Formal Specification
Wittchim (Feather-Stalking Performance) — Formal Specification
Australia · festival-ritual · Confidence: low
← Back to game pageImplementation Suitability
Suitability: adaptation-requiredFidelity: abstractedGap-fills: 1
# Wittchim — AI-optimized specification
## General idea
A judged performance-art contest where the content is dance and mime quality, not a decision problem — a spec can describe the procedure and its social prize-sharing custom, but there is no meaningful agent-decision problem here, consistent with this catalogue's treatment of other performance games (compare kandomarngutta).
## Formal model
- **Abstraction taken:** none is applied to the performance itself — the source documents a judged aesthetic display, not a game with legal-move structure.
- **Player/agent count:** 1 performer at a time, judged by the assembled group.
- **Turn model:** sequential — one performance at a time.
## Formal rules
R1. A performer performs `stalk_feather()`, a mimed sequence with no formal legal-move structure.
R2. Judges perform `score(performer)`, a subjective quality assessment. [gap-fill] no scoring rubric is given in the source beyond "judges decide on the best performance"; inventing one would misrepresent an aesthetic judgement as a formal metric.
R3. The performer with the highest judged score wins, per R2, and performs `share_prize()` — repeating the performance and presenting the feather to the other competitors.
## Edge cases
- **Tie at the end condition:** not applicable — resolved by judge decision, not a comparable score.
- **Player with no legal move:** not applicable — stalk_feather() is always available.
- **Repetition / cycles:** not applicable.
- **Simultaneous claims:** not applicable — one performer at a time.
- **Mid-round elimination and turn-order repair:** not applicable.
- **Resource exhaustion:** not applicable.
- **Termination guarantee:** not applicable in the sense of a game-tree bound; each session simply concludes once all performers have gone.
## State representation
Persistent: none beyond the roster of performers.
Per-performer: judged score, once assigned.
Information set: fully observable — performances are public.
## Action space
- `stalk_feather()` — precondition: actor is the active performer.
- `score(performer)` — precondition: actor is a judge.
- `share_prize()` — precondition: actor won.
## Playability notes
- **Single-player-vs-AI:** not applicable — there is no adversarial or cooperative decision structure to play against.
- **Group games (n>4):** scales trivially by adding more performers in sequence, which reflects the activity's nature, not a design reduction.
- **Physical-skill model:** not modelled — performance quality is not a mechanic this framework's dexterity model can meaningfully proxy without trivializing the art form.
- **What is lost:** everything of actual interest — the performer's individual stalking technique, the corroboree costume and props, and the prize-sharing ritual's social function — none of which a formal decision-problem spec can or should attempt to capture.
## Implementation hints
Do not force this into a decision-tree implementation. If an implementer wants a digital analogue, it belongs in an animation/choreography or judged-scoring showcase tool, not an agent-decision framework — there is no strategic content here.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Wittchim (Feather-Stalking Performance) (Australia). (n.d.). Games of Countries. https://gamesofcountries.com/games/au-australia/wittchim
@misc{gamesofcountries-au-australia-wittchim,
title = {Wittchim (Feather-Stalking Performance) (Australia)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/au-australia/wittchim}
}