Digital urbanization is the deliberate organization and governance of an enterprise digital world, treated as a territory inhabited by people, applications, data, and AI agents. The series develops that idea from my preprint Toward a Theory of Digital-World Urbanization (Rodas López, 2026), available on the paper’s page and on Zenodo under a concept DOI that always resolves to the latest version.
The digital environment of a mid-sized or large organization grows by accretion, one project, acquisition, and spreadsheet at a time, and now agents too. Each piece makes sense where it was born. Taken together, they leave you with duplicated services, inconsistent identities, and data nobody can trace. The city is my working analogy, and each urban term is translated into its enterprise equivalent the first time it appears.
What the series covers
The framework
The enterprise as an inhabited territory introduces the framework’s seven systems and Northstar Consumer Group, the fictional company that runs through the series. The forgotten tradition of information-system urbanization revisits the French-speaking school that was already zoning the information system and looks at what it lacks once cloud platforms and agents arrive.
The shape of the city
- Zoning the enterprise by business capability deals with districts, business capabilities with an explicit owner.
- Infrastructure before density: shared services asks for identity, audit, observability, and notifications to mature before applications and agents multiply.
- Point-to-point integrations are improvised roads takes on mobility (APIs, events, queues, contracts) and the cost of coordinating it.
- Governing the data subsurface no one sees starts from a margin that Northstar calculates three different ways.
What grows outside the plan
An environment that’s hard to live in pushes people to build around it, and what they build around it makes it harder still to live in.
- Shadow IT as informal settlement: map it first proposes mapping unsanctioned spreadsheets and assistants before choosing between recognition, integration, relocation, and retirement.
- Digital habitability: what architecture doesn’t measure asks whether people can find services, get access, and know whom to turn to when something breaks.
- Urban debt and renewing the city without evicting it treats legacy modernization as continuous renewal while the business keeps running.
The new inhabitants
AI agent governance: civic identity and jurisdiction proposes a civic profile for every agent (sponsor, district, scoped permissions, revocation) and applies it to Northstar’s purchasing agent, which currently runs on credentials inherited from a developer.
Using the framework and testing it
- The digital urbanization canvas: 10 diagnostic questions turns the framework into ten questions for a diagnostic workshop, with no formal evaluation yet.
- How I would know this theory is wrong lists the tests the framework could fail and which parts I’d have to drop if it fails them.
The standard the series holds itself to
The ideas in this series come from building Grupo Diveco’s corporate platform, Portal Diveco, which we call “Ciudad Diveco” internally. We’ve been building it since August 2025, and it runs in production. As of September 2026, its registry holds some fifty tools that share identity, audit, and row-level filtering, and that sit on a data lake with shared semantics, alongside AI agents. The framework was abstracted from that practice. The portal never adopted DWU’s vocabulary, and its urban language lives in the brand, such as the “Ciudadano DIVECO” role every user has received by default since August 2025. In September 2026 we also published a 3D model that draws that ecosystem as a city, with each tool as a building. It’s a communication map, curated by hand on static data.
That shows the constructs can be operationalized in a real organization. The six propositions are still hypotheses with no empirical validation, and the framework hasn’t been through peer review. Diveco is a single case, and almost everything built there predates the written framework. It’s precisely the practice the constructs came from, so it can’t be both origin and proof. I also have no comparative measurements of duplication, integration fragility, authorization incidents, or adoption. What Diveco can offer is material for a retrospective case study, if the organization authorizes one.
Northstar Consumer Group is fictional. It’s the composite case that runs through every article, and it gathers patterns from my practice that the literature already documents. It illustrates without proving cause and effect. Several articles also include the real Diveco case, and details from one case are never attributed to the other.
Every article names where its mechanism breaks down or gets expensive. Shared infrastructure can become a bottleneck; zoning can harden silos. When Diveco appears, the article also says which part of the framework the platform doesn’t implement. A framework that lists only benefits reads too much like a brochure.
The practice behind the series adds up to about three years in enterprise platform architecture. The part documented in repositories covers about fourteen months, and it’s Diveco’s. I’m both that platform’s architect and the framework’s author, with the risk of confirmation bias that implies.
Who it’s for
Enterprise architects, platform teams, and data or AI leads in mid-sized and large organizations with several platforms, legacy systems, and early agents in production. For a small company with one or two isolated systems, the framework adds little. It sits alongside TOGAF, data mesh, and zero trust, and places them inside the same territory.
How to cite
Rodas López, J. R. (2026). Toward a Theory of Digital-World Urbanization: A Conceptual Framework for Enterprise Platforms, Data, and Intelligent Agents (Version 1.0) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21523863