Why we didn't go through ICANN

The question we kept getting asked

Before we built anything, before we had a single address registered or a single line of smart contract code deployed, the question that came up in almost every early conversation was the same one: “Why didn’t you just go through ICANN?”

It’s a fair question. ICANN is the organisation that runs the global domain name system. It’s where .com lives, where .au lives, where .nyc and .london and every other recognised extension on the internet is born and sanctioned. If you want a piece of internet real estate that plays nicely with the world’s browsers and search engines, the conventional wisdom says you start there. You file your application. You wait. You pay. You comply. You operate.

We looked at that path carefully. We sat with it. We understood it. And then we walked away from it — not because it was hard or expensive, though it is both of those things, but because it was philosophically incompatible with what we were trying to build.

This post is our attempt to explain exactly why. Not as a critique of ICANN for the sake of it, and not as a defence of blockchain for the sake of that either. But as an honest account of the reasoning behind a decision that shaped everything else about Queensland Foundation.

What ICANN actually is

Most people who own a domain name have never thought about ICANN. That invisibility is, in a sense, a testament to how well the system works technically. You type an address into a browser and something appears. The plumbing that makes that happen — the infrastructure that converts a human-readable name into a machine-readable IP address — functions reliably in the background, and most users never need to consider who built it or who controls it.

ICANN was founded in 1998 as a nonprofit public benefit corporation. Its mission directs it to coordinate the allocation and assignment of names in the root zone of the Domain Name System. That sounds dry, but what it means in practice is that ICANN decides what top-level domains exist on the internet. It decides who can operate those domains, who can sell registrations under them, and under what conditions that authority can be revoked. It is, in the plainest terms, the gatekeeper of the global namespace.

Managing the domain name system is critical to the functioning of the internet. DNS servers provide automated lookup services, translating domain names into computer-readable addresses so that information will flow to and from the correct places. The internet as we know it depends on this translation layer, and ICANN sits at the top of it. That position gives ICANN — and the entities that hold agreements with it — enormous structural power over how digital identity is assigned, controlled, and maintained.

It is worth pausing on that last word: maintained. Because the traditional domain model is not a model of ownership. It is a model of maintenance. When you register a domain name through any conventional registrar — any of the household-name companies that sell .com or .com.au addresses — you are not buying that name. When you purchase a website domain, you essentially rent it. In plain English: you don’t really own it. You do this through an intermediary domain registration and management service. If you forget to pay the domain renewal fee, you relinquish your ownership rights.

That is the foundational premise of the entire traditional DNS model. Lease, not own. Rent, not hold. Permission, not right.

The layered machine you have to navigate

To understand why the ICANN path didn’t fit our mission, it helps to understand just how many layers the traditional system contains — and how many parties sit between an idea and a registered domain name.

At the top sits ICANN. Below it are registry operators: the companies that actually run the infrastructure for a given TLD. A registry is a database of all second-level domains under a top-level domain along with their full registrant information. A registry operator maintains the registry data, manages the domain registration process, and controls policies and registration requirements related to that extension.

Below the registries sit the registrars — the companies most people interact with when they buy a domain. These are retailers that sell domain names and act as the middlemen between registries and domain owners. And below them, sometimes, sit resellers: another layer of intermediary between the person who wants a name and the infrastructure that records that name.

Every layer in this stack adds cost, adds dependency, and adds a point of potential failure or policy change. Beyond ICANN, registrars also incur costs related to registry access. Every TLD is operated by a registry, and each registry has its own wholesale pricing, accreditation process, and compliance requirements. The result is a billing and administrative system of significant complexity. Fees vary depending on registry agreements and the structure of each TLD, meaning that registrars need to manage complex billing systems to handle multiple fee types across different domain extensions.

If we had chosen to operate inside this system — to apply for .queensland or .qld as ICANN-sanctioned gTLDs — we would have been entering every one of these layers simultaneously. We would have become registry operators. We would have needed accredited registrars. We would have been subject to ongoing ICANN oversight, compliance requirements, and contractual relationships that extend indefinitely and can be amended unilaterally by the body that holds the authority.

That is a significant operational commitment. But even that is not the real reason we chose a different path. The real reason runs deeper.

What it costs to apply for a TLD

Let’s talk plainly about the financial reality of pursuing an ICANN-sanctioned TLD, because the numbers are instructive in what they reveal about who this system is designed to serve.

The evaluation fee for a new gTLD application is USD $227,000. That is the floor — the single, non-refundable fee just to submit an application for one string. Each application requires $227,000 in ICANN fees, with no volume discount. We hold six TLDs: .queensland, .qld, .brisbane, .surfersparadise, .gold-coast, and .brisbane2032. Under the ICANN model, that would be six separate applications, each requiring that base fee — before a single domain had been registered, before a single Queenslander had been served, before we had done anything meaningful at all.

The application fee is only the beginning. These stack on top of the base fee, and they don’t include legal counsel, internal resource allocation, or ongoing registry operational costs post-delegation. Once a TLD is awarded, you also take on the operational costs of running a registry. Operating a gTLD is a technology-based service comprised of many requirements and expenses, such as acquiring equipment, establishing operations, hiring and training staff, and providing customer service.

For a generic gTLD, breakeven typically occurs between 50,000 and 100,000 registered domains, several years after launch. Sunk costs after five years of operation can reach $750,000 to $850,000.

There’s more. ICANN requires that registrars maintain Commercial General Liability Insurance coverage of at least $500,000. This coverage must be maintained in force throughout the term of accreditation. An applicant must demonstrate that it has adequate working capital available for operation. For applicants seeking initial accreditation, demonstration of the ability to procure liquid capital immediately available in the amount of US$70,000 or more will be deemed adequate.

And then there are the ongoing fees once you’re operating. Once your application is approved and you’re invited to sign the Registrar Accreditation Agreement, ICANN will require full payment of an annual fixed fee of $4,000 USD. On top of that, there is a transaction fee, which is a per-domain fee for domain name registrations, renewals, and transfers under certain gTLDs.

Now consider what this economic model demands of the people at the end of the chain — the actual humans who just want a name. Because every layer of cost in this system flows downstream. The registry charges the registrar wholesale. The registrar marks it up for retail. The registrant pays the retail price. And then they pay it again. Every year. Forever. The registry can increase the price by a certain percentage each year.

We want to charge $5, paid once, with no annual fees ever. That ambition is not compatible with a system built on recurring cost recovery at every level. The maths simply doesn’t work. But more importantly, the philosophy doesn’t work either.

The renewal model is a feature, not a bug — for the system

Here is something important to understand about the renewal model: it is not incidental to the traditional domain system. It is the engine of it.

Renewal fees are how registrars generate revenue. Most registries charge a wholesale price per domain, and some may also require upfront deposits, integration fees, or monthly minimums. The reason every .com renews annually, the reason you receive emails reminding you to pay your registration fee, the reason the domain industry sustains itself as a multi-billion dollar sector — all of it traces back to the fundamental model that a domain name is a leased asset, not an owned one.

High renewal costs are an intrinsic part of traditional domains. You don’t truly own a domain through a registrar — you lease it. This is not a flaw that ICANN or registrars are trying to fix. It is the revenue model. It is the business.

And that model creates a particular kind of relationship between a person and their address online. Your name on the internet — the place where your identity, your business, your presence lives — is conditional. It is conditional on your continued payment. It is conditional on your registrar remaining solvent and compliant. It is conditional on the registry continuing to operate under the terms ICANN allows. It is conditional on ICANN itself not revoking, reassigning, or restructuring the TLD.

When a registry is established, it essentially is a monopoly for a particular domain name. Because of conflict of interest issues, ICANN policy restricts gTLD registries from selling domain names directly through a retail channel. This is where the accredited registrar system comes into play.

So we have a system where a monopoly registry sells wholesale to accredited middlemen, who sell retail to registrants, who must pay every year to maintain their right to an address they can never truly hold. Every participant in that chain extracts value from the transaction. The person at the bottom — the individual who just wants their name — pays for all of it, perpetually.

We were not willing to replicate that for Queensland.

The question of control

There is a dimension to all of this that is more political than financial. It is worth naming directly, because it shaped our thinking as much as the economics did.

As the internet and World Wide Web became a global force, critics felt that the United States had too much control over ICANN. That debate has been ongoing for decades. ICANN executes its tasks through stewardship of the authoritative root zone file, which is “the global address book for data” that contains a definitive list of IP addresses and top-level domains.

Whoever controls that root zone file controls what exists on the internet. TLDs that are not in it are, from the perspective of the traditional web, invisible. This is why the root is so consequential. And it is why the question of who controls the root is never merely technical — it is deeply political.

Foreign governments have resented this control because top-level domain names are worth billions of dollars and have significant political salience. Control over domain names entails the potential for censorship, by regulating domain name use on a global basis.

During 2014, France publicly criticised ICANN, calling them an unfit venue for internet governance, and argued that alternatives should be sought. From almost the instant it was founded, ICANN came under fire from many quarters of the internet community. Activists questioned the way the board was picked and its decision to meet behind closed doors. The board’s subsequent decisions to charge fees on every domain to fund its operations further enraged activists, who worried that ICANN was moving beyond its technical management mandate to more broadly regulate the internet.

We are not making a geopolitical argument here. We are not at war with ICANN or the DNS. But we do recognise that when you operate inside the ICANN system, you operate inside a structure of permissions. The addresses you control exist because a centralised authority has decided they should exist. And that authority can, under certain circumstances, decide they should not.

For us, building permanent digital identity for Queenslanders — identity that should outlast any single organisation, administration, or contractual arrangement — that kind of dependency is a structural problem, not an administrative inconvenience.

What it means to operate outside the traditional domain system

People ask us whether being outside ICANN is a limitation. Whether our TLDs are “real.” Whether addresses on Queensland Foundation are somehow lesser than a .com.

We want to answer that question honestly, because it deserves a real answer and not a deflection.

In the traditional sense — the sense of a browser resolving an address through the global DNS without any additional configuration — yes, onchain TLDs operate differently. Traditional DNS does not integrate natively with blockchain apps or wallets. The web as it was built in the 1990s runs on infrastructure that predates the blockchain by decades, and that infrastructure was not designed to accommodate permanent, owned, immutable addresses. It was designed to accommodate rented, managed, revocable addresses.

So when someone asks whether our addresses “work” in that system, the honest answer is: they work differently. They are a different kind of thing. They are not a lesser version of a .com. They are a different infrastructure with different properties and different guarantees.

Unlike traditional domains, which users rent annually through centralized registrars, blockchain domains function as permanent, on-chain assets, fully eliminating yearly renewal costs. Once purchased, they belong to the owner indefinitely — no recurring fees, no risk of expiration, and no intermediary controlling access.

That last phrase is the one that matters most to us. No intermediary controlling access. Your Queensland address is yours because the blockchain says it is yours. Not because a registrar is holding it on your behalf. Not because ICANN has authorised the registry to authorise the registrar to record your name. Because a smart contract on an immutable public ledger records your ownership, and no single entity — not us, not any government, not any company — can unilaterally revoke that.

Blockchain domains are typically owned by users rather than leased from a registrar. This enables users to fully control their domain names, including selling or transferring them to other parties.

We think that distinction matters enormously, especially when we are talking about identity. Identity is not a subscription. Your name should not be something that lapses, that gets recycled and resold to someone else because you forgot to renew it. Your digital address should be as permanent as a street address. You buy it once. You hold it for life. It can be transferred — like property — but it cannot be taken from you by an administrative failure or a missed payment.

On the word “permanent”

We use the word permanent carefully and deliberately, because it is a word the traditional domain industry cannot use.

Missing renewal deadlines in the traditional DNS can result in permanent loss of valuable domains, particularly for short names or those with established reputations. That is an extraordinary thing to sit with. The name you have built your business around, your professional identity, potentially decades of accumulated link authority and recognition — all of it can vanish because a credit card expired or a renewal email landed in spam. This happens to individuals, to small businesses, to community organisations, to institutions. It is not rare. It is a regular feature of a system built on the annual renewal.

ENS domains, even within the blockchain ecosystem, require annual renewal payments to maintain ownership, with a grace period following expiration after which domains enter a premium auction phase where registration costs decrease gradually over several weeks before returning to standard pricing.

Even some blockchain-based naming systems have replicated the renewal model. They have moved the infrastructure to the chain but kept the commercial logic of recurring fees. That was not a mistake we wanted to make. If you are going to rebuild identity infrastructure from the ground up — if you are going to argue that this model is fundamentally better for people — then the model has to actually be better. Not incrementally better. Structurally better.

ENS domains are similar to traditional web domains in that they require an annual renewal fee. In contrast, some onchain domains have no renewal mechanism. Once you buy it, you own it.

That second sentence is the entire product philosophy behind Queensland Foundation, stated as plainly as it can be stated.

Why Queensland specifically

You might reasonably ask: even if the onchain model is philosophically superior, why Queensland? Why these six specific TLDs? Why not build a more generic product and let people put anything before the dot?

The answer goes to the heart of why place still matters in a digital world.

There is something particular about Queensland — about the coast, the climate, the character of the people who live there — that resists being dissolved into the generic internet. People who are from the Gold Coast are not simply from “Australia.” People who grew up watching the sun rise over Surfers Paradise have a specific relationship to a specific place. That place deserves its own digital address space.

.queensland, .qld, .brisbane, .surfersparadise, .gold-coast, .brisbane2032 — these are not arbitrary strings. They are the names that Queenslanders use to describe themselves and their home. We secured them not because there is commercial demand for generic digital real estate, but because the people of Queensland deserve to own their own corner of the internet, expressed in their own language, tied to their own identity.

And the onchain model makes that possible in a way that ICANN cannot. Under ICANN, a geographic TLD requires government endorsement, international review, and ongoing compliance with a framework written by an organisation headquartered in Los Angeles. Some names require government support or have special restrictions, particularly geographic names. A community wanting to stake a permanent digital claim to its own identity must first seek permission from a body whose primary mandate is global DNS stability — not community empowerment.

We believe communities should be able to own their names without intermediaries. That is a statement about sovereignty, digital identity, and the relationship between people and the places they come from. ICANN cannot deliver that. The blockchain can.

The long process of getting here

We want to be honest about one thing: this was not an easy or quick decision. We did not dismiss the ICANN path because it was inconvenient. We studied it seriously. We understood exactly what we would be getting by entering that system — global browser compatibility, search engine recognition, a familiar user experience, and the legitimacy that comes from operating inside the established order.

Those are real things. We did not pretend they were not.

But we also understood what we would be giving up. We would be giving up permanence. We would be giving up the ability to charge a one-time fee. We would be giving up the ability to offer true ownership rather than licensed access. We would be giving up the possibility that a Queensland address could ever be immutable — that it could live on infrastructure that nobody can shut down or reprice or revoke.

Applying for a gTLD is a considerable investment and responsibility, and beyond the reach of some organisations. That sentence, from ICANN’s own documentation about the application process, tells you something important. The traditional TLD system was designed for organisations with resources and legal departments and long operational runways. It was not designed for community-first projects with a mission to make digital identity accessible.

We are building something that starts at $5 and ends at “yours forever.” That ambition required a different foundation.

What the blockchain gives us that ICANN cannot

Let us be specific about the properties that onchain infrastructure provides, because the word “blockchain” is used loosely enough that it has become almost meaningless in some contexts. We are not invoking it as a buzzword. We are invoking it because of specific technical guarantees that matter for permanent digital identity.

The first is immutability. Once an address is registered on-chain, the record of that ownership is written into the blockchain’s history. It cannot be altered, rolled back, or deleted by any single party — including us. This is architecturally different from a database-backed registry, where records can be modified by anyone with administrative access.

The second is transferability without intermediary. Blockchain domains are owned by users rather than leased from a registrar, enabling users to fully control their domain names, including selling or transferring them to other parties. In the traditional system, transferring a domain involves the registrar as a necessary party. On-chain, ownership transfer is a direct transaction between two wallets. No intermediary. No transfer fee paid to a third party. No delay while a company processes your request.

The third is permanence of record. ENS records — and by extension, onchain address records built on similar infrastructure — are secured by Ethereum and cannot be destroyed. We do not need to maintain a server for your address to continue existing. We do not need to pay ICANN’s quarterly fees for your registration to remain valid. The blockchain is the registry, and the blockchain runs on distributed infrastructure that no single entity controls.

The fourth is trust minimisation. In the traditional system, you trust your registrar not to lose your domain, not to go bankrupt, not to lock you out during a transfer dispute. You trust the registry not to deaccredit or change its wholesale pricing in ways that make your registrar nonviable. You trust ICANN not to revoke or restructure the TLD. That is a lot of trust spread across a lot of parties, each of whom has their own interests. With onchain ownership, the trust is in the code — in the smart contract that records your ownership — not in a chain of human institutions.

A different kind of infrastructure means a different conversation

We acknowledge — again, openly — that onchain addresses occupy a different part of the technical landscape than traditional ICANN-sanctioned domains. They are not currently resolved by default in every browser. They require different tooling for certain use cases. The internet as it is built today was not designed for them.

But neither was the internet in 1993 designed for what it became in 2003. Infrastructure that seems incompatible with the mainstream has a consistent history of eventually becoming the mainstream, when the properties it offers are genuinely superior and the adoption curve catches up with the vision.

We are not asking anyone to take that on faith. We are asking people to think clearly about what a domain name is actually for. If it is a leased address that you maintain through annual payments, then the ICANN model serves that purpose reasonably well. But if it is a piece of digital identity that you should hold with the same permanence and confidence that you hold a piece of land, then the ICANN model is structurally incapable of providing that. It is not designed to. It was never designed to.

By moving domain names to the blockchain, projects in this space have challenged the centralised control of the traditional domain name system, offering users greater ownership, security, and flexibility.

That challenge is what Queensland Foundation is participating in. Not recklessly, not as a provocation, but because the people of Queensland deserve digital addresses that are genuinely theirs — permanently, irreversibly, unconditionally.

What we built instead

So: what did we build?

We secured six permanent onchain TLDs — .queensland, .qld, .brisbane, .surfersparadise, .gold-coast, and .brisbane2032 — on blockchain infrastructure. These TLDs are not registered with ICANN. They are not managed by a centralised registry organisation in the traditional sense. They are recorded and governed by smart contracts on a public blockchain. The ownership of every address registered under them is held by the person who bought it, in their own wallet, with no ongoing relationship to us required for that ownership to remain valid.

When someone buys a Queensland address, they are not entering into a lease. They are not establishing a relationship with a registrar who holds their name in trust. They are not becoming a subscriber to an annual renewal cycle. They own it, in the same way they own any other asset — freely, transferably, and permanently.

That sentence describes exactly what we built. It describes why we did not go through ICANN.

The case for community ownership of digital space

There is a broader argument here that extends beyond Queensland, and we want to name it plainly even though it takes us beyond the specifics of our project.

The traditional domain system, for all its engineering elegance, was built on a set of assumptions about how digital identity should work — assumptions that made sense in 1998 and have been quietly constraining the possibilities of the internet ever since. The assumption that names are licensed, not owned. The assumption that a centralised authority must validate what exists in the namespace. The assumption that the people best placed to manage digital identity are large organisations with legal teams and compliance departments, not individuals and communities.

Those assumptions built a functional internet. They also built a system where identity is always conditional, where access is always mediated, where the permanent ownership of a digital address is simply not available at any price.

As one of the architects of the DNS noted in an early memo, concerns about “rights” and “ownership” of domains were considered inappropriate. The appropriate frame was “responsibilities” and “service to the community.”

We respectfully disagree with that framing — not with its original intent, which was sensible in a world where domain names were being assigned to technical nodes rather than personal identities, but with its application to a world where a domain name has become the most fundamental unit of a person’s presence online. In that world, rights and ownership matter. The question of who holds an address, and on what terms, and for how long, and whether anyone else can take it — those are not administrative technicalities. They are questions about the relationship between people and their digital lives.

We believe those questions deserve better answers than the current system provides. Building on the blockchain is our attempt to provide them — for Queensland first, and as a model for what digital identity infrastructure can look like when it is designed around permanence and true ownership rather than recurring revenue and centralised control.

A final word on what “outside the system” means

We want to be clear about something that sometimes gets misunderstood. Choosing not to go through ICANN does not mean we are against the internet, against interoperability, or against the idea of shared infrastructure. The internet is the most powerful communications technology ever built, and the DNS is a remarkable piece of engineering that keeps it working at a scale that should astonish anyone who thinks about it clearly.

What we are against is the conflation of “the internet” with “the ICANN-controlled namespace.” Those two things are not the same. The internet is a protocol. The DNS is a layer on top of that protocol. The ICANN-sanctioned namespace is a specific set of policy and commercial arrangements built on top of the DNS. Challenging one layer does not mean rejecting the whole.

Alternate roots and alternative approaches to naming have historically been established by individuals and groups who want to express their disinterest in participating in the centralised processes for management of the authoritative root, or to protest the policies developed under those processes.

We are not protesting anything. We are building something. We are building permanent, owned, community-rooted digital addresses for Queenslanders who deserve to hold a piece of their digital future in their own name, without asking anyone’s permission to keep it.

That is why we didn’t go through ICANN.

Permanent Queensland addresses from $5. No renewals. Ever.

Claim Your Address →

Why we didn't go through ICANN