Document checksum-substrate join keys in the federation docs - #481
Merged
Conversation
Adds md5/sha256 content-hash join keys to public-export-v1.md (draft addition, matching its HOLD posture), a linking paragraph in federation-v1.md, and a Tier 6 note in readiness-audit.md. States the sha256-pairing rule and the counted-once-per-md5-group federation consequence from #473, and the discovery-via-hash-exchange-not-crawling boundary.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Owner-directed (2026-07-25): document the md5/sha256 identity-group work's (#473) federation benefits and discovery features in the federation docs, before that build's own PRs land.
docs/federation/public-export-v1.md— new### Exact-match join keys: md5/sha256 content hashessubsection under §2 (Keying + tooling), clearly dated/marked as a draft addition consistent with this doc's existing BUILD-hold posture (not a ratification of anything new). States, in substance, md5 identity groups: checksum substrate, evidence transfer, and group-level vote pooling #473's own ratified pairing rule (md5 alone is not collision-resistant; a trust context requires sha256 pairing; an md5-match/sha256-mismatch is a loud anomaly, never a silent fallback) and the counted-once rule's federation consequence (an importedFEDERATEDverdict keyed on a content hash that matches a whole local md5 group applies once to that group, never once per member). Also adds the "discovery, not crawling" boundary — sibling-copy discovery happens through hash exchange over a published export, never by crawling Google Drive (no public-corpus search surface exists, and sources are community-submitted).docs/federation-v1.md— one paragraph in the "Interchange keys & stability contract" section linking the checksum substrate (md5 identity groups: checksum substrate, evidence transfer, and group-level vote pooling #473) as an exact-match complement to the existingcontent_phashjoin-key path, and citing (not restating)docs/theory.md§6's Dawid-Skene framing for peer-reliability estimation over shared content hashes.docs/upstreaming/readiness-audit.md— one line in the existing Tier 6 "Federation verdict-exchange consumer component" entry, noting content-hash join keys are what make that component's payoff concrete once built (still gated on the same "no live peer yet" precondition, not a new readiness question).docs/upstreaming/extractable-primitives.md— not touched, deliberately: PR Add Card.md5_checksum substrate and a listings-rewalk backfill command #477's still-open branch already adds a CLEAN row for the real checksum-substrate primitive (md5_backfill.py's listings-walk reconciliation); adding a second row here would duplicate it. The intake-triage pattern (issue Intake triage: price a candidate source by checksum overlap before any fetch #478, "filed minutes ago") has no code yet, and the ledger's own stated methodology is "read directly from the repo, not inferred from doc comments" — a row for unbuilt work would violate that. Both skips are judgment calls, not oversights.References #473, #451 (item 5, federation join key).
Test plan
python3 .github/scripts/docs_lint.py --strict— cleanpython3 .github/scripts/tests/test_docs_lint.py— 26 tests OKpython3 .github/scripts/tests/test_publish_readme.py+ manual regenerate-and-diff — no driftdocs-lint.yml) — not watched from this session per policy; will confirm on the PR🤖 Generated with Claude Code