I guess you’re saying:

* Learn by doing activities
* You will need some content (explanations / examples / tools / etc) to help you do the activity (so you know what / why / how)
* You’ll also need experts who can help you
* When you do an activity you should produce something (related to your goal for learning in the first place, a report, a web site, a project plan, etc)
* It’s a good idea to check if the product was a good one or not
* As the designer of this learning experience (and for ‘stakeholders’) you should check if the product and the parts that make it up could be improved

I think if you take a practical activity (like learning soccer) and test if this would work and then apply it to knowledge work that’s a good way to check if it will work. This is because when dealing with learning for knowledge work (or comparing it with learning at school which is sometimes learning for the sake of learning) it’s easy to get stuck on the treadmill of knowledge acquisition (part of the process) instead of getting better at doing stuff (goal).

One reaction I had has to do with the value of context–or application–to really be able to evaluate any learning model. That is, a more formal and academic learning environment would need a far tighter relationship between assessment and learning activities. Certain socioeconomic learner groups would need more diverse forms of community-interaction (where “mentoring” itself might be inadequate), and so on. Maybe what I’m getting at is that there are few conditions that are universal, and so any learning model must be inherently flexible. Adaptable. Truly learner-centered so that a learner or community can show ownership over that process, molding it with “human hands.” The model must also be as flexible or diverse as the needs for learning to begin with. When I develop learning models–or rather, sketch them out for fun–I try to start there: reasons for learning, because that changes everything.

I am going to think more about this framework. Very cool thinking!