Why it matters:Digital systems have become public infrastructure. If they cannot be fixed, inspected, and handed over, they become another form of hidden debt.
Public infrastructure is usually imagined as concrete, rails, and pipes. Yet a city now runs on ticketing systems, benefit portals, school platforms, and maintenance databases. When these systems fail, the consequences are just as practical as a broken road: people cannot pay, apply, register, or appeal.
Digital infrastructure is often governed by procurement rules designed for one-off purchases. A supplier delivers a platform, signs a support contract, and keeps the inner workings proprietary. For a time, the service looks efficient. Then the supplier changes terms, the contract ends, or the software becomes too expensive to replace.
Repair begins with access
A repairable digital system has clear interfaces, documented data, and a route for another team to take over. Its source code may not need to be public in every case, but public bodies should own enough of the design and data to avoid being trapped.
This is not a philosophical preference. It is continuity planning. A public service that cannot be inspected or changed without the original vendor becomes fragile even when it appears stable. The risk is hidden in renewal dates and contract terms rather than visible on the street.
Procurement can create or dissolve lock-in
Contracts can require open standards, exportable data, and the right to maintain the software through a third party. They can also specify that documentation and test suites are part of the deliverable. These requirements need to be present from the beginning; they are difficult to retrofit after the system is embedded.
Smaller public agencies often lack the internal capacity to evaluate such clauses. Shared digital teams can help by providing common contracts, security reviews, and technical support. The alternative is that each town negotiates the same problem alone.
Legacy systems are not only old
A system can be new and still become legacy if it is hard to update. The age of the code is less important than the ability to understand it. Good public digital infrastructure is designed for succession: another team can read the documentation, run the tests, and make a change without reverse-engineering the entire service.
Repairability is not a technical nicety. It is the difference between owning a service and renting access to it.
Make maintenance visible
Governments often receive more credit for launching a service than for keeping it reliable. Budgets reflect that imbalance. Maintenance is treated as an operating cost that can be squeezed, while new projects attract announcements.
A more mature approach publishes service health, tracks unresolved defects, and funds staff who maintain the system over time. It also gives the public a place to report failures. A repairable public system is not one that never breaks. It is one that can be fixed before trust erodes.
Reference desk
Sources and further reading
Fieldwork publishes explanatory journalism and independent analysis. This story follows our editorial standards.
Updated throughout the editorial cycle.
