Civic information architecture
DSarchitek Inc. develops data governance, technology platform, innovation and modernisation strategies for the entire citizen journey that government acts through. The work is done by senior-level consultants, start to finish, and produces custom deliverables tailored to each client's needs and circumstances, not generic templated guidance.
What civic information architecture is
Civic information architecture is the design of the data platforms and systems government acts through. Registries record who owns land, who owns a vehicle, who was born and who has died. Identifier systems decide how government recognises a person. Policies decide who may see personal information and what AI tools may do with it.
These choices last a long time. A registry built today will serve governments decades from now, facing conditions no one can foresee. The practice provides solutions for that full lifespan. It asks what the system will allow, what it will protect, who will inherit it, and who the citizens it serves are.
Four questions, one design
Consulting usually splits public-sector data work into separate specialties, each handled by a different expert. The practice treats four of them as parts of one design question, because an answer to any one of them changes the others.
The four registers the practice designs together.
- Technical
- How the data is structured, which records are authoritative, and where information can flow.
- Constitutional
- What the state may know about a person, what it may act on, and what the person can take back.
- Political
- What the system lets future governments do, what it rules out, and what it keeps open to audit.
- Service delivery
- How a person experiences government across life events such as a birth, a land purchase or an estate settlement.
Designing the four together avoids two common failures: systems that work technically but treat people badly, and systems that meet the law on paper but cannot be audited in practice.
Stewardship, not only efficiency
Much data work asks how to make today's process faster or cheaper. The practice asks a longer question: what is the institution building, who will inherit it, and will it still hold up in thirty years? The different questions lead to different paths: different designs, models and strategies. DSarchitek's practice answers the second question, which provides greater value over a longer period of time.
Problems the practice works on
Each area below has its own page with what the practice does and the work behind it.
Registry and master data modernisation
Moving registries off decades-old mainframe systems while every record keeps its legal meaning. Building one reliable record of each person across programs that each kept their own.
Citizen identifier and digital identity architecture
Identity projects often mix up three separate questions. Which record is the person's official one (the canonical identifier)? How does the person prove who they are (the authentication identifier)? What does the person control themselves (citizen-controlled identity)? The practice designs an answer to each question separately, so a technology choice does not quietly decide all three.
AI deployment policy
The practice designs AI policy around one rule: AI tools do not get direct access to secured personal information. The rule then shapes how AI is implemented across government data sources, how vendor claims are tested, and how data use is overseen.
Data across jurisdictions
Federal, provincial, Indigenous and municipal governments describe the same people and places in different ways. The practice reconciles those differences in the design, not by patching them later. Indigenous data governance, including the OCAP and CARE principles, is a starting input to the design options when required, rather than an add-on.
Sovereign data and cloud infrastructure
Data residency and data sovereignty are not the same thing. Residency means the data is stored in Canada. Sovereignty means it stays under Canadian legal control. A Canadian cloud region run by a foreign provider can still be reached by foreign law, such as the US CLOUD Act, so residency alone does not keep public data under Canadian authority.
The practice designs cloud and data platform strategies and solutions for public bodies that treat that exposure as an architecture and contract risk to be managed in the design. That means deciding which workloads may run where, what the vendor contract must say, and how keys, backups and administrative access stay under the government's control.
Sovereign data centre enclave designs
Some government work cannot share infrastructure with anyone: sensitive personal records, security and defence workloads, and the AI systems trained on them. An enclave is a separated part of a data centre, with its own physical, network and personnel controls, built to hold that work. Demand for enclaves is growing as governments bring AI into services that handle protected information.
The practice designs these enclaves and the path to certifying them. It maps federal security standards, including ITSG-33 and the Protected B, Medium Integrity, Medium Availability (PBMM) profile, TEMPEST, COMSEC and the Canadian Program for Cyber Security Certification, into a phased certification roadmap, so a facility can take on more sensitive workloads as each phase is met. Where the facility is Indigenous-led, the governance design keeps that control in place as the facility grows.
Long-horizon policy documents
Frameworks, design principles and procurement guidance written to be read, audited and revised by people who were not in the room when they were written.
What the practice does not do
The practice does not manage delivery projects, run technology selection bake-offs, provide general training, or staff large teams. It does not certify anyone's compliance. When a problem needs those capabilities, the practice says so early and, where it can, points to a firm that does that work.
How engagements work
Engagements are tailored and focused on the core problem, with senior decision-makers. Typical examples are a ministry choosing its master data architecture, a government setting citizen identifier strategy, a public body writing its AI deployment policy, or a registry planning a long modernisation.
You work with the consultant directly, from the first conversation to the final document. There is no hand-off to a delivery team. That keeps the cost down and keeps the person who understands your problem on it throughout.
Read the principal's record for the work behind this, or go to the Engage page. The first conversation settles whether there is a fit.