The question is not only who holds the keys to the digital state. It is whether the state was ever designed as a unified digital architecture.
The recent analysis by INAT titled “Who Holds the Keys to the Digital State?” has reopened a discussion that I consider absolutely necessary. Not because it belongs to a specific political space or moment, but because it touches a much deeper question: what does digital sovereignty really mean for a modern state?
The INAT analysis raises the question of whether the public sector will remain a permanent customer of technology solutions or become a state that acquires knowledge, control, and sovereignty over its critical digital infrastructure. It also refers to the need for public computing capacity, secure public data, and public procurement models that do not simply buy ready-made solutions, but create technological capability.
Many of these issues are not new to the public debate. In previous articles, long before this discussion became more visible, I have repeatedly written about similar themes: that data is not just information but infrastructure; that trust is not a matter of communication but of architecture; that compliance without real resilience creates illusions; and that a public digital project should not be judged only by whether it was delivered, but by whether it leaves behind operational capability.
I do not mention this as a matter of personal validation. I mention it because it shows something important: the problem is not occasional. It does not concern one study, one government, one project, or one public body. It is structural. And as long as the discussion about digital transformation remains limited to how many services have moved online, we miss the essential question.
The Greek public sector does not simply need more digital services. It needs a unified digital architecture.
The metaphor of digital city-states
Ancient Greece was not a unified state in the modern sense. It consisted of city-states: Athens, Sparta, Corinth, Thebes, and many others. They shared a language, a cultural world, and common references, but they had different institutions, different laws, different strategies, and different views of power and public life.
Something similar seems to have happened, unintentionally, in the Greek digital public sector.
Each ministry, each organization, each agency, each registry, and each information system has often evolved as its own territory. With its own data, its own processes, its own forms, its own technology, its own contractors, its own interfaces, its own documentation, and, in many cases, its own version of truth.
This does not mean that every such project was wrong. Nor does it mean that this happened out of bad intent. Many systems were created to serve real needs, under real pressure, with limitations of time, funding, legislation, and administrative complexity.
But the result is visible: the Greek public sector often looks less like a unified digital state and more like a confederation of digital city-states.
The problem is not that there are many “cities.” A state should not be one huge monolithic information system. That would be technically dangerous, administratively rigid, and operationally unrealistic.
Public bodies need autonomy. A hospital does not operate like an urban planning office. A municipality does not operate like the tax administration. The Cadastre does not have the same operational needs as a social insurance fund.
The problem is not autonomy.
The problem is autonomy without common architecture.
In other words, it is not wrong to have digital city-states. What is wrong is not having a common constitution, a common language, a common road network, and common technical conditions for cooperation between them.
Authentication is not integration
In recent years, an important improvement has taken place in the citizen experience: common authentication. Citizens can now access many public services using common credentials and a relatively unified access model. This matters. It solved real problems and improved everyday interaction with the state.
But here lies a critical misunderstanding.
Authentication is not interoperability.
Authentication answers the question:
Who are you?
Real interoperability answers much harder questions:
What data does the state already hold about you?
Which public body is the official source of truth?
Who informs whom when something changes?
How does a case move from one agency to another without requesting the same information again?
How do we avoid duplicate records, contradictions, and different versions of the same reality?
How can we prove who changed what, when, under which authority, and with what result?
These are the questions that determine whether a state is truly digital or merely has digital entrances to old administrative silos.
We should be accurate: Greece does not lack interoperability altogether. The Interoperability Center exists as an infrastructure for exchanging operational data between public bodies. There are also web services and data exchange mechanisms. The problem is not that nothing exists. The problem is that these connections have not yet evolved into a complete, unified operational and data architecture for the state.
It is one thing for two systems to connect when needed.
It is another thing for them to be designed from the beginning as parts of a shared institutional data infrastructure.
The Cadastre on G-Cloud: an important step, but not the end of the road
The recent announcement that the Hellenic Cadastre has fully moved to G-Cloud is a very relevant example of this discussion. According to the announcement, all applications, systems, and data of the Hellenic Cadastre have been transferred to the Government Cloud, which is the unified digital infrastructure of the Greek public sector for hosting critical information systems.
The description of the project states that the Cadastre’s digital operations are moving from old, fragmented infrastructure to a unified and secure Government Cloud. It also notes that decades of technologies, applications, and data that had previously been fragmented across different environments are now being consolidated into a modern infrastructure.
This is positive. It is a step toward greater security, availability, resilience, and operational continuity. According to the official description by the General Secretariat for Information Systems and Digital Governance, G-Cloud provides modern cloud infrastructure that can be used by government bodies to host information systems, and it is part of a broader strategy to concentrate public-sector information systems into a unified structure.
So this should not be underestimated. Moving critical infrastructure into a unified and secure government environment is necessary.
But here we need the crucial distinction.
Moving to a common infrastructure does not automatically mean a unified state architecture.
G-Cloud primarily answers the question:
Where are the systems hosted?
It does not, by itself, answer the deeper questions:
How does Cadastre data communicate with the tax authority?
How does it connect with municipalities?
How does it connect with urban planning authorities?
How does it connect with notaries, engineers, courts, property transfers, permits, spatial planning, and investment processes?
What is the common data model?
Which system is the source of truth for each critical field?
How are all related systems updated when information changes?
The Cadastre is critical national infrastructure not because it is a large database, but because it is connected to property, taxation, transactions, investment, spatial planning, justice, and the functioning of public administration.
Therefore, the move to G-Cloud is a necessary infrastructure step. But the real next step is different: making the Cadastre a reliable, interoperable, auditable, and reusable source of truth for the entire state.
That is the difference between infrastructure consolidation and state capability.
Interoperability is not a technical feature
Interoperability is often treated as a technical task. Build an API. Pass a field from one system to another. Create a connection. Open a web service.
All of these are necessary. But they are not enough.
Real interoperability is a design principle for the state.
It means shared semantics.
Common data vocabularies.
Clear sources of truth.
A common logic of identities.
Rules for updating registries.
Historical traceability.
Audit trails.
Transparent workflows.
Reusable infrastructure.
And the ability for a case to move from one public body to another without being lost in translation.
Europe has already recognized this need. The Interoperable Europe Act entered into force on April 11, 2024, with the aim of strengthening interoperability in the public sector and enabling administrations to cooperate so that public services can function seamlessly across territorial, sectoral, and organizational boundaries.
That is exactly the point: interoperability is not a luxury. It is a prerequisite for modern public administration.
The citizen as a messenger between systems
The most visible symptom of system isolation is that the citizen becomes the carrier of data between public services.
The state already knows something, but asks the citizen to prove it again.
A document already exists in one public agency, but must be uploaded again to another platform.
Information has already been entered into one registry, but must be declared again in another application.
A change is made in one system, but another system is not automatically updated.
This is not a digital state. It is digitized bureaucracy.
Old bureaucracy asked citizens to move from office to office carrying paper. New bureaucracy risks asking them to move from platform to platform carrying PDFs.
The form changed. The logic did not.
A mature digital state is not the one with more online forms. It is the one that asks less from the citizen because it already knows more in a lawful, secure, transparent, and auditable way.
From project delivery to state capability
This is the central point.
A public digital project should not be evaluated only by whether it was delivered, whether it works, or whether it serves a number of users. It should be evaluated by whether it strengthens the overall capability of the state.
Does it leave behind shared data?
Does it leave documentation?
Does it leave APIs?
Does it leave clear data contracts?
Does it enable reuse?
Does it provide mechanisms for auditability?
Does it transfer knowledge to public bodies?
Does it allow the state to change contractors?
Does it leave something on which the next project can build?
If the answer is no, then the project may be useful, but it is not strategic. It solves one problem while creating a new dependency.
The deliverable is not the platform.
The deliverable is the capability that remains when the project is over.
We do not need a super-system
The solution is not to centralize everything into one gigantic system. That would be a mistake. It would create technical risk, administrative rigidity, and a new form of centralized dependency.
The right model is more mature: a federated digital architecture.
Each public body keeps its operational autonomy, but works within a shared framework:
common identities,
common data vocabularies,
common interoperability standards,
clear sources of truth,
clean APIs,
rules for updating registries,
mandatory documentation,
traceability,
auditability,
data portability,
and the ability to reuse infrastructure.
Autonomy without interoperability becomes isolation.
Decentralization without common architecture becomes fragmentation.
Digitization without shared data becomes digital bureaucracy.
The next stage
Greece has made important progress in digital public services. That should be acknowledged. But the next stage is more difficult and less impressive from a communication point of view.
It is not only about more applications.
It is not only about more online services.
It is not only about better portals.
It is not only about faster authentication.
It is about moving from the digitization of services to the architectural maturity of the state.
The question is no longer whether citizens can submit an application online. The question is whether the state can understand, process, cross-check, document, and complete that case without turning the citizen into a messenger between isolated systems.
The question is not whether we have digital doors.
The question is whether behind those doors there is a unified state or many digital city-states.
That is the real challenge.
The Greek public sector does not simply need more technology. It needs better architectural thinking. It needs to design digital projects not as isolated implementations, but as parts of a shared institutional infrastructure.
Because the digital state is not the sum of its platforms.
It is its ability to operate as a coherent, reliable, interoperable, and auditable system of public value.
Otherwise, we will have many digital city-states.
But not a truly digital state.
