Felt is a simple story sifting and simulation engine for emergent narrative play experiences. It provides narrative system developers with tools for defining story sifting patterns that match narratively potent sequences of events, and for building simulations in which characters make use of story sifting to reason about the world.
For example, here's a Felt sifting pattern that captures a "violation of hospitality" microstory, in which a traveling character enters a town, is shown hospitality by a resident of this town, and then experiences harm at the hands of this same town resident character:
; match a sequence of three events...
(eventSequence ?e1 ?e2 ?e3)
; ...in which the first event involves a guest entering town...
[?e1 eventType enterTown] [?e1 actor ?guest]
; ...the second involves a resident of the town showing them hospitality...
[?e2 eventType showHospitality] [?e2 actor ?host] [?e2 target ?guest]
; ...the third involves the same resident harming the guest...
[?e3 tag harm] [?e3 actor ?host] [?e3 target ?guest]
; ...and in which the guest didn't leave town between the first and last "anchor" events
(not-join [?eMid]
(eventSequence ?e1 ?eMid ?e3)
[?eMid eventType leaveTown] [?eMid actor ?guest])By providing emergent narrative systems with the capacity to identify and extract these kinds of evocative microstories from the "raw material" of character-centric narrative simulation, we hope to improve the ability of these systems to understand the emergent stories they're creating—and, consequently, the effectiveness of these systems at collaborative storytelling with a human interactor.
This repository contains a provisional standalone version of Felt. So far, active Felt development has mostly taken place in a series of repositories for specific projects that make use of Felt, including Why Are We Like This? and Diarytown. For examples of how to define Felt sifting patterns or simulation domains, it's probably best to start with one of these repositories.
In addition to the felt.js file included here, you'll also need some version of DataScript included in your project to work with Felt.
For more information about Felt, see the following publication:
- Felt: A Simple Story Sifter. Max Kreminski, Melanie Dickinson, and Noah Wardrip-Fruin. International Conference on Interactive Digital Storytelling (ICIDS), 2019.