Your National eID Works Beautifully. Inside One Country's Portal

These systems are not bad. Most of them are excellent, and they solve the hard part, real identity verification at national scale, better than anything in this industry. The limitation is not quality. It is where the proof stops working.

This page makes no claim about Solidus at all. It is about a structural pattern in somebody else's infrastructure, and our own position is stated separately, unflatteringly, and in detail.

The pattern, across dozens of national systems

Read enough national eID briefs and the same architecture appears almost everywhere:

• Centralized. One authority operates the identity, the directory and the trust decision.
• Portal-scoped. The credential is a login to a government estate. Its authority is real and its perimeter is the portal.
• Built on X.509 and PKI. Qualified certificates, smart cards, SIM-anchored signatures, mature, audited, decades-proven technology.
• No verifiable-credential surface. Typically no W3C decentralized identifier, no verifiable credential a holder can carry off-portal and present somewhere else.

Every one of those is a reasonable engineering choice for the problem as it was posed: prove a citizen's identity to their own government's services, reliably, at scale, for decades.

What that architecture genuinely achieves

A level of assurance nothing in decentralized identity comes close to. The state can bind an identity to a person because the state issued the underlying documents, runs the registry, and can compel correction. No cryptographic scheme substitutes for that.

And operational maturity. These systems have run for years, survived audits, handled fraud, and been used by populations rather than pilots. That is not a starting point anyone else in this field has.

Where it stops

At the edge of the estate that issued it. A national eID proves who you are to that country's services. Take the same identity to:

• a private company, usually no path, or a bespoke integration per company;
• another country, mutual recognition is a treaty question, not a technical one;
• an offline or cross-sector context, often nothing at all.

The proof does not travel, because it was never designed to. The architecture assumes the verifier is inside the same estate as the issuer, and inside that assumption it is close to optimal.

What decentralized identifiers change, and what they do not

The change is one thing only: the verifier no longer has to be inside the issuer's estate. A credential carries its own proof, a signature checkable against a public key, so a party with no relationship to the issuer can verify it.

What that does not change:

• It does not create assurance. A credential is only as good as the issuer behind it, and the strongest issuers are exactly these national systems.
• It does not answer who to trust. That is a governance question no format solves.
• It does not make a national eID obsolete. The interesting future is the two composing, a state-grade identity becoming a portable credential, not one replacing the other.

Why the European framework matters more than the technology

A self-issued identifier is worth nothing until a relying party trusts the issuer, and no amount of cryptography manufactures that trust.

eIDAS 2.0 manufactures it for more than 450 million people, on a December 2026 deadline, which is why the interesting question in this field is not "DID or eID" but "when do state-grade identities become portable, and who is allowed to carry them." Our position on that framework, including what we do not hold.

What a researcher should take from this

1. Do not read "centralized" as a criticism. For national identity it is a design that works.
2. Read "portal-scoped" as the actual constraint. That is the thing decentralized identifiers address, and the only thing.
3. Assurance and portability are separate properties, and most arguments in this field confuse them.
4. The strongest possible outcome is composition, not replacement, and it is a policy outcome before it is a technical one.

Why there are no numbers on this page

Our briefs carry user counts, service counts and adoption figures for these systems. We have not verified them, so we are not republishing them. Being wrong about a national service is worse than saying nothing, and the structural point does not need a scoreboard.