I do not believe the two perspectives are mutually exclusive, as Wayne Hodgins as noted many times, but there are distinct differences in these approaches to object-oriented instructional design. If we have to assign labels (of course we do, right?), I fall more on the side of the by-hand approach to LOs. I see a place in the future for greater automation of some tasks but near-term there are many limitations. My fear in most of the current research on automated systems approaches is that we tend to lose sight of the most important aspect of learning – the learners. Some of what I saw was excellent for teaching skill- or procedure-based learning such as algebra or computer programming but I would be less comfortable teaching interpretive literature, philosophy, or interpersonal selling skills through such adaptive techniques. (A brief footnote here, there are some fantastic simulations being built that do provide some ability to begin to address broader learning goals but these types of technologies were not the focus of the discussion in Montreal).

Another divisive issue is a deceptively simple one: language. Instructional designers and computer science folks just talk different. Sometimes, I think we are trying to do the same things but much gets lost in weak attempts at translation. My simple talk followed several brilliant presentations outlining systems and standards that are groundbreaking and need to be shared. But, as Joe Scandura and I discussed at dinner after the session, we need more communication between those who are making advances in systems and those making advances in instruction and learning.

An update on thoughts from ISPI in Vancouver will appear soon!