v0.1.0 status: everything is a draft (on purpose) — release staged, review invited

[maehr]

TextRefs v0.1.0 — the first tagged release of the standard, the site, and the seed registry — is staged and open for review before it ships. The release PR is textrefs.org#4 (staging → main), and it doubles as the community-review venue for the three architecture decisions it adopts. Everything below is what that PR ships.

The headline decision: every record in the registry is now status: draft. TextRefs is not stable yet, and we've made that a feature of the data model rather than a disclaimer.

The three ADRs (decisions/)

• ADR-0001 — dcterms:conformsTo replaces target_kind (from target_kind: prefer a dereferenceable IRI + dct:conformsTo over an enumerated scheme list #6): mapping targets no longer carry a curated scheme label backed by an ever-growing appendix table; an optional conforms_to IRI carries the conformance claim instead, following Linked Art practice. The identifier IRI stays authoritative.
• ADR-0002 — deterministic identifiers (adr: purpose of deterministic UUIDs (compiler reproducibility vs. offline minting) #15, feat(spec)!: seed CanonicalReference UUIDs from the semantic identity tuple (ADR-0002) #21): a reference UUID is computed from (work_key, citation_system_key, locator) alone — normalization_version is gone from the data model. Anyone holding the canonical fields can compute the registry UUID offline; the registry stays authoritative for which locators are canonical.
• ADR-0003 — draft lifecycle (adr: review lifecycle and ephemeral draft state (retractable pre-promotion identity) #22, feat(spec): draft lifecycle with retractable pre-promotion identity (ADR-0003) #23): the identifier-persistence promise attaches at promotion (draft → candidate, expert review), not at publication. Drafts may be corrected or retracted freely, without tombstones; from candidate on, identity is permanent and tombstone rules apply.

Standard changes

• Dereferenceable-location guidance strengthened MAY → SHOULD (Dereferenceable locations: change MAY to SHOULD #7) and a self-contained multi-translation worked example with @context (Add a complete worked example including @context #8) — both from community review.
• Citation-system profiles MUST declare canonical ASCII forms — leading-zero policy and letter case (standard: require canonical ASCII digit and case forms in citation-system profiles #13). The Bible profile is pinned to case-sensitive OSIS book codes; Bekker accepts the full corpus from page 1 and forbids leading zeros; Stephanus likewise.
• Succession is provenance, not equivalence: tombstoned records carry superseded_by (dcterms:isReplacedBy); MappingAssertion stays reserved for work-level equivalence with a Work-IRI subject (standard: MappingAssertion.subject Work-IRI constraint conflicts with tombstone/versioning successor links #10).
• Published Work records carry direct skos:exactMatch/skos:closeMatch edges, so SKOS-aware clients can follow mappings without dereferencing the reified assertions (standard: JSON-LD output does not emit direct SKOS mapping triples #11).
• dcterms:license has a single IRI-typed range: authored SPDX ids are emitted as canonical SPDX IRIs (standard: license and license_url both map to dcterms:license (mixed literal/IRI range) #12).
• Editorial erratum batch: RFC 9562 for UUIDs, application/ld+json for the JSON-LD siblings, UUID-namespace clarifications, tombstone/alias prose fixes (standard: erratum — editorial batch (RFC 9562, ld+json media type, aliases.json, tombstone rationale, namespace note) #14).

Registry data (textrefs/registry)

• All 12 works, 10 citation systems, and 24 mappings demoted to draft — a one-time bootstrap demotion; the promotion ladder is one-way from here.
• Citation systems now carry a required published description; examples and normalization_version are gone.
• Locator regex fixes: Bekker pages 1–9999, digit-initial OSIS books (1Cor), no leading zeros anywhere they aren't attested.
• Non-SPDX license values cleaned up; registry CI now validates against the branch registry data actually integrates with.

Site and infrastructure

• Draft records are flagged with a banner and excluded from search indexing on all record and browsing pages.
• Astro 7 / Starlight 0.41 with the full plugin stack and all dependencies at latest (Upgrade to Starlight 0.41 / Astro 7 (blocked on starlight-llms-txt) #20).
• Reference pages rebuilt as minimal static HTML (3.3 KB instead of 40.7 KB each): the deployed site went from 1.9 GB to 270 MB, fully readable without JavaScript, with COinS for reference managers.
• Governance statutes (EN/DE) updated for the draft entry and promotion tracks; get-started guides rewritten.

What this means for you

This is exactly the moment to propose works. New data merges as draft after technical review — no long-term commitment on either side, identifiers can still be fixed if wrong, and a re-proposed record regains the same UUID by construction. Expert review then promotes records to candidate, which is when the persistence promise kicks in.

The contribution funnel:

1. Propose a work / citation system as an issue or PR in textrefs/registry (YAML, ~40 lines for a work; see the authoring guide).
2. Technical review (schema, locator regex, canonical form) → merges as draft, appears on the site flagged as draft.
3. Expert review → promotion to candidate; the identifier becomes permanent.

Where we want help

• Review the release PR, especially the three ADRs: textrefs.org#4.
• #9 — locator validation (help wanted): is locator_regex the right machine-checkable contract, or do some citation traditions need more? Design input from people who know their citation systems is wanted.
• More works: theology, classics, philosophy, law — if your field has an edition-independent citation system, it belongs here.

Thanks to Julien A. Raemy for the pre-release review that prompted ADR-0001 and shaped the §13 worked example and the dereferenceability guidance.

Release notes and the v0.1.0 tag follow once the PR merges.