{
  "abstract": "Long-running work with language models fails less often because the model cannot generate text than because the surrounding system cannot preserve what is currently true. Transcripts accumulate proposals, corrections, reversals, sources, and outputs, but they do not reliably distinguish what was said from what was accepted. A later model must reread history and reconstruct state, often blending obsolete instructions with current intent. Long-Context State Memory (LCSM) is a model-agnostic runtime for durable canonical state, provenance, authority, journaled transition, and reconstruction. Backboard is the reference workbench that makes the protocol visible. Conversation is intake; bounded notes and canonical objects represent current recognized truth; immutable versions preserve change of mind; proposals are validated and gated before they move a canonical pointer; receipts explain why the current state exists. The architecture is summarized by one chain of command: The model proposes. The runtime validates. The policy gate authorizes. The journal records. The canon pointer moves. The user can inspect. LCSM is not an attempt to place an infinitely long transcript inside a model. It removes the obligation for the model to own memory. The model receives a scoped working set compiled from authoritative state and relevant evidence. The runtime retains the durable object.",
  "author": "Hayden Lindley",
  "category": "Continuity & Canon",
  "claims": {
    "claims": [
      {
        "claim_id": "P02-C01",
        "evidence_refs": [
          "P02-S01",
          "P02-S05"
        ],
        "limitations": [
          "Production persistence implementation requires independent verification."
        ],
        "public": true,
        "scope": "Protocol and reference workbench.",
        "statement": "LCSM separates immutable historical versions from a current canonical pointer and records change through append-preserving journal operations.",
        "status": "SPECIFIED"
      },
      {
        "claim_id": "P02-C02",
        "evidence_refs": [
          "P02-S01"
        ],
        "limitations": [
          "A live model classifier is not validated by the reference mechanism alone."
        ],
        "public": true,
        "scope": "Fixture-based workbench design and acceptance contract.",
        "statement": "The Backboard reference loop can contain replaceable semantic judgment inside deterministic validation, authority, journal, and projection mechanics.",
        "status": "VERIFIED_MECHANISM"
      },
      {
        "claim_id": "P02-C03",
        "evidence_refs": [
          "P02-S01",
          "P02-S03",
          "P02-S04"
        ],
        "limitations": [
          "Matched comparative study remains outstanding."
        ],
        "public": true,
        "scope": "Comparative human/model evaluation.",
        "statement": "State-based working sets should reduce obsolete-context errors and resumption cost relative to transcript-only memory.",
        "status": "HYPOTHESIS"
      }
    ],
    "disclosure_note": "Claims marked non-public in the authored ledger are omitted from this public artifact.",
    "paper": "p02-lcsm-backboard",
    "version": "0.1.0",
    "withheld_claim_count": 0
  },
  "date": "2026-07-27",
  "diagrams": {
    "diagrams": [
      {
        "brief": "Left lane shows a transcript containing conflicting statements. Right lane shows an archive feeding versioned canonical notes with one current pointer.",
        "format": "svg",
        "id": "P02-F01",
        "path": "/diagrams/p02-lcsm-backboard/P02-F01.svg",
        "status": "source-generated conceptual architecture",
        "title": "Transcript versus canonical state"
      },
      {
        "brief": "LLM proposes → runtime validates → policy authorizes → journal records → canon pointer moves → projections refresh, with human inspection at every consequential gate.",
        "format": "svg",
        "id": "P02-F02",
        "path": "/diagrams/p02-lcsm-backboard/P02-F02.svg",
        "status": "source-generated conceptual architecture",
        "title": "The LCSM chain of command"
      },
      {
        "brief": "Side-by-side current and candidate notes, including sources, authority, information loss, dependency effects, and acceptance controls.",
        "format": "svg",
        "id": "P02-F03",
        "path": "/diagrams/p02-lcsm-backboard/P02-F03.svg",
        "status": "source-generated conceptual architecture",
        "title": "Two-note commit"
      },
      {
        "brief": "One journal fans out to chat, board, FOVEA atlas, mobile strip, agent capsule, and public proof page; deleting any projection leaves canon intact.",
        "format": "svg",
        "id": "P02-F04",
        "path": "/diagrams/p02-lcsm-backboard/P02-F04.svg",
        "status": "source-generated conceptual architecture",
        "title": "Projection reconstruction invariant"
      }
    ],
    "paper": "p02-lcsm-backboard"
  },
  "evidence_state": "VERIFIED_MECHANISM",
  "id": "p02-lcsm-backboard",
  "institution": "Glyphd Labs",
  "integrity": {
    "claim_ledger_sha256": "0526fb72770877bbda31e0ed8a7301f2053885f5c4f9219f295f65ceb0a03030",
    "paper_mdx_sha256": "aa66756d383af07430631b945b2867f27ed2054c0c0bd297196be3e56fe4b2db"
  },
  "public_disclosure_boundary": "This edition publishes the state architecture, transition doctrine, and public-safe reference loop. It excludes private user data, confidential classifier prompts, unpublished policy thresholds, security-sensitive storage details, and patent-claim drafting. The term “canonical” describes governed recognized state, not an assertion of universal truth.",
  "publication_class": "public-safe working paper",
  "publication_url": "https://www.glyphd.com/papers/p02-lcsm-backboard",
  "revisions": {
    "paper": "p02-lcsm-backboard",
    "revisions": [
      {
        "date": "2026-07-27",
        "summary": "First synthesized public-safe institute edition compiled from the Glyphd project corpus.",
        "version": "0.1.0"
      }
    ]
  },
  "schema_version": "1.0.0",
  "slug": "p02-lcsm-backboard",
  "sources": {
    "notes": "Internal source paths are provenance references. Public publication must expose only approved public-safe excerpts or archived public artifacts.",
    "paper": "p02-lcsm-backboard",
    "sources": [
      {
        "date": "2026-07-04",
        "id": "P02-S01",
        "path": "LCSM-WORKBENCH-E2E-SPEC.v0.md",
        "title": "LCSM Workbench E2E Specification v0"
      },
      {
        "date": "2026-07-12",
        "id": "P02-S02",
        "path": "GLYPH-DESK-MASTER-SPEC.v1.md",
        "title": "Glyph Desk Master Specification v1"
      },
      {
        "date": "2026-07-21",
        "id": "P02-S03",
        "path": "CONTINUITY-CORE-DESIGN-REVIEW2.md",
        "title": "Continuity Core Design Review 2"
      },
      {
        "date": "2026-07-18",
        "id": "P02-S04",
        "path": "ZEKE-V1-CANON-RECOVERY-AND-GAP-AUDIT-v0.1.md",
        "title": "Glyphd V1 Canon Recovery and Gap Audit"
      },
      {
        "date": "2026-07-25",
        "id": "P02-S05",
        "path": "MCE-1-SYSTEM-CONTRACT-v0.1.md",
        "title": "MCE-1 System Contract"
      }
    ],
    "version": "0.1.0"
  },
  "subtitle": "Durable state, provenance, authority, and inspectable change of mind for AI systems",
  "title": "LCSM / Backboard",
  "version": "0.1.0"
}