Context Vigilance
Search/

← Corpus / reach-edu-hub / narrative

reach-edu-hub Draft

Citation Registry for reach-edu-hub Narratives

One hex code per unique article, held once per client. Narratives cite [^abc123] inline; definitions resolve from here. Codes are minted once and never renumbered — reorder, split, or scramble content freely without breaking claim–source pairing.

Path
narratives/citations/README.md
Authors
Michael Staton
Augmented with
Claude Code (Fable 5)
Tags
Citations · LFM · Hex-Codes · Registry · reach-edu

Citation Registry — reach-edu-hub narratives

Per LFM rule 2 (lossless-flavored-markdown skill): inline references use stable hex codes ([^1ucdcd]), never sequential numbers. Sequential numbers break the moment content is reordered, split, or copy-pasted between documents — the memopop assembler’s positional-renumbering pain (manage-memo-citations skill) is the counterexample this design avoids. Render order is resolved at build time; the code is the durable identity.

Files

  • registry.csv — source of truth. One row per unique article: hexcode, title, source_url, published, capture_path, origins, added.
    • hexcode — minted once (random 6-char lowercase hex, ≥1 letter), never reused, never renumbered. If an article drops out of every narrative, its row stays (codes are permanent identities).
    • capture_path — path into augment-it/clients/reach-edu/corpus/ where the verbatim Jina capture lives (empty = not yet captured).
    • origins — where the article entered the registry (mega-gifts-research, narrative:<slug>), pipe-separated.
  • definitions.md — generated, paste-ready footnote definitions in the form [^abc123]: 2026, Jan 04. [Title](https://…). Regenerate from the CSV; do not hand-edit (edit the CSV, then regenerate).

Discipline

  1. Citing in a narrative: find the article in registry.csv, use its code inline with a single space before each mark, Obsidian-friendly (…claim. [^abc123] [^def456] — never …claim.[^abc123][^def456]), and copy its definition line from definitions.md into the narrative’s ## Citations block. Definitions may sit anywhere in the doc; render matches them by code.
  2. New article: add a row to registry.csv with a freshly minted code (check uniqueness against the hexcode column), then cite it. Capture the article into the augment-it corpus inbox when it matters enough to preserve verbatim, and backfill capture_path.
  3. Never renumber, never recycle. A code that exists means that pairing is claimed forever, even across future decks and one-pagers.
  4. Same article everywhere. All Reach surfaces (strategy narratives, decks, one-pagers) cite the same code for the same article — that’s why the registry is per-client, not per-document.

Provenance

Seeded 2026-07-28 from two pools: the mega-gifts-by-topic funding research (augment-it/clients/reach-edu/outputs/2026-07-28_mega-gifts-by-topic/, 81 articles, all captured in the corpus) and the source lists of the seven June-2026 strategy narratives (46 articles, mostly uncaptured). 125 unique articles after overlap.