Gartner 2025 Extensions — Practice Guide¶
Background¶
The Gartner 2025 alignment analysis identified three gaps where Linked.Archi had partial coverage of Gartner's EA themes. Three extensions were built to close those gaps:
| Extension | Prefix | Addresses Theme | Key Classes |
|---|---|---|---|
| AI Ethics & Governance | aigov: |
T4 — AI Strategy & Governance | AISystem, RiskClassification, ConformityAssessment, BiasAssessment, HumanOversightPlan |
| Financial Architecture | fina: |
T5 — Financial Acumen | CostModel, CostItem, InvestmentCase, CostBenefitAnalysis, Budget |
| EA-as-a-Service | easvc: |
T1 — EA as Internal Consultancy | EAService, ServiceCatalog, ServiceEngagement, SatisfactionAssessment, EAMaturityAssessment |
Each extension follows the standard Linked.Archi pattern: ontology, taxonomy (SKOS), SHACL shapes, metamodel manifest, and reference data. Each has a dedicated practice guide for detailed usage:
- AI Ethics & Governance Practice Guide
- Financial Architecture Practice Guide
- EA-as-a-Service Practice Guide
This document provides the cross-extension view — how the three compose with each other and with existing Linked.Archi modules.
1. Extension Summaries¶
1.1 AI Ethics & Governance (aigov:)¶
Namespace: https://meta.linked.archi/ai-governance/onto#
Imports: arch:core
Artifacts: extensions/ai-governance/ — onto · tax · shapes · metamodel · reference-data
Provides vocabulary for managing AI system governance — risk classification (EU AI Act), conformity assessment, bias assessment, explainability documentation, human oversight plans, governance policies, incident tracking, and transparency records.
Reference data: 4 EU AI Act risk levels (Unacceptable, High, Limited, Minimal), 7 AI principles (OECD + EU), 3 human oversight modes (Human-in-the-Loop, Human-on-the-Loop, Human-in-Command).
Viewpoints: AI Governance Overview, AI Risk Assessment, AI Transparency & Disclosure, AI Incident Tracking.
SHACL enforcement: Every AI system must have a risk classification. High-risk systems must have conformity assessments, bias assessments, and human oversight plans. Every conformity assessment must reference principles and have a result.
1.2 Financial Architecture (fina:)¶
Namespace: https://meta.linked.archi/financial-architecture/onto#
Imports: arch:core
Artifacts: extensions/financial-architecture/ — onto · tax · shapes · metamodel · reference-data
Makes cost a queryable dimension of the architecture knowledge graph. Provides TCO models, cost allocation to capabilities and value streams, investment case tracking (Run/Grow/Transform), cost-benefit analysis for decision options, budgets, and FinOps cloud cost management.
Reference data: 10 cost categories (License, Infrastructure, Labor, Support, Integration, Compliance, Opportunity, Technical Debt Interest, Training, Decommissioning), 3 investment types (Run, Grow, Transform), 4 cost periods.
Viewpoints: Cost Overview, Investment Portfolio, Technical Debt Landscape, Cloud FinOps.
SHACL enforcement: Every cost model must have a currency and at least one cost item. Every investment case must have a type and amount. Every CBA must have estimated cost, benefit, and currency.
1.3 EA-as-a-Service (easvc:)¶
Namespace: https://meta.linked.archi/ea-service/onto#
Imports: arch:core
Artifacts: extensions/ea-service/ — onto · tax · shapes · metamodel · reference-data
Worked example: examples/ea-service/ea-service-example.ttl
Models the EA function as an internal management consultancy. Provides service catalogs, service requests, engagement tracking, outcome measurement, stakeholder satisfaction, EA capability portfolios, and maturity assessment.
Reference data: 10 EA service types (Architecture Review, Technology Radar, Portfolio Assessment, Target State Design, Standards Guidance, Impact Analysis, Modernization Planning, AI Readiness Assessment, Compliance Validation, Onboarding Consultation), 6 engagement statuses, 5 maturity levels (Gartner-aligned).
Viewpoints: EA Service Catalog, Engagement Tracking, EA Capability Portfolio, EA Maturity Assessment, Stakeholder Satisfaction.
SHACL enforcement: Every service must have a type and label. Every engagement must have an assigned architect and status. Every satisfaction score must be 1–5. Every maturity assessment must assign a level.
2. How the Extensions Compose¶
The three extensions are independent — each imports only arch:core and can be used standalone. But they are designed to compose with each other and with existing Linked.Archi modules.
2.1 Composition Diagram¶
┌─────────────────────────────────┐
│ EA-as-a-Service (easvc:) │
│ Service catalog, engagements, │
│ satisfaction, maturity │
└──────────┬──────────────────────┘
│ easvc:influencedDecision
│ easvc:producedDeliverable
┌──────────▼──────────────────────┐
│ Architecture Decisions (ad:) │
│ Decisions, forces, options │
│ ← existing extension │
└──────────┬──────────────────────┘
│ fina:hasCostBenefitAnalysis
│ ad:influencedByForce
┌────────────────┼────────────────┐
│ │ │
┌─────────▼────────┐ ┌────▼──────────┐ ┌───▼──────────────┐
│ Financial │ │ AI Ethics & │ │ TIME Framework │
│ Architecture │ │ Governance │ │ ← existing │
│ (fina:) │ │ (aigov:) │ │ │
│ TCO, CBA, ROI │ │ Risk, bias, │ │ Portfolio │
│ │ │ oversight │ │ rationalization │
└─────────┬─────────┘ └──────┬────────┘ └───┬──────────────┘
│ │ │
│ fina:hasCostModel│ aigov:* │ timefw:*
│ │ │
┌─────────▼──────────────────▼───────────────▼─────────────┐
│ arch:core │
│ + ArchiMate + C4 + Backstage + BPMN + TOGAF │
└──────────────────────────────────────────────────────────┘
2.2 Key Composition Points¶
EA-as-a-Service → Architecture Decisions:
Every EA engagement that influences a decision creates a traceable link via easvc:influencedDecision. This answers: "Which EA engagements led to the most impactful architecture decisions?"
Financial Architecture → Architecture Decisions:
Every decision option can have a cost-benefit analysis via fina:hasCostBenefitAnalysis. This answers: "Which option has the best NPV?" and enables cost-informed decision making.
Financial Architecture → TIME Framework:
Cost models can be attached to TIME-assessed applications via fina:hasCostModel. This answers: "What is the total migration cost for all Migrate-classified applications?"
Financial Architecture → ArchiMate Capabilities:
Cost models can be allocated to business capabilities via fina:allocatedToCapability. This answers: "What is the annual run cost per business capability?"
AI Ethics & Governance → Architecture Decisions:
AI governance choices become traceable decisions via arch:refines. This answers: "Which decisions shaped our AI governance framework?"
EA-as-a-Service → Financial Architecture: Cost models can be attached to EA services to track the cost of delivering EA itself. This answers: "What does it cost to run the architecture review service?"
3. Cross-Extension Examples¶
3.1 Cost-Informed Architecture Decision¶
An EA engagement produces a target-state design. The decision options have cost-benefit analyses. The chosen option has an investment case.
@prefix easvc: <https://meta.linked.archi/ea-service/onto#> .
@prefix easvcrd: <https://meta.linked.archi/ea-service/reference-data#> .
@prefix fina: <https://meta.linked.archi/financial-architecture/onto#> .
@prefix finard: <https://meta.linked.archi/financial-architecture/reference-data#> .
@prefix ad: <https://meta.linked.archi/arch-decision#> .
@prefix am: <https://meta.linked.archi/archimate3/onto#> .
@prefix skos: <http://www.w3.org/2004/02/skos/core#> .
@prefix : <https://model.example.com/cross-ext#> .
## The system under discussion
:PaymentPlatform a am:ApplicationComponent ;
skos:prefLabel "Payment Platform"@en .
## EA engagement that produced the target-state design
:Eng-PaymentTargetState a easvc:ServiceEngagement ;
skos:prefLabel "Payment Platform Target State Design"@en ;
easvc:engagementStatus easvcrd:Closed ;
easvc:assignedArchitect :SeniorArchitect ;
easvc:engagementFor :PaymentPlatform ;
easvc:hasOutcome :Outcome-PaymentTargetState .
:Outcome-PaymentTargetState a easvc:ServiceOutcome ;
easvc:influencedDecision :ADR-PaymentMigration ;
easvc:outcomeDescription "Produced target-state architecture and influenced migration decision."@en .
## The decision with cost-informed options
:ADR-PaymentMigration a ad:Decision ;
skos:prefLabel "ADR-042: Migrate Payment Platform to AWS"@en ;
ad:hasOption :Option-AWS, :Option-Replatform, :Option-Tolerate ;
ad:hasSelectedOption :Option-AWS .
:Option-AWS a ad:Option ;
skos:prefLabel "Migrate to AWS EKS"@en ;
fina:hasCostBenefitAnalysis [
a fina:CostBenefitAnalysis ;
fina:estimatedCost 650000.00 ;
fina:estimatedBenefit 1100000.00 ;
fina:netPresentValue 320000.00 ;
fina:discountRate 0.08 ;
fina:analysisHorizonYears 3 ;
fina:cbaCurrency "EUR" ;
] .
:Option-Replatform a ad:Option ;
skos:prefLabel "Replatform on-premises"@en ;
fina:hasCostBenefitAnalysis [
a fina:CostBenefitAnalysis ;
fina:estimatedCost 180000.00 ;
fina:estimatedBenefit 340000.00 ;
fina:netPresentValue 110000.00 ;
fina:discountRate 0.08 ;
fina:analysisHorizonYears 3 ;
fina:cbaCurrency "EUR" ;
] .
## Investment case for the selected option
:PaymentPlatform fina:hasInvestmentCase [
a fina:InvestmentCase ;
skos:prefLabel "Payment Platform Cloud Migration"@en ;
fina:investmentType finard:TransformInvestment ;
fina:investmentAmount 650000.00 ;
fina:investmentCurrency "EUR" ;
fina:projectedROI 35.0 ;
fina:paybackPeriodMonths 18 ;
fina:investmentStatus "approved" ;
] .
3.2 AI Governance with Cost Tracking¶
An AI system has both governance artifacts and a cost model — enabling queries that combine compliance status with financial impact.
@prefix aigov: <https://meta.linked.archi/ai-governance/onto#> .
@prefix aigovrd: <https://meta.linked.archi/ai-governance/reference-data#> .
@prefix fina: <https://meta.linked.archi/financial-architecture/onto#> .
@prefix finard: <https://meta.linked.archi/financial-architecture/reference-data#> .
@prefix skos: <http://www.w3.org/2004/02/skos/core#> .
@prefix : <https://model.example.com/cross-ext#> .
:CreditScoringAISystem a aigov:AISystem ;
skos:prefLabel "Credit Scoring AI System"@en ;
aigov:hasRiskClassification [
a aigov:RiskClassification ;
aigov:classifiedAs aigovrd:HighRisk ;
aigov:classificationRationale "Credit scoring — Annex III, Section 5(b)"@en ;
aigov:classificationDate "2026-01-15"^^xsd:date ;
] ;
aigov:hasConformityAssessment [
a aigov:ConformityAssessment ;
aigov:assessedAgainstPrinciple aigovrd:FairnessPrinciple ;
aigov:assessmentResult "Pass"@en ;
aigov:assessmentDate "2026-02-01"^^xsd:date ;
] ;
aigov:hasHumanOversightPlan [
a aigov:HumanOversightPlan ;
aigov:oversightMode aigovrd:HumanOnTheLoop ;
aigov:oversightResponsible :CreditRiskTeam ;
] ;
fina:hasCostModel [
a fina:CostModel ;
fina:currency "EUR" ;
fina:tcoHorizonYears 3 ;
fina:totalTCO 480000.00 ;
fina:hasCostItem [
a fina:CostItem ;
fina:costCategory finard:InfrastructureCost ;
fina:annualAmount 72000.00 ;
] , [
a fina:CostItem ;
fina:costCategory finard:ComplianceCost ;
fina:annualAmount 45000.00 ;
skos:note "Annual conformity assessment and bias audit costs"@en ;
] ;
] .
3.3 EA Service Delivery with Satisfaction and Cost¶
An EA service has a cost model (what it costs to deliver) and engagement outcomes with satisfaction scores.
@prefix easvc: <https://meta.linked.archi/ea-service/onto#> .
@prefix easvcrd: <https://meta.linked.archi/ea-service/reference-data#> .
@prefix fina: <https://meta.linked.archi/financial-architecture/onto#> .
@prefix finard: <https://meta.linked.archi/financial-architecture/reference-data#> .
@prefix skos: <http://www.w3.org/2004/02/skos/core#> .
@prefix : <https://model.example.com/cross-ext#> .
:ArchReviewService a easvc:EAService ;
skos:prefLabel "Architecture Review Service"@en ;
easvc:serviceType easvcrd:ArchitectureReview ;
easvc:requiredCapability :CapArchimateModeling, :CapStakeholderFacilitation ;
fina:hasCostModel [
a fina:CostModel ;
fina:currency "EUR" ;
fina:hasCostItem [
a fina:CostItem ;
fina:costCategory finard:LaborCost ;
fina:annualAmount 45000.00 ;
skos:note "Estimated 0.25 FTE for architecture reviews (12 reviews/year)"@en ;
] ;
] .
4. Cross-Extension SPARQL Queries¶
For each AI system: risk level, annual cost, and compliance status¶
PREFIX aigov: <https://meta.linked.archi/ai-governance/onto#>
PREFIX fina: <https://meta.linked.archi/financial-architecture/onto#>
SELECT ?system ?label ?riskLevel
(SUM(?amount) AS ?annualCost)
(COUNT(DISTINCT ?ca) AS ?assessments)
WHERE {
?system a aigov:AISystem ;
skos:prefLabel ?label ;
aigov:hasRiskClassification ?rc .
?rc aigov:classifiedAs ?riskLevel .
OPTIONAL {
?system fina:hasCostModel ?cm .
?cm fina:hasCostItem ?ci .
?ci fina:annualAmount ?amount .
}
OPTIONAL { ?system aigov:hasConformityAssessment ?ca }
}
GROUP BY ?system ?label ?riskLevel
EA service delivery cost vs satisfaction¶
PREFIX easvc: <https://meta.linked.archi/ea-service/onto#>
PREFIX fina: <https://meta.linked.archi/financial-architecture/onto#>
SELECT ?service ?serviceLabel
(SUM(?cost) AS ?annualDeliveryCost)
(AVG(?score) AS ?avgSatisfaction)
(COUNT(?eng) AS ?engagementCount)
WHERE {
?service a easvc:EAService ;
skos:prefLabel ?serviceLabel .
OPTIONAL {
?service fina:hasCostModel ?cm .
?cm fina:hasCostItem ?ci .
?ci fina:annualAmount ?cost .
}
OPTIONAL {
?service easvc:hasEngagement ?eng .
?eng easvc:hasOutcome ?outcome .
?outcome easvc:hasSatisfactionAssessment ?sat .
?sat easvc:hasSatisfactionScore ?score .
}
}
GROUP BY ?service ?serviceLabel
Which EA engagements influenced decisions with the highest NPV?¶
PREFIX easvc: <https://meta.linked.archi/ea-service/onto#>
PREFIX fina: <https://meta.linked.archi/financial-architecture/onto#>
PREFIX ad: <https://meta.linked.archi/arch-decision#>
SELECT ?engagement ?engLabel ?decision ?decLabel ?npv WHERE {
?engagement a easvc:ServiceEngagement ;
skos:prefLabel ?engLabel ;
easvc:hasOutcome ?outcome .
?outcome easvc:influencedDecision ?decision .
?decision skos:prefLabel ?decLabel ;
ad:hasSelectedOption ?option .
?option fina:hasCostBenefitAnalysis ?cba .
?cba fina:netPresentValue ?npv .
}
ORDER BY DESC(?npv)
Investment portfolio by Run/Grow/Transform with EA engagement traceability¶
PREFIX fina: <https://meta.linked.archi/financial-architecture/onto#>
PREFIX easvc: <https://meta.linked.archi/ea-service/onto#>
SELECT ?type (SUM(?amount) AS ?totalInvestment) (COUNT(?ic) AS ?count)
(COUNT(DISTINCT ?eng) AS ?eaEngagements)
WHERE {
?element fina:hasInvestmentCase ?ic .
?ic fina:investmentType ?type ;
fina:investmentAmount ?amount .
OPTIONAL {
?eng a easvc:ServiceEngagement ;
easvc:engagementFor ?element .
}
}
GROUP BY ?type
5. Validation¶
All three extensions can be validated independently:
# Syntax validation
.scripts/validate.sh --syntax extensions/ai-governance/
.scripts/validate.sh --syntax extensions/financial-architecture/
.scripts/validate.sh --syntax extensions/ea-service/
# SHACL validation (when profiles are registered)
.scripts/validate.sh --shacl ai-governance
.scripts/validate.sh --shacl financial-architecture
.scripts/validate.sh --shacl ea-service
References¶
Gartner 2025 Sources¶
- Gartner 2025 Alignment Analysis — Gap analysis that motivated these extensions
- Enterprise Ontologies — Mandatory — Practice guide on the ontology mandate
Extension Practice Guides¶
- AI Ethics & Governance Practice Guide — Detailed usage with EU AI Act alignment
- Financial Architecture Practice Guide — TCO, CBA, FinOps, TIME composition
- EA-as-a-Service Practice Guide — Service catalog, engagements, maturity
Worked Examples¶
examples/ea-service/— Acme Corp EA function (EA-as-a-Service)