Přejít na hlavní obsah

Open Manuscript Studio — OMI-SPEC-240 Implementation Profile

Document metadata

FieldValue
Document typeReference implementation profile / implementation evidence report
Profile version0.1.0
StatusDraft
Target specificationOMI-SPEC-240@0.1.0 — Rendering and Publication Profile Model
Target specification statusActive Draft
Reference implementationOpen Manuscript Studio
Studio evidence baseline63e391d8a3e0191ec2b24e494664df30d66f43e3
Evidence review date2026-09-18
Formal conformance claimNone
Approved conformance suiteNot published
Evidence classificationPartial implementation

1. Purpose

This document records how the current Open Manuscript Studio implementation maps to the requirements of OMI-SPEC-240 — Rendering and Publication Profile Model.

It has three purposes:

  1. make the relationship between the Draft specification and the reference implementation auditable;
  2. distinguish implemented behaviour from partial or unverified behaviour;
  3. define the remaining work required before Studio can make a stronger implementation or conformance statement.

This document is an implementation profile, not a publication profile authored under OMI-SPEC-240. It describes the behaviour of one implementation of the specification.

2. Conformance boundary

OMI-SPEC-240 is an Active Draft specification. It was activated as version 0.1.0 on 2026-09-09.

Open Manuscript Studio implements a substantial subset of its model, including publisher profiles, reusable publication styles, semantic-role-based rendering, print and interactive PDF targets, semantic HTML, IDML style import, publication diagnostics, deterministic build provenance and target-specific rendering.

However, this document does not claim that Studio is formally conformant with OMI-SPEC-240.

A formal conformance claim requires, at minimum:

  • a declared final conformance class and supported capability set;
  • stable machine-readable profile/style schemas where required;
  • requirement-mapped fixtures and tests;
  • an approved conformance suite and result format;
  • closure or explicit permitted treatment of every applicable normative requirement;
  • evidence that the tested implementation version is the version being claimed.

Until that framework exists, the strongest statement made here is Partial implementation with requirement-mapped evidence.

The presence of a feature in Studio does not make that feature normative, and the presence of a requirement in OMI-SPEC-240 does not prove that Studio implements it.

3. Declared implementation surface

3.1 Roles

The current Studio implementation provides evidence for the following OMI-SPEC-240 roles:

  • profile producer — users can select and edit publication profiles and reusable publication styles;
  • profile consumer — Studio loads profile/style data and applies it during validation and rendering;
  • renderer — Studio generates presentation-oriented publication output from canonical manuscript state;
  • validator-adjacent implementation — Studio performs publication-readiness, semantic-fidelity, target and schema checks, although no complete OMI-SPEC-240 conformance validator is claimed.

3.2 Output targets

The implementation profile currently covers these rendering targets:

TargetStudio implementation stateNotes
Print/archive PDFImplementedPinned server-side Vivliostyle renderer; active hyperlinks removed for the print/archive variant.
Interactive PDFImplementedUses the same publication source while retaining usable scholarly and external links.
Semantic HTMLImplementedScript-free semantic HTML package; print-only pagination behaviour is omitted.
CSS style exportImplemented interoperability outputExports style rules, but CSS is not itself the canonical OMI publication-profile model.
JATS XMLAdjacent validated publication outputUses profile-aware publication gates, but JATS semantic mapping is primarily governed by the publishing/export layer rather than being treated as a visual rendering target in this profile.
EPUB and DTP-oriented formatsOutside this profile's verified target claimStudio can export additional formats, but they are not treated here as evidence of complete OMI-SPEC-240 target conformance.

3.3 Architecture

The implemented publication path follows this separation:

Canonical OMI manuscript

├── scholarly semantics
├── metadata
└── references / notes / structured objects


Publisher / publication profile

├── publisher identity and legal display policy
└── selected reusable publication style


Target adaptation

┌──────────┼──────────┐
▼ ▼ ▼
print PDF interactive PDF semantic HTML
│ │ │
└──────────┴──────────┘


artifact + publication-build provenance

Typesetting corrections and presentation overrides are kept separate from canonical scholarly text and are applied in the publication layer.

4. Verified implementation evidence

The evidence baseline includes, among other implementation surfaces:

  • src/services/publicationStyleExport.ts — publication-style loading, semantic rendering and target handling;
  • src/document/publicationStyles/ — reusable style data and the reference journal style;
  • src/components/PublicationStyleEditor.tsx — named styles, page geometry, typography, paragraph-style inheritance, hyphenation and typesetting controls;
  • src/model/publicationParagraphStyles.ts — reusable paragraph-style definitions and presentation properties;
  • src/components/PublisherProfileEditor.tsx — publisher identity and publication policy editing;
  • src/components/PublicationProfilePanel.tsx — profile, style, HTML, JATS, PDF and publication-artifact surfaces;
  • src/services/idmlPublicationStyleImport.ts — Adobe IDML style-set import with explicit trust boundaries;
  • src/services/publicationBuildSidecar.ts — portable publication-build provenance;
  • src/services/printHyphenation.ts — language-aware print hyphenation;
  • src/services/exportPdf.ts and the Vivliostyle service path — separate print and interactive PDF behaviour;
  • HTML/JATS/PDF publication release-gate tests and provenance tests;
  • docs/architecture/publication-build-provenance.md;
  • docs/architecture/vivliostyle-pdf-artifacts.md.

The related OMI implementation documentation is:

5. Requirement mapping

The statuses in this table are evidence classifications for this implementation profile. They are not OMI lifecycle states and are not conformance results.

  • Implemented — repository evidence supports the stated requirement for the declared implementation surface.
  • Partial — relevant behaviour exists, but some required generality, diagnostics, target coverage, or requirement-specific evidence remains incomplete.
  • Not verified — the review did not establish sufficient evidence for the requirement.
  • Not applicable — the requirement does not apply to the declared implementation surface. No current requirement is excluded solely to improve the profile status.

5.1 Core rendering and profile separation

RequirementStatusStudio evidence / limitation
REQ-RPP-001ImplementedPublication rendering derives from canonical manuscript state. Publication styles and typesetting corrections are applied in the publication layer rather than persisted merely to satisfy rendering.
REQ-RPP-002ImplementedPublisher identity/profile policy and reusable publication-style typography are represented and edited separately.
REQ-RPP-003Not verifiedNo general OMI-SPEC-240 unknown-extension round-trip contract has been verified for arbitrary profile/style extension data.
REQ-RPP-004PartialJATS/HTML/PDF paths expose diagnostics and release blockers, but generic unsupported publication-style rules are not yet covered by a single complete cross-target diagnostic model.
REQ-RPP-005PartialCommitted revision digests, profile digests, pinned renderers, deterministic HTML timestamps and renderer-input fingerprints provide reproducibility evidence. Equivalent-result guarantees are not yet formally tested across every declared resource and target combination.

5.2 Publication profile and reusable style

RequirementStatusStudio evidence / limitation
REQ-RPP-010ImplementedStudio keeps an active/selected style within the available named style set and retains at least one valid style.
REQ-RPP-011ImplementedPublisher identity is separated from manuscript contributor/authorship and scholarly metadata state.
REQ-RPP-012ImplementedPublisher branding, identifiers and legal display data can change independently of canonical manuscript content and reusable typography.
REQ-RPP-013ImplementedTypography/page geometry are edited in the style layer and do not redefine publisher identity, licence semantics or manuscript metadata.
REQ-RPP-020ImplementedRendering maps OMI semantic roles to presentation. IDML names may help mapping but source-application style names are not required as manuscript semantics.
REQ-RPP-021ImplementedStyle application is presentation-oriented and does not intentionally rewrite object identity, contributor identity, citation targets or scholarly values.
REQ-RPP-022PartialParagraph-style BasedOn inheritance exists. A complete requirement-specific proof that every inheritance cycle is detected and rejected for all profile/style imports is not yet recorded.
REQ-RPP-023ImplementedIDML style import uses recognizable external style names as mapping evidence and produces reusable OMI publication-style roles rather than requiring InDesign semantics in the manuscript.

5.3 Targets, resources and cascade

RequirementStatusStudio evidence / limitation
REQ-RPP-030ImplementedPrint/archive PDF and interactive PDF are distinct target modes because they intentionally differ in hyperlink behaviour.
REQ-RPP-031ImplementedSemantic HTML omits fixed page size, running headers, page numbers and forced print pagination.
REQ-RPP-040PartialFont fallbacks and controlled resources exist, and renderer inputs are fingerprinted. A uniform manifest diagnostic for every materially layout-changing resource substitution is not yet verified.
REQ-RPP-041ImplementedCommercial fonts are not bundled or redistributed automatically; publication documentation treats missing publisher fonts as external resources.
REQ-RPP-050PartialStyle/profile/target/override resolution is deterministic in the implemented paths, but a complete property-by-property precedence fixture suite mapped to this requirement has not yet been published.
REQ-RPP-051ImplementedTypesetting and presentation overrides are explicit, removable publication-layer state and do not intentionally mutate canonical scholarly semantics.

5.4 Citations and notes

RequirementStatusStudio evidence / limitation
REQ-RPP-060ImplementedCitation rendering changes presentation while citation targets, locators and scholarly identity remain in the manuscript/reference model.
REQ-RPP-061PartialFootnotes/endnotes and their references are semantic objects and target renderers retain note relationships. Complete cross-target requirement fixtures are still needed.

5.5 Failure behaviour and diagnostics

RequirementStatusStudio evidence / limitation
REQ-RPP-100ImplementedRendering is non-destructive. Failed IDML import/render paths do not intentionally partially rewrite the canonical manuscript.
REQ-RPP-101PartialPublication release gates and fidelity diagnostics identify known unsupported/fallback conditions, but there is not yet one complete generic RPP diagnostic vocabulary covering every degraded profile/style property.
REQ-RPP-102Not verifiedThe review did not establish requirement-specific evidence that every failed profile/style edit preserves the last valid state in all editing/import paths.

5.6 Security, privacy and provenance

RequirementStatusStudio evidence / limitation
REQ-RPP-200ImplementedManuscript, profile and imported IDML/XML values are processed as untrusted data at publication/import boundaries.
REQ-RPP-201ImplementedPublication import/render paths reject or avoid executing supplied scripts, template code, processing instructions and active content.
REQ-RPP-202ImplementedIDML/XML trust boundaries reject unsafe declarations/processing instructions; JATS validation uses pinned local resources rather than accepting untrusted remote schema substitution.
REQ-RPP-203PartialPDF rendering blocks unsafe external resource loading and uses self-contained input. A single profile-wide external-resource policy covering every future target/resource type is not yet formalized as RPP evidence.
REQ-RPP-204PartialStudio has role-aware and double-anonymous review projections, but a dedicated renderer-boundary test matrix proving that every publication target excludes all unauthorized/private objects has not yet been published.
REQ-RPP-205ImplementedUntrusted IDML/parser errors are isolated from raw DOM rendering and fixed/safe diagnostics are used in the hardened import path.
REQ-RPP-206Partial.omi-build.json records artifact/manuscript/profile digests and renderer identity. The sidecar is not yet treated here as a separately signed archival attestation, so evidentiary integrity protection remains incomplete.

5.7 Accessibility and internationalisation

RequirementStatusStudio evidence / limitation
REQ-RPP-220ImplementedSemantic HTML and publication rendering preserve heading structure and supported accessibility metadata such as image alternative text and table semantics rather than styling them away.
REQ-RPP-221ImplementedThe semantic HTML package emits meaningful document structure instead of flattening the manuscript into generic visual containers.
REQ-RPP-222Not verifiedStudio does not currently make an OMI-SPEC-240 accessible-PDF conformance claim. Tagged-PDF structure and requirement-level evidence remain future work.
REQ-RPP-223PartialInteractive HTML/PDF retains usable links where supported. A complete target-specific accessibility test set for link text, destinations and keyboard interaction has not yet been mapped to this requirement.
REQ-RPP-230PartialFont fallback chains and system-font handling exist, but a complete missing-glyph detection and diagnostic contract has not been verified.
REQ-RPP-231ImplementedLanguage-aware hyphenation is applied during print rendering and typesetting without replacing the canonical manuscript text.

6. Implementation-specific capabilities beyond the minimum mapping

Studio currently contains publication capabilities that are useful implementation evidence but do not, by themselves, establish additional conformance:

  • visual publication editing against the current manuscript;
  • reusable named publication styles;
  • InDesign-like paragraph-style inheritance and next-style behaviour;
  • IDML style-set import;
  • publisher identity and legal-display configuration;
  • separate export and print stylesheets;
  • language-aware print hyphenation;
  • print/archive and interactive PDF variants;
  • semantic HTML packages;
  • validated JATS publication output;
  • publication-build sidecars with SHA-256 provenance;
  • OJS/OMP transfer of provenance-bound publication artifacts.

These features may evolve faster than the Draft specification. Where Studio and OMI-SPEC-240 diverge, the divergence must be resolved explicitly rather than treating implementation behaviour as normative by precedent.

7. Gaps before an Implemented or Tested classification

The following work is required before this profile should be promoted beyond Partial:

  1. publish a canonical machine-readable profile/style schema or an equivalent normative serialization contract;
  2. define and test unknown-extension preservation/loss reporting;
  3. add generic unsupported-rule and resource-substitution diagnostics to the publication manifest;
  4. publish deterministic cascade/precedence fixtures;
  5. add inheritance-cycle fixtures for imported and edited style graphs;
  6. map note-reference preservation across every declared target;
  7. add renderer-boundary authorization/visibility tests;
  8. define missing-glyph detection/fallback diagnostics;
  9. define the accessible-PDF capability level and, if claimed, add tagged-PDF/read-order tests;
  10. publish requirement-mapped OMI-SPEC-240 fixtures and automated result evidence;
  11. adopt the future OMI conformance framework when the relevant conformance specification and suite are active.

8. Evidence and release policy

A Studio release MUST NOT be described as OMI-SPEC-240 conformant solely because:

  • the publication-profile UI exists;
  • a profile JSON can be exported;
  • PDF/HTML/JATS rendering succeeds;
  • the implementation profile lists a requirement as Implemented;
  • Studio is the OMI reference implementation.

A future conformance statement must identify:

  • the exact Studio release/build;
  • the exact OMI-SPEC-240 version;
  • the claimed conformance class or capability profile;
  • the conformance-suite version;
  • the test result/report;
  • any permitted limitations.

Until then, user-facing language should use formulations such as:

Studio implementation profile for OMI-SPEC-240@0.1.0 Draft. Implementation coverage is documented separately; this is not a formal conformance claim.

9. Relationship to OMI-SPEC-230

OMI-SPEC-230 defines the publication workflow and publication-output lifecycle. OMI-SPEC-240 defines the rendering/profile contract used inside that workflow.

The boundary is:

OMI-SPEC-230
publication job / publication state / output lifecycle


OMI-SPEC-240
resolved profile + style + target rendering contract


publication artifact + diagnostics + provenance

Studio's publication build pipeline exercises both domains, but evidence for one specification must not be treated automatically as evidence for the other.

10. Maintenance

This profile should be reviewed when any of the following changes:

  • OMI-SPEC-240 version or lifecycle state;
  • publication profile/style data structures;
  • rendering targets;
  • Vivliostyle or other renderer versions/policies;
  • IDML/style import behaviour;
  • publication-build provenance format;
  • authorization projection in publication rendering;
  • accessibility claims;
  • formal OMI conformance tooling.

Requirement statuses should be changed only when repository evidence supports the change.

11. Change history

  • 0.1.0 — 2026-09-18: Initial Studio implementation profile for OMI-SPEC-240@0.1.0; records the declared target surface, requirement-by-requirement implementation evidence, non-conformance boundary and gaps to stronger evidence.