PCon3 is presented here: http://www.indie-rpgs.com/forum/index.php?topic=17387.0

One of the advantages of examining the whole of play content is that active content can be classified in multiple ways, without dependence on the imagined or real nature of the content.

The easiest way to classify play content is essentially encyclopedic. While this method is not wholly reliable, we can often feel quite safe making claims that if active content typically consists of matters dealing with elves and their fictional culture then the players are learning about elves and their culture. And more so, if play is consistently of this type, we could also claim that the players are realizing this as a goal.

However this alone does not suggest what about elven culture is being learned, while the matter may be deduced, the form is more complex. Are the players learning specific details about elven culture? Are they learning how to better mimic an elf within this culture? Are they attempting to discover what these elves mean personally and to each other? It could be any of these, or some combination.

This leads to a more difficult manner of classification, the form of the learning. Loosely this can be reduced to three categories:

• Cultural - learning associated with actual people, whether present or abstracted. This includes empathic and emotional learning, learning about social structures, learning about values and beliefs, and general social reinforcement commonly found in play. In short this learning necessitates the understanding of an actual person, possibly oneself, on some level.
• Declarative - factual, although not necessarily true or real facts, learning. This is the common idea of learning, the direct acquisition of knowledge, rather than practice.
• Procedural - active, although not necessarily physical, learning. This sort of learning is closely associated with practicing tasks, usually with performance reinforcement.

Using these two approaches we can fruitfully classify active play content by it's topical matter and it's form of learning, at least on the theoretical level. The difficulty lies in extracting the form of learning from active play content. In order to do this we must consider the structure of views for each of these forms of learning.

The simplest view is that of Procedural learning. In this case the view is focused on the procedure being practiced, as well as any feedback. This is a cyclic process, but anything outside of that process is scarcely incorporated within the view. Thus a cyclic structure of play content suggests that the form of learning is procedural (in much the same way as a surfeit of content about elven culture suggests that the topical matter is elven culture). Declarative is more complex, but loosely speaking since factual learning is a process of building connections, declarative play patterns tend towards random walks, gaining complexity by unconstrained movement, but keeping it manageable by locality constraints (successive active content depends on prior active content in a strong way). Lastly, Cultural content is the most complex. The view in this case focuses on the relations of the people involved (and some number of others, whether specific or generalized), but this network is not arbitrarily extensible, as the walk in declarative learning. Likewise this restriction permits cultural learning to be more sporadic in terms of relations.

Now in each case this analysis assumes that only a single matter and form is being learned. In practice views can be much more complicated. However as a simple model of views it is useful to treat them as a small set of atomic pairs of topical matter and form of learning.

    - Mendel Schmiedekamp