An Empirical Strategy

May 10th, 2021

Here’s an odd idea that emerged in the Zoom discussion yesterday. I am here presenting a developed version of the brainstorm we got. 

We begin with a preliminary trial using six people: a Drama Manager and five Actors. The Drama Manager concocts an interesting dramatic situation rife with interpersonal conflict. The specification of the situation includes definitions of each of the characters, their backgrounds and desires. The author transmits these specifications as a written document to each of the Actors for them to study. 

The group meets in a text-only chat room where the DM selects one of the Actors to take the first action. The selected Actor writes a sentence or two describing the action he takes in the chat box. Next, the actor selects an emotional face to accompany the decision text. Finally, the Actor sends a private text to the DM explaining the reasons for his decision, presumably couched in terms that can be reasonably reflected in an algorithm. 

Now the DM selects another Actor to present a reaction to the first Actor’s action. The second Actor follows the same procedure as the first Actor, entering his overt action into the chat box along with a facial expression, and secretly explaining his motivation to the DM.

The DM goes through the cast, selecting other Actors to react, although there is no requirement that the DM select them in any order, and in fact the DM might wish to restrict the action to just two Actors at first. 

This process continues until the story reaches a satisfactory ending or the Actors tire.

The DM repeats this process with a different group of Actors, using the same scenario as previously. After perhaps a dozen such trials, we will have a large set of actions, reactions, and motivations. We must next combine similar actions.