Environmental Grounding Dossier
Six environment-grounded requirements are derived from the housing access and representation problems diagnosed in Chapter 2 and Chapter 3. Each requirement card records the problem observed in practice and the affected stakeholder groups. Consequence, cost dimension, derived requirement, and acceptance criterion are declared for each entry. Requirements established here provide the evaluative basis for Chapter 4 and the demonstration cases in Chapter 10.
Requirement ID | Problem Statement | Stakeholder Group | Consequence | Cost Dimension | Derived Requirement | Acceptance Criterion | Confidence |
— | — | — | — | — | — | — | — |
ER-01 | Diachronic needs are managed by synchronous artefacts and approvals | Households, disability residents, certifiers | Adaptations trigger repeated reinterpretation of prior intent | Time | Representations must carry change-validity conditions across handovers | Change event can be evaluated through declared checks without full artefact rediscovery | High |
ER-02 | Meaning decays across handovers in monolithic representations | Designers, builders, compliance reviewers | Verification depends on tacit reconstruction | Cognitive | Constraints and invariants must be explicit, queryable, and portable | At least one formal invariant set is bound to each change-relevant interface | High |
ER-03 | Late discovery of constraint conflicts increases adaptation burden | Providers, project owners, occupants | Rework and delay accumulate after commitments are sunk | Real | Requirement checks must run earlier and be replayable after change | Baseline vs artefact workflow reports reduced reconstruction effort per case | Medium |
ER-04 | Accessibility intent is not reliably transportable in delivery chains | SDA participants, assessors, policy actors | Feature-level compliance can still fail integrated usability | Physical | Semantics must be encoded as first-class objects, not only geometry annotations | Demonstration cases show route-level and relation-level checks, not only dimensional checklists | Medium |
ER-05 | Hidden coupling expands verification scope for local edits | Multi-actor delivery teams | Small changes propagate into global verification debt | Skill | Interfaces must declare degrees of freedom, invariants, and trigger checks | Each local transformation log includes a finite triggered-check set | High |
ER-06 | Platform governance fails when complements evolve without shared rules | Governance teams, toolchain maintainers | Drift, incompatibility, and non-comparable revisions | Time | The governed kernel and the governed instance library must be explicitly versioned and auditable | Evaluation workbench records versioned rule compliance and exception budgets | Medium |
Evidence grounding for each requirement is provided by Chapter 2 (Sections 2.3–2.6) and Chapter 3 (Sections 3.2–3.4). Full traceability is documented in Appendix: Requirements-Design-Evaluation Traceability Matrix. Six cost dimensions are covered: time, cognitive, real, physical, skill, and governance. Together, the requirements define a complete problem specification against which the thesis artefacts are evaluated. Therefore, this dossier establishes the evidentiary basis for the evaluation framework in Chapter 4.