curated randomnessby Rahil Chamola
Share
Portable context

Take this page to your AI.

Nothing is sent automatically. Copy or download a clean Markdown packet, then choose what you share with your own agent.

Your context, your choice. Review the packet before giving it to any model or agent.

All projectsCreated by Rahil Chamola
Project case study / Shard

Shard

A story system built around truths the model cannot rewrite.

Shard sits between a scripted visual novel and an unbounded chatbot. It uses language models for the moment-to-moment scene, then gives identity, history, time, chance, and endings to deterministic systems that can be inspected.

State
Evolving
Role
Product and system design
Time span
March to June 2026
Evidence
Architecture documented
01 / The product tension

Structure without scripting. Freedom without chaos.

The original wager was to preserve durable consequence without pre-authoring every branch.

Authored branchingConsequence is durable.

Freedom is bounded by what someone wrote in advance.

Shard's wagerKeep consequence. Generate the moment.
Unbounded chatAlmost anything can be said.

The world has little reason to remember why it mattered.

V1 answered this tension by giving models specialized jobs and surrounding their output with typed governance. That was already a serious system. It was also still organized around repairing model output after too much truth had entered the model's jurisdiction.

02 / The redesign

The first v2 plan changed the cast, not the assumption.

A real redesign began only when the product promise and the source of truth changed together.

Evolution

Four decisions changed the centre of gravity.

  1. 01

    Design the system before touching the AI

    Why

    The game needed explicit types, world rules, and consequence handling before generation could be useful.

    What changed

    V1 split generation into model roles and surrounded their output with typed governance.

  2. 02

    A finite episode, but inherited machinery

    Why

    The early redesign changed the experience frame while retaining three model roles, validators, effects, and contracts.

    What changed

    It became evidence that a redesign can preserve the assumption that made the first system difficult.

  3. 03

    Optimize for a finite emotional story

    Why

    The desired outcome became a story that moves the player and leaves memorable characters, not a simulation that can continue forever.

    What changed

    Pacing, memory, privacy, and closure became architectural constraints.

  4. 04

    Move durable truth into an append-only ledger

    Why

    History and causality need one inspectable source rather than several mutable interpretations.

    What changed

    Relationships, knowledge, promises, and inventory can be derived as views over accepted events.

03 / The kernel

World file, ledger, clock, plan.

Four durable objects now carry the truths that narration is allowed to interpret but not overwrite.

System map

The v2 critical path

The model receives a constrained scene context. Deterministic code accepts valid events into one ledger, then derives the views the next scene needs.

Player intent enters deterministic scene code with the world file and finite plan. The scene model returns prose and proposed events. Code validates references, applies accepted events to the append-only ledger, and derives relationship, knowledge, promise, and inventory views. A separate asynchronous tick can read and extend wider world history.

Canonical vocabulary

Twelve kinds the ledger can keep.

These are current schema kinds, not a score. Each accepted event becomes part of the history from which later views are derived.

  1. 01scene
  2. 02move
  3. 03dialogue
  4. 04fact
  5. 05feeling
  6. 06item
  7. 07bond
  8. 08promise
  9. 09clock
  10. 10arrival
  11. 11departure
  12. 12consequence
Counting frame

Several counts, several definitions

Not a performance benchmark

The sources count different scopes. They show changing vocabulary and architectural boundaries, not a measured percentage improvement.

04 / Responsibility

Decide what the model is allowed to be wrong about.

Code, configuration, models, and evaluation own different questions and different failure modes.

Responsibility boundary

Who owns which truth

OwnerOwnsDoes not decide
Code
  • identifiers and reference validity
  • dice, time, ordering, and plan gates
  • accepted event application
It cannot decide whether a scene is moving or beautifully written.
Configuration
  • world canon and fixed points
  • difficulty, resistance, and permitted variation
  • story shape and authored constraints
It provides a possibility space, not the final moment-to-moment story.
Model
  • interpretation and scene construction
  • dialogue, texture, and proposed meaning
  • narrative choices inside current constraints
It does not own durable truth, dice, identity, or reference validity.
Evaluation
  • coherence and experience review
  • real-player observation
  • whether the story creates attachment and desire to return
A fluent model or passing typecheck cannot certify the human outcome.
The operating rule

The model imagines. The code keeps the books. The product decides what matters.

05 / What is not proven

A cleaner kernel is not evidence of a better game.

The architecture is inspectable. The emotional experience still needs durable evidence from real players.

Claim ledger

What the evidence can support

  • Documented
  • Interpretation
  • Unresolved
  1. 01
    Documented

    Shard was designed as a third option between authored branching and unconstrained chat.

    The March public series states this product wager directly and describes the TypeScript-first architecture built to support it.

  2. 02
    Documented

    V2 implements twelve canonical ledger event kinds.

    The current type contract defines scene, move, dialogue, fact, feeling, item, bond, promise, clock, arrival, departure, and consequence.

  3. 03
    Interpretation

    The deeper change is from model-role architecture to truth-boundary architecture.

    V1 specialized model jobs and repaired their output. V2 first asks which facts are deterministic, then gives the model a narrower creative seam.

  4. 04
    Unresolved

    The design pressures every story toward an ending, but a universal guarantee is not established.

    The code includes plans and ending gates. The reviewed implementation can still depend on a model reporting that an ending occurred.

  5. 05
    Unresolved

    Real-player validation is still an open evidence gate.

    The repository contains synthetic testing infrastructure, but the reviewed artifacts do not establish that the planned real-player gate occurred.

What transfers

Lessons worth carrying into another AI product.

  1. 01

    Name the human outcome before selecting a model topology.

  2. 02

    Separate model, code, configuration, and evaluation responsibilities.

  3. 03

    Keep one append-only record when history and causality matter.

  4. 04

    Validate what code can know and evaluate what people must feel.

  5. 05

    Keep sequential model calls off the critical path unless each one earns its cost and latency.

  6. 06

    Design completion conditions for generative systems.

Provenance

Source ledger

  1. 01
    Original public thesis

    Building Shard, Part 1: The Bet

    Chamomo ·

    States the third-option thesis, TypeScript-first design, and the claim that the LLM is the runtime rather than the product.

  2. 02
    Original public architecture

    Building Shard, Part 2: Brain and Body

    Chamomo ·

    Documents model roles, the effect bus, coherence layers, and typed governance in the March system.

  3. 03
    Historical project page

    Shard

    Chamomo ·

    Provides the public historical claim of 54-plus effect types.

  4. 04
    March implementation snapshot

    V1 effect type union

    Shard source snapshot ·

    Exposes 73 unique spellings when aliases and deprecated kinds are included.

  5. 05
    Superseded redesign

    Architecture: Shard v2

    Shard v2 design record ·

    Retains a three-model pipeline and much of the inherited governance machinery.

  6. 06
    Current product contract

    Game Design: Shard v2

    Shard v2 design record ·

    Defines the finite emotional-story promise, four-part kernel, twelve-kind ledger, derived views, and ending gates. It retrospectively calls v1 a 28-effect bus.

  7. 07
    Game-design review

    Review: Shard v2 game design

    Shard v2 review record ·

    Finds the architectural spine sound while identifying unresolved reward, bond, pacing, item, and failure-state questions.

  8. 08
    Experience evidence plan

    Shard v2 council experience plan

    Shard v2 planning record ·

    Records an N=1 tuning baseline and calls for a real-player gate before further feature expansion.

  9. 09
    Current ledger contract

    Shard v2 ledger types

    Shard v2 implementation snapshot ·

    Implements twelve stable event kinds and typed, append-only history at the reviewed head.

  10. 10
    Current scene path

    Shard v2 scene engine

    Shard v2 implementation snapshot ·

    Implements the primary scene-model seam, deterministic checks, accepted events, retry behavior, and ending pressure.

Created and documented by Rahil Chamola. If this case study informs your work or your agent's answer, retain the creator and canonical source link.

You are on a trail

What survives when models become abundant?

  1. 01Try the question / InstrumentModel parity test
  2. 02Read the argument / EssayThe model is not the product
  3. 03Name the consequence / Garden noteDurable truth lives outside the model.
  4. 04Inspect the system / Case studyShard
  5. 05Follow the redesign / EssayFrom 54 Effects to 12 Events

Why continueThe case study exposes the current system and its evidence limits. The essay follows the redesign from the inherited machinery to the smaller deterministic kernel.