Bitcoin Ordinals and Runes — How Casey Rodarmor Brought NFTs to Bitcoin

Bitcoin was designed to do one thing well: move value between addresses. Its scripting language was deliberately limited. Its block space was deliberately scarce. The people who built Bitcoin made a series of intentional choices to prevent it from becoming a general-purpose computation platform — because general-purpose computation introduces attack surfaces, and attack surfaces jeopardize the monetary properties that make Bitcoin valuable.

For over a decade, this design choice held. Then, in January 2023, a software engineer named Casey Rodarmor launched Ordinals, and Bitcoin acquired NFTs.

## What Ordinals actually is

To understand Ordinals, you need to understand two things: satoshis and the Taproot upgrade.

A *satoshi* is the smallest unit of bitcoin — one hundred millionths of a bitcoin (0.00000001 BTC). There are approximately 2.1 quadrillion satoshis in Bitcoin's total supply. Satoshis are not individually distinguishable in Bitcoin's accounting model. They are fungible units; a satoshi sent in one transaction is, technically, indistinguishable from a satoshi sent in another.

Rodarmor's *ordinal theory* assigns a sequential number to every satoshi, based on the order in which it was mined. The first satoshi mined in Bitcoin's history is ordinal 0. The last satoshi to ever be mined will have ordinal 2,099,999,997,689,999. This numbering system is external to Bitcoin's protocol — it is a convention, not a protocol change. But it enables something useful: any satoshi can be uniquely identified by its ordinal number, and therefore tracked.

*Inscriptions* use this uniqueness. The Taproot upgrade, implemented in November 2021, introduced a new transaction structure called SegWit v1. A key feature of Taproot was the ability to include arbitrary data in a transaction's witness field — data used to prove transaction validity, which is stored on-chain but not subject to the same size constraints as transaction outputs. Rodarmor's insight was that you could inscribe arbitrary content — images, text, audio, video, code — into the witness data of a transaction that creates a specific satoshi, and track that satoshi as a carrier of that content.

## The fee market response

The launch of Ordinals in January 2023 and its rapid growth through 2023 had a direct and measurable effect on Bitcoin's fee market. Inscription transactions competed for block space with regular payment transactions. As inscription demand rose, average transaction fees rose.

Here's the uncomfortable truth about this: the reaction within the Bitcoin community was sharply divided along philosophical lines. One faction saw fee increases as unambiguously positive — miners earn transaction fees, and higher fees mean greater miner revenue, which strengthens the security budget as the block subsidy declines over successive halvings. The incentive alignment was, by this view, exactly correct.

Another faction saw inscription data as polluting a monetary network with nonmonetary content, bloating the blockchain with cat pictures, and introducing attack vectors by normalizing unusual script patterns. Some developers proposed changing Bitcoin's consensus rules to filter out "non-standard" inscription transactions. This effort failed, partly because filtering would have required a consensus change and partly because the economic argument for higher fee revenue proved persuasive to miners and economically-minded participants.

## BRC-20 tokens and their limitations

By spring 2023, developers built on top of Ordinals' inscription mechanism to create fungible tokens. *BRC-20* tokens used JSON text inscriptions to define token issuance and transfer rules. The convention was simple: you inscribe a text file defining the token, and subsequent inscriptions moving the token follow the same convention.

The numbers suggest something peculiar: BRC-20 tokens at peak created more on-chain transactions than all other Bitcoin usage combined. The mechanism was inefficient — every transfer required an inscription transaction, taking up substantial block space — and the tokens had no on-chain enforcement of the rules (the JSON conventions were interpreted by indexers, not by Bitcoin nodes). But the demand was real, driven by speculation.

The limitations of BRC-20 were structural, not incidental. Bitcoin's scripting language cannot enforce token rules at the consensus layer. Every BRC-20 transfer relied on all parties trusting the same off-chain indexer to interpret the JSON correctly.

## What Runes changed

In April 2024, coinciding with the Bitcoin halving, Rodarmor launched the *Runes protocol* — a more efficient mechanism for fungible tokens on Bitcoin's base layer.

Runes moved from the inscription mechanism to OP_RETURN outputs. OP_RETURN is a script opcode that creates an unspendable output used to embed data in a transaction. It is more space-efficient than using witness data, and it avoids creating unspendable UTXO entries (a concern with BRC-20 tokens, which left "dust" outputs in Bitcoin's UTXO set).

The Runes protocol defines a token by a *runestone* — an OP_RETURN message that specifies token creation, issuance rules, and transfer instructions. The protocol is designed to be compatible with Bitcoin's existing UTXO model: Rune balances are held in UTXOs, and transfers move balances between UTXOs in a way that Bitcoin nodes can validate (though the Rune-specific logic still requires dedicated indexers).

It's worth noting that neither Ordinals nor Runes changed Bitcoin's consensus rules. Both are application-layer protocols that use Bitcoin's existing transaction structure in ways its designers did not anticipate. Bitcoin's consensus layer remains agnostic about their existence.

## The structural implications for Bitcoin

Two questions follow from Ordinals and Runes that are worth examining.

The first is the fee market question. Bitcoin's security model after 2140 — when the block subsidy reaches zero — depends entirely on transaction fees to compensate miners. The introduction of non-monetary use cases for Bitcoin's block space has expanded the potential fee base. Whether this effect is large enough to matter at the scale of Bitcoin's future security budget is uncertain, but the direction is positive from a miner revenue perspective.

The second is the fungibility question. Ordinal theory, by assigning unique identities to satoshis, creates a technical mechanism for fungibility degradation. If some satoshis are valued differently — because of the inscriptions attached to them — it could, in theory, complicate Bitcoin's use as money. In practice, this effect has been limited: the wallets and exchanges that handle Bitcoin's monetary use cases treat satoshis interchangeably, and the market for "rare" satoshis remains a small submarket.

> **Key Takeaway:** Ordinals and Runes are application-layer constructions that used Taproot's expanded data-carrying capacity to bring NFT and fungible token activity to Bitcoin. The significance is not primarily in the tokens themselves, but in what they reveal: Bitcoin's block space has more uses than its designers anticipated, and the fee market consequences of that expanded utility may be the most durable structural change these protocols introduce.

Bitcoin Ordinals and Runes — How Casey Rodarmor Brought NFTs to Bitcoin

// COMMENTS

ON THIS PAGE