tl;dr: It is best to approach this course like standalone textbook rather than a MOOC. There is a lot of great information but zero engagement from the course staff. Success will be based on the learner’s discipline and self-motivation.
I started this course a few months ago because I felt it would present the opportunity to learn how analytics are applied to big data. As I understand it, this is similar to the on-campus course at Indiana University and the professor made the lectures freely available online. There was no option to receive a certificate for completion; the only reward for completing the course would be knowledge. I gave it my level best effort to complete the course ended up dropping it a few months later after only getting to the 6th unit. Here are my thoughts:
- The instructor presents a lot of great material. Most popular literature and blog posts about big data use abstract terms or focus on the tools and buzzwords: Python vs R, Hadoop, and distributed processing. This course excels at going past the hype and talks about how things work and the challenges involved. Prof. Fox’s expertise is readily apparent. The professor also aggregates information from many leading sources to give comprehensive lectures. After all, this is a full semester, graduate level course at IU.
I found the video lectures to be poorly organized. The course is presented using the sage-on-the-stage method where the instructor uses a webcam and slide show. In an attempt to keep the learner engaged, the lectures are broken into 10-15 minute segments which the instructor feels is the length of the student’s attention span. Where this falls short is that the lecture videos aren’t self-contained; they are merely a continuous lecture split into multiple videos. Arranging the course this way makes it difficult to view the videos without dedicating enough time to complete the full lecture. I found I was often going back to watch the last few minutes of the prior video to know where the current one was starting from.
The course doesn’t feel like a final product. It’s well known that presentation matters but there is definitely room for improvement in this low budget production. The instructor regularly coughs throughout the lectures but there was no effort into editing or re-recording the videos. While I can’t fault him for having a cold, it is quite distracting and there’s really no excuse for it in recorded lectures. This is the equivalent of a textbook have misspelled words and leaving the corrections in the margins when it goes to print. I feel like the professor wasn’t willing to put in the extra effort to give a polished final product and was left wondering what other areas weren’t fully developed.
The course materials were not engaging. The lectures do not incorporate exercises or demonstrations to get the learner involved in the material. There are projects midway through the course but you will need to make a pretty big commitment just to get there. his style is in complete contrast to the interactive style used in Andrew Ng’s Machine Learning course on Coursera or MIT’s Analytics Edge on edX. Both are phenomenal courses which I highly recommend. The key difference between these great courses and Big Data Applications and Analytics is that they include regular quizzes and practice exercises along the way to solidify understanding.
The discussion group on Google+ is not actively moderated. There was more spam posts than course discussion. This is probably the biggest fault of the course is not have a moderator to clean the discussion boards and to lead the discussion. In my experience with other MOOCs, the majority of learning happens outside of the lectures. An active discussion board could have enriched the experience and made up for the other shortfalls in the course. Finally, at last look, there were only 148 members to the discussion group. This hardly qualifies the course as a Massive Open Online Course (MOOC).
To summarize, I think the course is best approached like a standalone textbook rather than an interactive MOOC. Know that there will not be a strong network of peer learners to lean on for discussions or Q&A. There is a lot of outstanding material but you will have to work for it.