Paper Review: Producing and Delivering a MOOC on Pattern-Oriented Software Architecture for Concurrent and Networked Software

Schmidt, D. C., & McCormick, Z. (2013). Producing and Delivering a Coursera MOOC on Pattern-Oriented Software Architecture for Concurrent and Networked Software. SPLASH ’13, October 26–31, 2013, Indianapolis, Indiana, USA. Retrieved from https://my.vanderbilt.edu/zachmccormick/files/2013/09/posa.pdf

Summary:
Two faculty members from Vanderbilt University produced and delivered a Massive Open Online Course (MOOC) on “Pattern-Oriented Software Architecture for Concurrent and Networked Software” (known as POSA MOOC) in Spring of 2013. This paper descries their experience.
Launched on 4th March 2013, POSA MOOC ran for ten weeks and had an enrolment of more than 31,000 students. Its section zero provided course overview with an hour of introductory video. Section one provided an introduction to concurrent and networked software using 3.5 hours of videos. Section two provided an introduction to patterns and frameworks with 6 hours of videos. Section three contained 6 hours of videos and focused on how to develop concurrent and networked software by applying patterns and frameworks and grouping patterns into pattern languages. The final section, which was an option one, provided a case study of “Gang of Four” Patterns with 3.5 hours of background videos on object-oriented design and patterns. Students could engage in one of two tracks. Normal track was meant for students who had time/interest in taking the auto-graded quizzes and final exam, but who did not have time/interest to complete the peer-graded short essay questions and peer-graded programming assignments. Students in normal track received a Statement of Accomplishment. Students in Distinction track received a Statement of Accomplishment with Distinction by scoring more than 70% score. The score was based on the weekly quizzes (35% of the final grade), peer-graded short essays and peer-graded assignments (55% of the final grade) and a final exam (10% of the final grade).

Authors have noted following observations and lessons learned:

• An enormous amount of time was needed to prepare the content prior to MOOC launch. Work involved filming high quality video lectures and creating student assessment mechanisms.
• An even larger amount of time is needed to manage a MOOC after launch. Authors spent 40+ hours per week on discussion forum largely to accelerate and amplify the learning process, to dispel common misconception, to build good will and to reward constructive student participation.
• The POSA MOOC student diversity was both challenging and rewarding. Challenged included much greater level of scepticism from experienced students, increased workload to fill knowledge gaps, generalizing from limited perspectives impedes learning and overly narrow focus on programming. Rewarding aspects included highly stimulating discussions with expert software developers and greatly improved course structure and content.
• Assessing student performance in a “design-oriented” MOOC is harder than in “fact-oriented” MOOCs.
• The Coursera platform, which was used to run this POSA MOOC, is a work-in-progress.
• Innovations helped make the POSA MOOC more like a “real” course. The innovations included virtual office hours using Google Hangout and YouTube channel, crowd-sourced programming assignments and grading calculator.
• Interpret MOOC enrolment statistics carefully. Out of 31,000+ enrolments, ~20,000 ever logged in and participated in some way. Only ~1600 of these participants received some form of statement of accomplishment.
• MOOCs can enhance student-centred learning opportunities by increasing asynchrony in courses and by providing location-agnostic learning.

Authors have mentioned following benefits to faculty:

• Significantly better on-campus courses
• Fostering global life-long learning communities that connect students who possess a range of experience

Authors have mentioned following benefits to Vanderbilt:

• Expanding the brand value of a Vanderbilt education
• Better opportunities for engagement with alumni and prospective students.

Besides these benefits, authors have also observed following drawbacks based on their experience:

• Potential for “deskilling” education and educators
• MOOCs require substantial institutional investment and the payoff isn’t clear (yet)
• Detecting and dealing with plagiarism is tedious

Based on their experience, authors are planning to expand their offering with an intentionally-coordinated, trans-institution sequence of MOOCs that focus on patterns and frameworks for mobile device programming. It would showcase how intentionally-coordinated MOOCs can create life-long learning communities that (1) cross-cut traditional institutional/disciplinary boundaries and (2) would not be feasible without the MOOC paradigm and MOOC platforms like Coursera.

Assessment:
This paper is an experience-sharing paper, which documents observations and lessons learned along with commentary on pros and cons of MOOCs. It has also identified a research gap regarding correlations between completion rate and various diversity related factors (such as broadband penetration, English proficiency, age, and types of software jobs available in various countries).

Reflection:
Authors have expressed fear that administrators at some cash-strapped institutions would diminish the quality of higher education via wanton replacement of experienced faculty with MOOCs and inexpensive lecturers. While this fear is very much valid, it may also be looked at from different perspective. Many cash-stripped institutes, particularly in developing countries can actually enhance quality of their delivery of education by combining MOOCs produced and delivered by top-notch educators with in-person mentoring by local inexpensive lecturers. Obviously it requires paradigm change where in the responsibility of such lecturers would change from that of teaching in classroom to facilitating learning experience of students through one-on-one or group mentoring sessions.