On the same day that I set out to write about the importance of active community building and management in MOOC’s, Coursera rolls the addition of student profile pages. It’s a small step that I hope signals a dawning realization in the massive potential and need to facilitate and harness a growing community. The addition of profiles itself is nothing revolutionary, and in fact,it is something that I’ve always wondered why they neglected to implement long ago. It made little sense that students functioned as scattered masses and classes as disjointed entities. Having profiles is a very rudimentary first-step to addressing a growing need & problem of effectively leveraging the community. It provides the basic foundation, i.e. a continuous identity which you need for building any meaningful network . But, as said, I believe that it is only the most basic beginning a problem/solution that no one has quite figured out how to tackle yet—even the rest of the internet.
My thoughts on this post were triggered by this great article on the preliminary outcomes of the recent Coursera peer assessment experiment in the humanities. A big hurdle in the peer assessment of required essays has quickly emerged: many students were not fluent in English and produced broken work through Google Translate. My takeaway from this article though is not some solution to fix persistent grading problem or bridging the language gap (or perhaps it is, eventually & with several leaps of logic) but the self-serving potential in the growing organic community on these platforms. Garcia, a web developer in the article, suggested that perhaps the solution to the language problem would be to connect students that speak the same language with each other and do the assignments in their language.
But what if foreign students were allowed to write in their native language? “At the beginning of the course, you could [administer] a survey where you ask the person if they fluently speak another language, and then perhaps you could route assignments when you do peer review,” Garcia continues. “For example, I speak Spanish and French, so you could route those [essays] to me.”
By itself, this is a solution that could pose other philosophical problems to the MOOC experience (1*), but it made me realize that perhaps a viable solution to many of MOOC’s problems lies in the better management of their communities as a priority and not afterthought. The platform should proactively give students the right tools that would allow them to self-regulate and mimic a better classroom experience.
If you are as geeky as me about following the development of MOOC’s, you would probably be familiar with the distinction of MOOC’s by xMOOC’s and connectivist MOOC”s (2*). The xMOOC’s focuses on content mastery and pushing out cotent unidirectionally for consumption; Coursera and the likes tend to fall into this category, at least for now. The connectivist model emphasizes social learning and participation through a decentralized learning process. It would seem like I am trying to say that the connectivist model is better or more successful in tackling the community management aspect, but in fact I am not. The connectivist model is more visionary in that it understands that one of the most dynamic assets of an unbound open learning system is the people, but in execution there is a lot of same chaos in identifying, absorbing, and building upon meaningful contributions. At the end of the day it is still too many voices overflowing seemingly never-ending streams.
Why should MOOC’s care about community building? I think this is very much an investment of necessity and not some auxiliary benefit. I would even say that in order for MOOC’s to really be revolutionary, this is the key. Two of the greatest problems with MOOC’s are 1) information flow outside of the dictated content and 2) the emotional experience of connecting with others in your learning. Right now, both of these are addressed by messy and haphazard discussion threads on the forums. For example, in my attempt at Machine Learning this time around, I had to dig through the forum to find study groups based in San Francisco because I wanted to have some support in my study process. In order to do that, after I found the relevant threads, I had to join a Google+ group and a Facebook group and a Meetup group, all three of which are outside the Coursera platform. It made very little sense that suddenly I have to monitor 3 additional web presences on top of the ML course itself; ideally this would be something that is built-into Coursera whether as a proprietary platform or integrated with external services. A thoughtful system should reduce friction, increase efficiency, and facilitate the camaraderie of a real classroom experience.
So what would better community management look like? What does it entail in terms of both psychology and infrastructure ? We now venture into a territory that nobody has quite figured out yet the most effective design practices in facilitating mass communication and collaboration—it would likely entail many iterations & failed experiments in interaction design. But here are a few thoughts on the key goals to address (and some examples of web services that do that well):
1) Managing information: purposing a meaningful discussion
Without question the massive information flow is the hardest thing to tackle and it is a problem that every engagement-focused network is facing. The key is to build a meaningful conversation that allows you to stay focused, discover the important things, and excite you enough to contribute too. No one is doing this perfectly (and likely no one ever will), but here are some thoughts:
Setting up the proper framework & controls
- Much of how productive a discussion turns out is derivative of the initial design and controls of the discussion system. An effective framework identifies the right motives for that context and reinforces that through effective UX. Some of the most effective systems that I’ve seen is stack overflow (an all too frequently cited example I am sure), Wikipedia (likewise), and OpenIdeo. Stack Overflow implements an effective review and voting system based on your credibility points. Wikipedia has extensive guidelines and version/citation mechanisms around each article. And OpenIdeo has built an extremely thoughtful stage-by-stage concepting mechanism that provides guided discussion for each stage. All of these systems gave careful forethought to the larger product they are trying to build and what necessary constraints and hints need to be in place to shape the right discussion. What would this look like for a MOOC? I don’t know yet, I suspect it will be contingent upon the specific motives of each MOOC, with many of them sharing common problems. Maybe I’ll give this another go in another blog post.
Facilitating digestion of information
- Without doubt the “freeflowing” stream is broken, yet it is still the primary mechanism of information flow. To some extent this is mitigated by the voting system, but still not so much right now because there isn’t an overwhelming incentive to actively vote up and review something. Again, this is something that stack overflow does particularly well, in no small part due to their enthusiastic community that is able to utilize tools well-designed in the first place (they are one of the few places that truly use tags well). The answer to the question is probably some combination of effective search, topical tags, and community initiative to manually review and organize. Again, exactly what this would look like for MOOC’s is question for another day.
- The Coursera profile is a first step to tackling this issue because a participant can now start to build valuable identity on its platform. This is, overall, not as big of a problem for MOOC’s because people voluntarily chose to participate out of interest—but perhaps the question is how to sustain meaningful participation over time. I think the answer lies in enabling people to build up credibility and achievements on the platform, and this might involve some mechanism of badges or points for your participation.
2) Facilitating action: tangible engagement beyond words
To take a MOOC beyond a virtual experience would require a platform that facilitates people to build meaningful connections and not just passive consumption of content. I think this is actually something easier to fix than point one. The infrastructure around it is easier to build (and has been done before); the trick here is probably more about human attention devoted than infrastructure. I imagine there needs to be a team dedicated to helping meaningful connections happen.
Forging meaningful connections
- A system should be able to help its participants connect easily. Right now students have to resort to outside services to do that, but this is something that should not be difficult to fix. Imagine how valuable these MOOC’s could be if they could cultivate a transparent and accessible community of smart capable people, who are potential resources for furthering your learning or hiring? A side story: I started this blog at first on tumblr but quickly migrated to WordPress. It was because I realized WordPress had real people who will invest their time to make meaningful comments & connections with you rather than just <3 your post. I wanted true and substantial connections, and I came to the right community for it.
Translating into real-life interaction
- I think this is actually the easiest part of everything because physical events are naturally more tangible (and hence I would argue manageable) than virtual ones. I have already talked about setting up systems to facilitate geocentric meetups. One step further is to have active support to coordinate these meetups. A recent great example of this is the preparation for Udacity Global Meetup in which Udacity proactively procured a meetup space for one of their key cities, San Francisco.
* * *
I have meandered much and need to stop breaching my own preaching. Bottom line is I would love to see more attention shifted to community management and treat participants as resources to leverage and not just passive audiences. This would require a lot of prior planning and vision in UX, usability studies with real students, and ongoing community management rather than haphazard afterthought and manual pruning. It is perhaps a task that requires greater imagination than any new course to be built.
* * *
(1*) Language barrier or not, I can see that there is tremendous value in being able to connect with students from all over the world. Segregating students by languages could easily take away this magic, especially for minorities from impoverished areas that perhaps benefit the most out of it.
(2*) There is also a distinction proposed by Lisa Lane to separate them into Content-based, Task-based, and Network-based. But I think the latter two tend to share somewhat similar characteristics, at least in the examples proposed. DS 106, for eg, which is listed as a task-based MOOC, in fact very much derives its value out of sharing of work between students. The connectivist examples, such as the MOOCMOOC and the recent Edtech Startup 101 are actually very much task-based too, even though they tend to require more/some collaboration.