Adventures in Learning — Part 11

What Has Gone Before

In Parts 1-5, I discussed the Team-Based Learning methodology (TBL), course design for an undergraduate INF 202 Introduction to Data and Databases class, and drew parallels between TBL and roleplaying games (RPGs). Part 6-10 reviewed the first several weeks of classes.

Where We Are Going

This entry branches off in a more philosophical direction. An important component of good roleplaying game execution is creating a sense (and a reality) of player autonomy. The players’ choices have to be meaningful and they have to be true choices. The notions of freedom and autonomy are reflected in a book I recently finished (kudos to my bro-in-law, James Morss, for gifting it to me for the holidays).

Future Perfect

In Future Perfect (2012), Steven Johnson champions something he calls “peer progressivism.” Setting up a contrast, he starts with the “Legrand Star” – a concept derived from the French approach to railroad planning during the mid-1800s.


( Johnson writes: “Think of the Legrand Star as a kind of shorthand for the ways that states like to organize the world. They concentrate power in a central location; they make the peripheries, the edges of the network, feeder systems for the main core; they simplify; they favor broad strokes over unpredictable swerves; they prefer master planners over local knowledge. They look best from above.” (6)

In contrast, Johnson describes a Baran network, based on an alternative approach to military communications detailed by RAND Researcher Paul Baran during the Cold War. This “fishnet” model “didn’t involve a central core but instead relied on a dense network of connections to shuffle information across the country.” (12)


( Baran also conceived of breaking up messages and reassembling them at the reception point (another researcher labeled this “packet switching”, the name that stuck). This distributed network “proved far more resilient” than a centralized, hierarchical one.  As you’ve probably guessed, this communications system became the Internet, a means of sharing information on a global scale.

Johnson generalizes from the Baran web and other successful “distributed network” enterprises to flesh out a political, social, and organizational approach based on peer networks. He amalgamates the concepts of decentralized control, local expertise, dense and diverse participation, open exchange of ideas, constant and evolving value assignment to promote positive deviations and discourage negative ones, and layered platforms of collaboration (new platforms built on top of earlier ones) into something he calls “peer progressivism”.

Johnson goes on to discuss peer progressivism ideals and experiments in the context of communities, journalism, technology, incentives, governance, and corporations. Only two sections touch on education: a review of the federal Race to the Top program (a prize-backed competition for secondary schools and their communities) and public school-wide performance incentives (not teacher-specific). Neither of these relates to undergraduate teaching, and indeed the whole concept may collapse when applied to a single classroom. Still, that won’t stop me from taking a stab.

Distributed Networks and TBL

In general, TBL moves the classroom away from the “Legrand Star” of the instructor dolling out the knowledge, being the font of all wisdom, and standing as the keeper of all grades. Team activities and decision making, facilitated discussion between the teams, and peer evaluations (for grades) decentralize the teaching process. For that reason, Johnson’s peer network concept is clearly applicable. Other aspects are not so apparent and a number of intriguing suggestions arise.

I’ve often considered how to further decentralize control in the classroom. What I’d like to see is greater student control over what goes on in the classroom, how the lesson plays out, and where to go next with the learning. As in a good roleplaying game session, the participants should feel they have significant control over where the storyline takes them. They are free to explore where they will, assemble clues in a variety of shapes and patterns, and influence the world in a variety of free-form and spontaneous ways. Even if I achieve the goal of creating multi-branching activities and scenarios, the overall (even if loose) structure of the lesson plan fences them in. Can I loosen the strings on how they learn? What would that look like?

One thought is to rely on “local expertise”. In the context of my class, that means first getting the students to step back from the individual content lessons and look at the course material as a whole, to think about what they are learning and how it relates to other possible data/database subjects (in particular, the ones they have no grasp or familiarity with), to consider what they could do differently in activities or content, and to experiment with what might (or might not) work. That means leaving enough time to go down false or ultimately unsuccessful paths. Heck, it means me going against basic instructor-protective instinct and allowing them to steer the class in a direction that I might consider a waste of time. We would need a discovery mechanism that revealed possible paths and a group-choice mechanism to determine which one to take. It would be exciting for me as instructor to design possible approaches depending on what the students viewed as an interesting problem or path. Such guidance would have to be adlibbed during class (as would be done during a roleplaying game session by the gamemaster) or worked up in the few days before the next class.

As if the above reconditioning for both students and instructor weren’t difficult enough, we could take the “local expertise” idea even further—to the level of pedagogy. That means getting the students to think about how they learn and what we can do as a class to make that process more efficient. I can imagine that such a process would be halting as they adjusted themselves to thinking more self-consciously about their processes. Asking them regularly to give their opinions about a class’s activities and how they might be improved is a start, but I’m not sure that would be sufficient. An ideal approach walks them through process analysis, gets them to brainstorm different formats, modes of discussion, and activities, focuses them toward determining mechanisms to measure the outcomes of different approaches, and encourages them to make reasoned decisions on what approach to try next. Creating a format for such an approach and internalizing it to the degree that we could employ it multiple times in one class would promote a highly iterative method of operating. Heck, we’re dreaming here, let’s go with the ultimate vision, right?

The open exchange of ideas would be crucial to setting up the above schema. That means getting the
students (and me) comfortable with criticism and the willing and supportive realization of wrong-headedness. At the same time, we’d need to inculcate a stubbornness to follow through with ideas that seem wrong-headed, particularly after a few initial failures.

The above operating procedures would also naturally require the development of a constant and evolving value assignment mechanism. Votes, prizes of some kind, community prestige, titles, status as an innovator are all things that could be used to promote and/or discourage, but we still need the measure. We need something like a “market” to judge value.

The final aspect of peer progressivism is dense and diverse participation. To the extent that we compare a single-source instructor to a 24-person class, we’ve got much more dense and diverse participation. That’s a built-in feature of TBL.  On the other hand, the scale that Johnson contemplates—the scale that makes the market and diversity of local expertise work—that’s not possible in a single class. Perhaps then the solution is to combine classes and draw on more students. Collaborating with students in a course covering a different (but related) Informatics subject could provide insight into the directions possible from any given point in the Data and Databases class. Likewise, the critiques could be flipped to aid in the non-Database class. Again, we need a mechanism to share ideas, preferably one that is simple, non-time-consuming, and convenient. A shared online forum could work if specific mandated assignments involved giving your impression of the other class’s current direction and possible future directions. Indeed, convincing the “non-database experts” in the other Informatics class of a certain direction could substitute for the valuing mechanism. Something to think about here (I’ll have more on inter-class collaboration in another entry).

Some novel, fascinating, and difficult-to-implement ideas here. Have to think about how to break down such a transformation into digestible chunks to begin experimenting. Heck, probably need to think more about whether the entire approach makes sense. Thus far, it seems to be worth exploring.

More to come,
M Alexander Jurkat


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s