Gavin Wood co-founded Ethereum with Vitalik Buterin and served as its first CTO before leaving to start his own chain. He filed a technical paper under his own name, describing how the network he was about to launch should be governed, and how that governance should eventually end:
Gavin Wood (Polkadot paper, November 2016)The body of token holders would maintain the ultimate legitimacy and form a supermajority to augment, reparameterise, replace or dissolve this structure, something we don’t doubt the eventual need for: in the words of Twain ‘Governments and diapers must be changed often, and for the same reason’.
“This structure” was the two-chamber committee the paper proposed: one chamber of bonded validators, one of major client developers and ecosystem players, both answerable to the token-holder supermajority. Token holders did exactly what the passage allows, more than once. In 2023 they dissolved the committee. In November 2024 and again in September 2025 they reached past it, into a formula Wood’s paper had written down: how many new DOT the network is allowed to print every year.
What the whitepaper says about itself
The paper’s title is Polkadot: Vision for a Heterogeneous Multi-Chain Framework, and its abstract opens by naming what it believes is wrong with every chain that came before it:
Gavin Wood (Polkadot paper, November 2016)Present-day blockchain architectures all suffer from a number of issues not least practical means of extensibility and scalability. We believe this stems from tying two very important parts of the consensus architecture, namely canonicality and validity, too closely together.
The preface lists five failures it holds responsible for blockchains never reaching real-world deployment — scalability, isolatability, developability, governance, applicability — and is blunt about its own limits: “In the present work, we aim to address the first two issues: scalability and isolatability.” Governance and applicability, named as failures in the same breath, are left for someone else to solve.
The scalability failure gets a number attached. The paper states that the Parity Ethereum client “can process in excess of 3,000 transactions per second when running on performant consumer hardware,” while “current real-world blockchain networks are practically limited to around 30 transactions per second” — a hundred-to-one gap the paper blames not on weak software but on consensus design itself, one that it says “applies equally to both proof-of-work (PoW) systems such as Bitcoin and Ethereum” and to proof-of-stake systems alike. Bitcoin is named here as an example of the handicap, not as a target. A few pages later the paper names Bitcoin again, this time as a peer rather than a problem:
Gavin Wood (Polkadot paper, November 2016)Polkadot may be considered equivalent to a set of independent chains (e.g. the set containing Ethereum, Ethereum Classic, Namecoin and Bitcoin) except for two very important points: Pooled security; trust-free interchain transactability.
The design that follows is not a rewrite of Bitcoin’s rules. Bitcoin appears in it not as something to change but as one of the chains to set side by side.
The relay chain, four roles, and a name for the election
Polkadot’s answer to its own scalability complaint is a division of labor: a central chain, the relay chain, that does nothing but provide shared security and message-passing, and any number of application-specific “parachains” plugged into it. The whitepaper is explicit that the relay chain carries no application logic of its own — parachains do all of that — which inverts Bitcoin’s single chain doing both jobs at once.
Four roles run the relay chain, and the paper defines each narrowly. Validators are “the highest charge” — they seal blocks and post the bond that makes misbehavior costly, and the paper likens them to “the mining pools of current PoW blockchains.” Nominators back a validator’s bond with their own stake and, in the paper’s own description, “have no additional role except to place risk capital.” Collators maintain a full node for one parachain and hand validators a candidate block plus a proof that it is valid. Fishermen are the odd one out: independent “bounty hunters” motivated by “a large one-off reward” for proving that a validator signed two conflicting blocks or helped ratify an invalid one. The reward for exposing a single validator is deliberately small and grows as more validators’ illegal signatures corroborate a genuine attack. The paper gives the whole combination a name — Nominated Proof-of-Stake, NPoS — and expected validators to be re-elected “at most once per day but perhaps as seldom as once per quarter.”
| Bitcoin | Polkadot | |
|---|---|---|
| Consensus participants | Anyone who can pay for electricity | A capped set of validators, backed by nominators |
| Block production and finality | One rule (longest chain) does both | Split: BABE produces, GRANDPA finalizes |
| Chains | One | One relay chain plus many parachains |
| Cost of a new chain | None — there is only one chain | Auctioned lease (2021–2024), now purchased execution time |
| Who can change the issuance rule | No one | Token holders, by vote |
On the one row this table leaves unscored, how decentralized the result is, Polkadot’s 176-entity Nakamoto coefficient is the highest of any chain in the proof-of-work versus proof-of-stake comparison, which also records that the chain spent its first three weeks under a single key.
What actually shipped refined the paper’s own sketch by dividing time. Slots of about six seconds build into four-hour sessions, and six sessions make a twenty-four-hour “era.” Block production runs on BABE, a lottery each validator enters per slot; finality runs separately on GRANDPA, which agrees on whole chains rather than single blocks as soon as more than two-thirds of validators attest to one. The active validator set is re-elected at each era boundary — closer to daily than the paper’s quarterly upper bound.
Connecting a new parachain to the relay chain also changed shape: from November 2021 through 2024, projects competed for a limited number of slots through DOT-collateralized auctions, locking tokens for leases of up to two years and getting nothing else in return for as long as the lease ran. Agile Coretime, live since September 2024, replaced the auction with a purchase: a parachain buys relay-chain execution time outright, monthly or block by block, rather than locking capital against a multi-year bet.
From “roughly 10%” to a vote-imposed cap
The paper’s own issuance design set no ceiling at all:
Gavin Wood (Polkadot paper, November 2016)funds coming from a token base expansion (up to 100% per year, though more likely around 10%) together with any transaction fees collected
A market-based mechanism was meant to nudge that expansion rate up or down to hold a target proportion of tokens staked — a target, not a cap, and one the paper’s own FAQ calls “unlimited.” The same document, in the same breath, states that “Polkadot tokens are neither intended nor designed to be used as a currency.”
The chain that eventually launched on May 26, 2020 bore little resemblance to the token-holder government the paper described. Genesis began under a single administrator key, Sudo, running as a proof-of-authority network. NPoS took over on June 18; Sudo was removed on July 20, handing governance to DOT holders for the first time; transfers were enabled on August 18; and three days later, on August 21, holders voted to redenominate the unit itself, one old DOT becoming one hundred new ones — 86 percent in favor, with Web3 Foundation and Parity among the abstentions.
Issuance first ran on that rate-based model, a roughly constant percentage that made the supply grow exponentially. In November 2024 holders switched it to a fixed 120 million DOT a year, and as the base grew, the same fixed number became a falling percentage. By September 2025, circulating supply had reached roughly 1.6 billion DOT, at which point 120 million a year was already under 8 percent.
That September, token holders voted anyway. Referendum 1710, filed under a track called “Wish For Change,” passed with 81 percent in favor. It ended open-ended issuance and imposed a hard ceiling of 2.1 billion DOT, replacing the flat annual mint with a stepped schedule: every two years, the yearly mint is reset to 13.14 percent of whatever supply still remains under the cap.
The new schedule went live with a runtime upgrade on January 27, 2026, and the first step landed on March 14, 2026, taking the yearly mint from 120 million to 55 million DOT — the cut of roughly 53.6 percent the proposal had projected. Polkadot’s own post on the result, as The Block quoted it, projected supply near 1.91 billion DOT by 2040 under the new rule, against roughly 3.4 billion under the old one. That projection does not fit the schedule as it took effect: worked forward from the 55 million first step, the rule gives about 2.05 billion DOT by 2040, close to the 2.06 billion a Motley Fool analysis on Nasdaq reached. Like Bitcoin’s 21 million, the cap is approached rather than landed on: the schedule steps toward 2.1 billion, and the proposal puts the year it reaches the cap at 2160.
Governance and initial distribution: an ICO, a frozen two-thirds, and a dissolved committee
Web3 Foundation ran Polkadot’s first token sale in October 2017, selling half of the 10 million DOT supply of the time and raising roughly $145 million in ether. Within days, roughly two-thirds of it nearly vanished. A shared code library in Parity’s own multi-signature wallet software — Parity being the company Wood co-founded — had a flaw that let one user accidentally delete it, freezing every wallet built on that library at once. More than 500 wallets across several projects were affected; Polkadot’s own share came to about $98 million, locked alongside the rest. Web3 Foundation went back to the market twice. A 2019 private sale placed 500,000 DOT, 5 percent of the supply, at a claimed $1.2 billion valuation, with the proceeds undisclosed. A second private sale in July 2020 raised about $43 million over 72 hours, at a price kept undisclosed under NDA and with tokens locked for roughly five months.
The governance the 2016 paper sketched as two committees under a token-holder supermajority did get built, and did get replaced by the same supermajority. From 2020, Polkadot ran what the network calls Governance V1: a Council, a Technical Committee, and public referenda alongside them — the bicameral shape the whitepaper had described almost exactly. In June 2023, OpenGov retired all of it. Fifteen separate tracks now let DOT holders propose and vote directly, with no council seat standing between a proposal and a decision; a conviction-voting mechanism multiplies a vote’s weight by how long the voter is willing to lock their tokens, up to six times for the longest lock available. Referendum 1710 — the vote that capped a supply the 2016 paper had left uncapped — passed on this system, under the exact “supermajority” authority that same paper had named nine years earlier as the only body allowed to change the rules.
What Wood has said about Bitcoin
In an interview titled “Polkadot: A Bet Against Maximalism,” he turned to Bitcoin’s mining process directly:
Gavin Wood (Real Vision interview, January 2021)Bitcoin “uses up the equivalent of … some small country’s energy simply in securing itself.”
In the same conversation he raised a second objection, on throughput rather than energy: Bitcoin’s advertised ten-minute block target, he said, does not describe how long a payment actually takes to confirm in practice, where a confirmation can run closer to an hour. But Wood’s verdict on Bitcoin isn’t one-sided. In a separate post, he put it this way:
Gavin Wood (Tweet from Polkadot's official account)Bitcoin pioneered resilience. Ethereum generalized it. Polkadot took it one step further with scale.
Both objections — energy and throughput — are about what Bitcoin costs to run, not about whether its design works, and crediting Bitcoin as the origin of the resilience his own network claims to extend is not how someone talks about a design they think needs replacing. The whitepaper Wood wrote five years earlier had already made the same point, naming Bitcoin as one of the systems Polkadot considers itself “equivalent to” — a chain worth connecting to, not one whose consensus needed replacing.
Polkadot’s supply since Referendum 1710, plotted against Bitcoin’s cap and ten other currencies on one normalized index:
Significance to Bitcoin
The digital-gold structural-features analysis counts a fixed supply as one of six features that, held together, nothing else has matched — and names the party who can change that supply as the sharper test than the cap number itself. Polkadot answers that test in the plainest way available: the party who can change the rule has a name, a wallet, and a voting record. The fixed-supply-versus-adjustable-money comparison asks who holds the authority behind any monetary design; Referendum 1710 is that authority exercising itself in the open, on a formula its own author wrote into a whitepaper and handed to a vote. The twelve-chain design comparison records Polkadot’s cap in a single line: capped in 2026, by a vote that can vote again. Bitcoin’s 21 million has no comparable vote to win, because it has no comparable voter. Polkadot’s 2.1 billion has one, and in September 2025, that voter used it.
