous progress, meetings, write-ups (including final proposal and intermediate report), and presentation;
• 25%: Readings, including active reading and collective annotations and discussions (Perusall), response papers, class participation, leading discussion (in-person class attendance is mandatory), and peer reviews of project reports.

Prior Knowledge

Content of software engineering and programming.

Assessment

• Active reading of assigned papers, (collective) annotating and discussions (Perusall), response papers, and active class participation;
• Presentation and moderation of a discussion topic;
• Written proposal, scientific report, and oral presentations and discussions for a topic in software engineering;
• Feedback on reports written by others (peer review).

Learning Objectives

At the end of this course, students should:

• Have gained a deeper and broader understanding and knowledge of Software Engineering research by reading, analyzing, reflecting and discussing current and classic literature;
• Be able to identify and discuss research problems and research questions as well as identify relevant related work;
• Be able to write a research proposal in the form of a scientific report, and present and discuss ideas on an advanced topic in Software Engineering research with a focus on software evolution and developer productivity;
• Be able to provide constructive feedback on a research proposal.

Seminar Organization

Kick-Off & Beginning

At the beginning of the seminar, students will be assigned a topic, a paper (based on preferences), and a partner to work on this topic.

Response Paper & Discussion Phase: For each lecture / discussion session of Topics 1-5:

• Read the assigned papers on Perusall and collectively add annotations (comments and questions).
• For teams leading a discussion: Each of the two teams assigned to a paper will present their paper and another relevant related paper they identified on the topic to the rest of the class (max. 15mins). Make sure to put both papers in context of each other and don’t just summarize one after the other. After the presentation, moderate the discussion (25 mins), make sure to prepare for it. The assigned teams also have to make sure to have read the other paper on the topic.
• All other students: Identify one additional paper related and relevant with respect to the topic and the given two papers; write a short response paper (max. 200 words) of the additional paper. Do not just summarize it but provide your own perspective, thoughts on it. Name the short response LASTNAME_FIRSTNAME_WEEK[01-05].pdf.
• Note, the additional paper has to be a full paper (i.e. >=9 pages) and from a top tier conference (e.g. ICSE, FSE, CHI) or journal (e.g. TSE, TOSEM, ToCHI).

Research Proposal Phase:

• Each team has to draft a research proposal in the area of their topic and present their draft. The draft should be a maximum of 3 pages (+ pages for references) and should have a motivation, related work section showing the relevance of the research idea/question, and a section on the research question and approach. Note, the major focus of the draft should be on the related work overview and it should not just be an enumeration or summary of other papers in the area.
• Until the end of the seminar, every team then refines their research proposal, writing a more complete related work section, specifying in more detail how they would go about addressing/answering the research question they pose (study method / study design), and what the motivation is. The final research proposal should be maximum 6 pages (+ pages for references). Again, the main focus is to have a good review of related work.
• At the end of the course, the proposals are submitted and peer-reviewed by other students (in EasyChair).
• At the last session the proposals are presented by the respective teams and the proposals will be peer evaluated by the reviewers.

Schedule and deadlines

Topics

Note that we’re using Perusall to read, reflect on and discuss on the papers below.