Abstract:

Real-time collaboration tools have become commonplace in the field of software engineering. Especially IT startups with distributed teams around the globe enormously rely on tools for agile development practices such as “Extreme Programming” (XP), which requires a close collaboration between co-workers even over vast geographical distances. The field of distance learning, especially in subjects which require continuous practice and assessments such as software development, show a similar setting. For this reason, the use of real-time collaboration tools for training future software developers seems a logical conclusion. Yet the differences in social dynamics and the overall structure between a typical work-place environment and an online lecture merit a closer look at the suitability of these applications for an (online) classroom.

In order to assess the feasibility of real-time collaboration tools in education, we performed several tasks with first-year undergraduate students. Depending on the complexity of the exercises different collaboration tools were used. "Etherpad Lite" was used to work on simple programs or code snippets. For more complex tasks "Saros" was introduced in class since it integrates into already existing software development environments such as "Eclipse"

Students were given the opportunity to take part in interactive online programming-sessions which were supervised by an instructor. After a short theoretical introduction and explanation of the programming tasks, the students completed the exercise collaboratively. During the whole activity (around 45 minutes), the lecturer supported the students by providing valuable feedback, both verbally and in form of ad-hoc modifications and corrections to the student's code.

In the short follow-up questionnaire on the collaborative task the students reacted generally positive to this new way of teaching. Particularly the immediate feedback from the instructor was seen as a great benefit since it prevented students from making further mistakes and further gave the participants an opportunity to ask follow-up questions. However, students criticised the additional complexity introduced by yet another E-learning solution. Especially the workflow of joining a live coding session with the tool "Saros" was found to be quite unintuitive.

In conclusion, students participated eagerly in the activities and thus they were more motivated to practice their skills during the online exercises. An analysis of the collected collaboration data shows that the majority of the written code, however, was written by just a couple of students Nonetheless, nearly all students were actively involved in editing at least some parts of the code, which contributed to the overall solution of the exercise in question.

Despite the generally positive reception and feedback from the students, it has also become evident that a customization of the tools is needed to better integrate them into our already existing E-Learning infrastructure making live-coding sessions effortlessly possible for students. However, a separate set of user-accounts and manual steps to join the session are currently necessary which impedes the learner’s experience. Additional limitations arise due to the fact that these tools are designed for pairs or small groups and therefore even simple tasks such as identifying changes by single users can become difficult and more complex due to a greater number of participants.

Keywords:

e-learning, distance learning, online collaboration, software engineering education, live editing.