Prove the model on one willing project. Then replicate it.
Every major project should leave Canada two assets: the infrastructure itself, and a secure record of how it was built, operated, and improved. This proposal is for a voluntary pilot program that proves the model project by project. No mandate is proposed or sought.
Status: a public idea, waiting for its proponent
This paper is published openly as the instrument for finding its proponent and funder. Those are the two gating items, stated plainly: nothing starts without a named, willing proponent with written interest, and a funder for the design phase. The paper's job is to be complete and honest enough that a real door can carry it into the right room. Ideas put out openly wait for their moment; this one is waiting for its project.
Why the proponent says yes
A voluntary program lives or dies on the proponent's incentive:
- Better project controls. Standardized outcome-linked records can improve construction delivery: faster root-cause analysis on defects and rework, clearer inspection evidence, and better tracking of corrective actions.
- What the schema adds over internal controls. The proponent's project controls serve the proponent. The common schema serves the record that outlives the project team: interoperable across contractors (no more fragmented proprietary systems), comparable across projects (benchmarks instead of anecdotes), and verifiable by third parties (regulators, insurers, future operators). Internal controls optimize one project; the schema compounds across all of them.
- Safety and regulatory value. A verified build record strengthens safety cases and regulatory relationships. Demonstrated diligence is an asset in every future approval.
- First-mover positioning. The pilot proponent helps write the standard others will follow. Standards written by operators fit operations; standards written without them don't.
- Optional licensing upside. Rights-cleared subsets of the record can be licensed. Upside, not the case for participating.
- Cost discipline. The first deliverable is a bottom-up cost estimate signed off by the proponent's own team. If the number doesn't work, the proponent walks away having spent only the design phase.
The pilot: North Coast Transmission Line (reference design)
Phase 1 is a 165-kilometre, 500-kilovolt transmission segment in British Columbia. Federal and provincial commitments total $3.9 billion for phases 1 and 2. First Nations may co-own up to 50 percent. Construction is active.
North Coast is the reference design — the project this proposal is scoped around. A proponent building at similar scale with similar data richness could substitute; the design travels.
Proponent: [to be named]. Funder: [to be named]. See Status above.
Why this project fits:
- The data already exists. Transmission construction generates terrain and weather records, construction sequencing, equipment telemetry, geotechnical results, inspection evidence, deviation and rework logs, vegetation management records, and commissioning results. The pilot standardizes what's already produced.
- Honest timeline. Governance agreement comes first, and the project is already building. Early-phase capture will be partially retroactive — assembled from existing project records — with live capture staged from agreement forward.
- Indigenous co-ownership is the governance. With a First Nations co-ownership opportunity of up to half the project, co-designed data governance under OCAP principles is the critical path. Timeline risk is accepted: the clock starts at governance agreement, not announcement. A workable model here would offer evidence for adaptation elsewhere—not a substitute for project-specific governance.
Legal pathway
The proposed voluntary pathway does not depend on a new statute. Outside counsel must confirm the authority and terms for the chosen project. Potential pilot instruments:
- Project agreement. A data-sharing and stewardship schedule negotiated directly with the proponent as a commercial term.
- Funding conditions. Where federal money flows, contribution agreements can carry proportionate data-deposit terms for the funded scope.
- Procurement. Where the Crown procures, contracts can specify data deliverables.
Limits stated plainly: these do not compel unwilling proponents or override provincial jurisdiction. Funding and Crown-procurement terms apply only within their funded or contracted scope; a purely private project would participate through a voluntary project agreement. The program scales by replication, not compulsion.
What capture costs
First deliverable, before data flows: a bottom-up estimate signed off by the proponent's own project-controls team. Categories: standardization (one-time schema mapping, reusable), incremental capture (gaps only, minimum necessary), stewardship (shared repository operation), governance (front-loaded co-design and rights clearance).
Working target: incremental capture cost well under 1% of project capital. If the estimate misses, that is a finding.
What the repository holds
The project record — design revisions and machine-readable as-builts; work packages, sequence, and site conditions; equipment telemetry and maintenance events; material batches, tests, and quality control; environmental measurements and mitigations.
The outcome layer, in two phases:
- Construction-phase outcomes (observable in-window): acceptance, defect, rework, and failure labels; inspection findings tied to the underlying task; safety events and corrective actions. These are real verified outcomes, and the pilot can demonstrate them within 18 months.
- Operational-phase outcomes (schedule-dependent): commissioning results, operating performance, longitudinal condition and maintenance. B.C.’s September 2026 update places Phase 1 completion at mid-2030, beyond an 18-month pilot starting soon. The pilot therefore demonstrates the construction-phase layer and designs for the operational layer; commissioning and longer-term performance are later-stage outcomes, with the schedule confirmed by the proponent.
The valuable unit is not volume. It is an event linked to a verified outcome.
Minimum-necessary collection: machine, process, and asset data by default. Identifiable worker data only where a defined use justifies it and consent and labour rules permit.
Governance
Rights and custody. Proponent retains background IP and protected commercial information. Repository receives a durable licence for preservation, regulatory use, and approved research. Commercial reuse only where every relevant right is cleared.
Indigenous data sovereignty. OCAP principles, co-designed from day one. Schema must accommodate Nation-specific access tiers without breaking interoperability.
Security and access. Canadian-controlled infrastructure. Tiered access: public, research, commercial, restricted, protected. Audit logs, purpose limits, revocable access.
Commercialization. License curated, rights-cleared datasets. Revenue first funds stewardship, then flows transparently to contributors, rights holders, Indigenous partners, and the repository.
Who operates this
Stated plainly, since a funder will ask:
- Design phase (months 0–6): proposed to be operated by Archive Signal, with two independent co-leads to be appointed. Indigenous governance co-design is facilitated by a neutral facilitator (to be named — not Archive Signal, which has a commercial interest in the outcome). Legal-pathway confirmation is led by outside counsel (to be named), not by the broker. Archive Signal leads schema mapping and the capture-cost estimate, the two deliverables that fit a data firm. Fixed scope; budget capped at $500K, final figure set by proponent costing.
- Repository (long-term): held by an independent steward, not Archive Signal. Candidates — a Canadian university, the National Research Council, or a purpose-built nonprofit — are unapproached at this stage; listing them is illustrative, not a commitment. The steward recommendation itself gets an independent check: the operator does not pick its own handover target without one.
- The common template. During the design phase, the participants will develop and openly publish a common template for project records. The proposal is for an open standard, not one controlled by a single firm.
The split is deliberate: Archive Signal moves the pilot from idea to evidence, then hands the asset to a neutral steward. The disclosure stands — a public standard benefits firms in the data layer including Archive Signal — but the repository itself is not ours, and the deliverables that need neutrality get it.
18-month pilot: replicate or retire
- Months 0–6 (from governance agreement): proponent agreement signed; cost estimate published; first common record template developed and openly published.
- Months 6–12: staged ingest (retroactive assembly + live capture); validation; first research access.
- Months 12–18: results published; replicate/retire decision in the open.
Decision thresholds (all must be met to replicate):
- Incremental capture cost under 1% of project capital (proponent-verified).
- Majority of ingested records linked to verified construction-phase outcomes.
- At least one demonstrated construction-phase use case: defect or rework root-cause analysis, inspection traceability, or safety corrective-action tracking, with value attributable to the record.
- Rights-clearance and access turnaround within published service levels.
- A second proponent expressing written interest in replicating.
Decision requested: fund the 6-month design phase — governance co-design, legal confirmation, capture-cost estimate — at an order-of-magnitude $300K–$500K (illustrative; final figure set by proponent costing). Funder to be named.
Precedent
Norway's Diskos National Data Repository (since 1995) proves structured submission, metadata standards, confidentiality windows, and tiered access can coexist with commercial rights. The petroleum enforcement model does not transfer — this proposal is voluntary where Diskos is compulsory — but the repository mechanics are working proof the design is sound.
Drew Clement is the founder of Archive Signal, which sources and brokers proprietary training-data corpora for AI builders. The author's commercial position is stated openly: a public data standard and repository would benefit firms operating in the data layer, including his own.
- Natural Resources Canada: North Coast project commitments and co-ownership opportunity.
- BC Hydro: project scope; B.C. September 2026 update: construction and mid-2030 Phase 1 completion.
- Norwegian Offshore Directorate: Diskos history and well reporting guidance.
- FNIGC: First Nations principles of OCAP®. Nation-specific governance is to be co-designed, not presumed.
Budgets, roles, thresholds and the pilot schedule are proposed design parameters, not funded commitments. Legal feasibility requires project-specific confirmation by outside counsel. No proponent, funder, independent co-lead or repository steward is appointed.
