Scott Wilson , Paul Sharples, Dai Griffiths and Kris Popat have an article up on their work embedding Wave-enabled widgets into Moodle using Wookie. (Try saying that ten times fast.) What they envision is very similar in a lot of ways to what my former SUNY colleagues and I were thinking about when we proposed a Learning Management Operating System. Of course, that was 2005, so we were thinking about portlets rather than widgets. The lower barrier to entry and client-side nature of widgets are game changers.
Anyway, it’s a very good piece, well worth reading. A couple of points jumped out at me. First:
Rather than just being a like-for-like replacement, the Widgets update in real-time without any page refresh, and this affects user behavior. Rather than clicking links to launch tools or to view content, the Widgets encourage more of a “monitoring” mode of operation, with users navigating to a course page, then leaving it open in the background, occasionally bringing it into focus to see if any new conversations were happening in the chat widget, or the voting results had changed.
If this finding bears out with further research, it could be a fairly big deal. For one thing, it has the potential of making the learning environment a lot stickier and more Facebook-like (in a good way). At the same time, it should force some pretty substantial refactoring of core LMS tools, which simply aren’t designed for monitoring. The two models are likely to clash.
Here’s another interesting—and problematic—bit:
Another consequence of Widgets is that far less tracking information is available at a micro-level, as interactions with Widgets are not made available to the VLE. This implies that a new model for tracking will need to be developed for such systems; this may be a useful opportunity to reconsider tracking in more sophisticated terms than page views and hits. For example, it may be useful to separate out measures of attention, using something like APML[12], and measures of user effort, using something like User Labor Markup Language (ULML[13]).
The thing is, we’re just now beginning to develop models where we can identify at-risk students and, more importantly, help them self-identify and self-remediate based on the data from the learning environment. The fact that this trend is in direct tension with the whole Web-2.0-in-the-learning-environment trend is underappreciated. We need to solve this problem.
