Standards Mapping

Standards Mapping

SIGNET is a profile-and-bridge, not a fork (design principle 1.2). The Canonical Data Model is a coherent superset that profiles existing standards: a SIGNET document can always be projected down to a conforming OCDS release, a Peppol BIS invoice, or an eForms notice, and data arriving in those formats can always be lifted into the CDM.

The mappings on this page are normative. A conforming implementation MUST be able to transform between the CDM and these formats without loss of the fields defined in the mapping (specification §8).

The normative mapping table

CDM object	Maps to / from	Mapping basis
Need	OCDS planning	OCDS 1.1 release
SourcingEvent, Lot	OCDS tender; eForms notice	OCDS 1.1; eForms
Submission	OCDS bid (bid extension); UBL Tender	OCDS bid extension
Award	OCDS award	OCDS 1.1
Contract	OCDS contract	OCDS 1.1
Order	UBL 2.3 Order; Peppol BIS Ordering	UBL 2.3
Catalogue	UBL 2.3 Catalogue; Peppol BIS Catalogue	UBL 2.3
Invoice	EN 16931; Peppol BIS Billing; UBL Invoice; Factur-X	EN 16931 semantic model
Party	OCDS parties; ePO Agent	OCDS org; ePO
Credential	W3C Verifiable Credentials 1.1	VC data model
Identifier (did)	W3C DID 1.0	DID core
Provenance	W3C PROV-O	PROV
Semantic layer (all)	EU eProcurement Ontology (ePO)	ePO OWL/RDF

What "without loss" means

• Down-projection (CDM → target). Where a CDM object has no counterpart in the target standard — most importantly the entire Agent Layer when projecting to OCDS — the projection omits it without error. SIGNET-original structures are always cleanly separable from third-party standards.
• Up-lift (target → CDM). Lifting data from a target format leaves the SIGNET-only fields empty; nothing is fabricated.

This separability is also an IP property: the SIGNET-original contribution (agents, mandates, decisions, policy-as-data, consent) never entangles the third-party standards it bridges.

The standards SIGNET bridges

OCDS — Open Contracting Data Standard

Provides the lifecycle vocabulary. The process layer's phase model (planning → tender → award → contract → implementation) is OCDS's. SIGNET adopts OCDS's successful low-friction, plain-JSON consumption approach (see Serialisation).

EN 16931 — European e-invoicing semantic model

Provides the invoicing semantics. The Invoice, InvoiceLine, and VatBreakdown objects carry EN 16931 Business Term (BT) and Business Group (BG) references field-by-field. This is the basis of the EU ViDA alignment — see EN 16931 & ViDA E-Invoicing.

UBL / Peppol BIS — document syntax & business interoperability specs

Provide the document layer for Order, Catalogue, and Invoice. The repository ships a working transform from a SIGNET Invoice to UBL 2.1 / Peppol BIS Billing 3.0.

W3C VC & DID — verifiable credentials and decentralised identifiers

Provide decentralised identity. Identifier supports the did scheme; Credential is a W3C Verifiable Credential reference, with optional selective disclosure.

W3C PROV — provenance ontology

Provides the provenance vocabulary. Provenance maps to PROV-O terms (wasGeneratedBy, wasDerivedFrom, wasAttributedTo) in the JSON-LD context.

EU eProcurement Ontology (ePO)

Provides the semantic alignment across the whole model. The JSON-LD context maps SIGNET terms to ePO where equivalents exist (e.g. procuringParty → ePO:hasBuyer, submittingParty → ePO:hasTenderer).

Solid — data sovereignty

Inspires the consent model. Consent expresses Solid-style access grants as CDM data.

A2A — Agent-to-Agent / Agent Card

Inspires capability discovery. SyntheticAgent and AgentCapability align to the A2A Agent Card model (/.well-known/agent.json).

SIGNET's original contribution

The Agent Layer — synthetic agents, capabilities, mandates, decisions, and policy-as-data — has no equivalent in any existing procurement standard. It is the part of SIGNET that is genuinely new, and the reason the CDM exists rather than a thin wrapper over OCDS.

Where to go next

• EN 16931 & ViDA E-Invoicing — the one mapping that is runnable and CI-verified.
• Serialisation (JSON-LD) — the @context that carries the semantic alignment.