Polychess

A bridge to a different kind of game.

This is not a game to win. It is a short crossing to a realization. Standard economics keeps score on one board, and the moves that win there are taking moves on boards nobody is counting. Make a few. Watch option space open and close. There is no high score to chase here — only the one thing that ends the play, and what you learn when it does.

The finite player's scoreboard · Board 1 0 the single GDP-like number, the score of one board. It only ever rises. Pursuing it is what ends the play.

Move 0 · all four boards open · 100 of 100 cells reachable · the play continues

The deck

Hover a card to preview its Δω on each board — directional only, a sign and not a magnitude. Click to play it. Every card is a real configuration-change drawn from the worked moves, plus a few obvious generative ones. Notice which cards lift the scoreboard, and what they take to do it.

Move log

  1. No moves yet. The board starts fully open.

An honest note on the toy

This is a deliberately legible toy model of a deliberately illegible reality. Its whole job is to make a structure felt — that one physical action registers as a move on several boards at once, scored differently on each, with king-pieces you didn't know you had. It does not simulate, measure, or predict.

The cell-counts are illustrative, not measurements. Global option space is not computable — that is precisely why the framework evaluates moves locally and directionally rather than scoring states globally (option-space-measurability). Crushing real biosphere dynamics into a 5×5 grid is exactly the legibility-for-truth trade the framework warns about (legibility-truth-tradeoff): the grid is legible and therefore lossy. Hold the feeling; discard the numbers.

The real apparatus is the directional rule on R_living, worked on six real cases with their evidence and their ambiguities, at /applications/moves. This page is the four-resolution teaching device for that rule, made playable.

The game on the other side of the bridge

James Carse drew the distinction this page is built on. A finite game is played to win: it has an agreed end, winners and losers, and it stops when someone wins. An infinite game is played to keep the play going: it has no end-state to reach and no score to top, and its players act to bring more players in and keep the game alive.

Chess is finite. The scoreboard above is finite. The framework's objective is not.

“The point isn't to win it or finish it, but to keep it going and bring others in.”

That is the whole crossing. The boards were never a puzzle to solve or a score to maximise — they were a way to feel, in a few moves, that the instinct to win a board is exactly the instinct that empties another. The viable objective is the infinite game's only rule, stated in the framework's own terms: make no move that asymptotically ends the play for any class of life (viable-objective). Not maximise this number. First, do no harm.

This is set out as a proposition — The Infinite Game — and the lineage to Carse is traced at /lineage. The rule it implies runs on real cases at /applications/moves.