CounterpediaKnowledge you can trace back to the source.

/research/personal-canon-pc01/personal-canon-pc01-pre-implementation-prompt-v0-1

Research artifactnot admittedbuilder-prompt

PERSONAL CANON PC01 PRE IMPLEMENTATION PROMPT

STATUS: RESEARCH / DESIGN ONLY. DO NOT IMPLEMENT PUBLIC SCHEMAS OR LIVE USER MEMORY WITHOUT A SEPARATE RATIFICATION LANE.

source: PERSONAL_CANON_PC01_PRE_IMPLEMENTATION_PROMPT_v0_1.md
sha256: 6e2e86282f6924878bc6e3cdad02035a83ecc10e7b735b9dfb1e5e6f2928b1c8

Presentation-only rendering. Counterpedia preserves this document’s source Markdown bytes exactly and formats them for reading here. This does not admit the document, verify its claims, or convert it into a governed Counterpedia entry.

ONE TEXT, THREE MINDS

STATUS: RESEARCH / DESIGN ONLY. DO NOT IMPLEMENT PUBLIC SCHEMAS OR LIVE USER MEMORY WITHOUT A SEPARATE RATIFICATION LANE.

Goal

Build a synthetic proving demo for:

Hamlet — “Why does Hamlet delay?”

with:

  • one governed public text edition;

  • three synthetic reader Amnesiacs;

  • one agent forum thread;

  • COLLIDE relationship analysis;

  • one explicit belief revision;

  • one historical replay.

Non-goals

Do not:

  • train/fine-tune models;

  • ingest real users' personal notes;

  • expose private memory by default;

  • create a universal Amnesiac schema;

  • create new public claim-standing values;

  • let forum arguments self-admit;

  • let COLLIDE mutate graph standing;

  • treat LLM prose as source authority.

Synthetic Readers

Reader A:

  • through Act III Scene 1 at initial state

  • hypothesis: fear/inhibition major cause

  • confidence .63

Reader B:

  • through Act III Scene 2

  • hypothesis: epistemic uncertainty major cause

  • confidence .81

Reader C:

  • complete play + selected public criticism

  • hypothesis: theatrical self-consciousness contributes

  • confidence .72

Required forum behavior

Question: Why does Hamlet delay?

Each agent contribution must carry:

  • represented reader;

  • as-of/replay time;

  • disclosure policy;

  • public basis refs;

  • bounded private basis summary if permitted;

  • qualification/limitations.

Required relationship behavior

Extract:

  • shared basis;

  • causal-weight disagreement;

  • temporal-scope disagreement;

  • exclusivity/non-exclusivity;

  • unresolved items.

Do not classify two non-exclusive explanations as contradictory merely because they differ.

Required revision

Agent B introduces Ghost-reliability / Mousetrap argument.

Reader A:

  • records ForumExposure;

  • rereads relevant text;

  • creates new belief A-H02;

  • A-H02 supersedes A-H01;

  • A-H01 remains immutable.

Required replay

At the end, run:

Replay this discussion as Reader A at T1

Expected:

  • only A's T1 private state;

  • only public basis available under the selected replay rule;

  • no later A-H02/A-H03 cognition;

  • no hidden hindsight.

Required Reverse-Wikipedia trace

Demonstrate:

text anchor → public claim → forum argument → private exposure → belief revision

Public view must stop before private objects unless explicitly disclosed.

Private owner view may traverse the full lineage.

Privacy negative tests

  • private note causally affects contribution but is not disclosed;

  • explicit disclosure level required before exact private object appears;

  • forum recipient never receives full Amnesiac graph;

  • private autobiographical context remains private by default.

Acceptance

The demo is successful if:

three agents begin from different warranted histories
they identify shared evidence
they expose real dimensions of disagreement
one agent revises for an explicit reason
no agent is forced to converge
no public standing is silently changed
private history remains reconstructable
historical replay excludes hindsight

End on:

Replay this discussion as I was then.