Ordinals

What Are Ordinals?

Ordinals are a numbering scheme for satoshis, the smallest unit of bitcoin (1 BTC = 100 million satoshis). Created by Casey Rodarmor and launched in January 2023, ordinal theory assigns each satoshi a unique serial number based on the order in which it was mined. This numbering allows individual sats to be tracked, transferred, and inscribed with data.

The term "Ordinals" refers to two related concepts: ordinal theory itself (the numbering system) and inscriptions (the process of attaching content to numbered satoshis). Together, they enable Bitcoin-native digital artifacts: content permanently stored on the blockchain, tied to a specific satoshi, and transferable using standard Bitcoin transactions.

Unlike NFTs on other blockchains that typically store metadata on-chain with the actual content hosted elsewhere, Ordinals inscriptions store the full content directly in Bitcoin block data. This makes inscriptions fully self-contained and as durable as Bitcoin itself.

How It Works

Ordinal Theory: Numbering Satoshis

Ordinal theory defines a deterministic scheme for numbering every satoshi that will ever exist. The numbering follows a simple rule: satoshis are numbered in the order they are mined, starting from 0. The first satoshi in the first block's coinbase reward is ordinal 0, the second is ordinal 1, and so on. When a transaction spends satoshis, ordinal theory uses a first-in, first-out (FIFO) convention to determine which specific satoshis move to which output.

Consider a transaction with two inputs and two outputs. The satoshis from the first input are assigned to the outputs in order, followed by the satoshis from the second input. This convention is entirely off-chain: the Bitcoin protocol itself has no awareness of ordinal numbers. Ordinal tracking is a social layer that participants agree to follow.

# Ordinal numbering example
# Block 0 coinbase: 50 BTC = 5,000,000,000 sats
# Ordinals 0 through 4,999,999,999

# Block 1 coinbase: 50 BTC = 5,000,000,000 sats
# Ordinals 5,000,000,000 through 9,999,999,999

# FIFO transfer tracking:
# Input 0: sats [100, 101, 102]
# Input 1: sats [200, 201]
# Output 0: 3 sats -> receives [100, 101, 102]
# Output 1: 2 sats -> receives [200, 201]

Inscriptions: Embedding Data

Inscriptions are the mechanism for attaching content to a satoshi. They leverage the Taproot upgrade (activated in November 2021) and specifically the witness data section of a transaction. Content is placed in a Taproot script-path spend using an envelope format within the witness.

The inscription envelope uses a series of OP_FALSE OP_IF ... OP_ENDIF opcodes to wrap the data. Because the content sits inside an unexecuted conditional branch, it does not affect transaction validation. Nodes store the data as part of the witness, but never execute it.

# Inscription envelope structure (simplified)
OP_FALSE
OP_IF
  OP_PUSH "ord"              # Protocol identifier
  OP_PUSH 1                  # Content-type tag
  OP_PUSH "image/png"        # MIME type
  OP_PUSH 0                  # Body tag (separator)
  OP_PUSH <raw image bytes>  # Actual content data
OP_ENDIF

The Taproot witness discount plays an important role here. Witness data receives a 75% discount on transaction weight compared to non-witness data, meaning inscription content costs roughly one quarter the fee rate of equivalent data placed in a standard transaction output. This discount, originally designed to incentivize Taproot adoption, makes storing large amounts of data economically viable.

The Inscription Process

Creating an inscription is a two-step process called the commit-reveal scheme:

1. Commit transaction: the inscriber creates a Taproot output whose script tree includes the inscription content. This transaction does not reveal the content itself, only the commitment (the Taproot output address).
2. Reveal transaction: the inscriber spends the commit output using the script path, which reveals the full inscription content in the witness data. The inscription is now permanently recorded on-chain.

The inscription is bound to the first satoshi of the reveal transaction's first output. From that point forward, ordinal theory tracks that specific satoshi as it moves through subsequent transactions.

Content Types

Inscriptions support any content type expressible as a MIME type. Common inscription formats include:

The practical size limit for inscriptions is the maximum Bitcoin block size of approximately 4 MB (the full weight limit of a block), though most inscriptions are far smaller. Larger inscriptions require higher fees and may take longer to be included in a block.

Derivative Protocols

Ordinals theory and inscriptions have spawned several derivative protocols that build on the inscription mechanism:

• BRC-20: a fungible token standard that uses JSON inscriptions to deploy, mint, and transfer tokens. BRC-20 tokens are tracked off-chain by indexers that interpret inscription content as token operations.
• Runes: a more efficient fungible token protocol created by Casey Rodarmor as an alternative to BRC-20. Runes use OP_RETURN outputs instead of inscriptions, reducing the creation of unnecessary UTXOs.

These protocols demonstrate how the Ordinals ecosystem has evolved beyond simple digital artifacts into a broader platform for tokenization on Bitcoin.

Rare Satoshis

Ordinal theory introduces a rarity framework based on Bitcoin's periodic events. Certain satoshis are considered more collectible based on when they were mined:

Rare sat collecting has become a niche market. Miners and collectors use specialized software to identify and isolate valuable satoshis within their UTXOs, carefully constructing transactions to avoid accidentally spending them as fees.

Use Cases

• Digital art and collectibles: artists inscribe original works directly on Bitcoin, creating provably scarce digital artifacts that require no external hosting or marketplace infrastructure.
• On-chain applications: developers inscribe HTML, CSS, and JavaScript to create interactive experiences that run entirely from Bitcoin block data.
• Token issuance: protocols like BRC-20 and Runes use ordinal theory and inscriptions as a foundation for fungible token systems on Bitcoin.
• Permanent data storage: inscriptions provide censorship-resistant, permanent storage for important documents, records, or cultural artifacts.
• Collectible satoshis: the rarity framework creates a market for specific satoshis based on their position in Bitcoin's issuance schedule.

Fee Market Impact

Ordinals have had a measurable effect on the Bitcoin fee market. During periods of high inscription activity, fees have spiked significantly as inscribers compete for block space with regular financial transactions.

In early 2023 and during subsequent waves of activity (particularly around BRC-20 minting events), average transaction fees rose substantially. Blocks filled with inscription data left less room for standard transactions, pushing fees higher for all users.

For miners, inscriptions represent additional revenue. The witness discount means inscriptions pay lower fees per byte than standard transactions, but the sheer volume of inscription data has increased total fee revenue per block. This dynamic is relevant to long-term fee market sustainability as block rewards decrease with each halving.

Risks and Considerations

Block Space Debate

The most contentious aspect of Ordinals is their consumption of block space. Critics argue that Bitcoin's limited block space should be reserved for financial transactions, and that storing images and other media on-chain is wasteful. Proponents counter that all valid transactions have equal right to block space, and that the fee market itself determines what data is valuable enough to include.

Some community members have proposed filtering inscription transactions at the mempool level or introducing protocol changes to restrict non-financial data. Others view this as a form of censorship incompatible with Bitcoin's permissionless nature.

Node Storage Requirements

Inscriptions increase the size of the Bitcoin blockchain. Large inscription waves have accelerated blockchain growth, raising concerns about the long-term cost of running a full node. While witness data can be pruned by nodes that do not need historical data, archive nodes must store all inscription content permanently.

Indexer Dependency

Ordinal numbers and inscription tracking are not enforced by Bitcoin consensus. They rely on off-chain indexers: software that follows ordinal theory's conventions to assign numbers and track ownership. If different indexers disagree on the rules (as has happened with edge cases), it can create confusion about which satoshi holds which inscription.

No Protocol-Level Ownership

Bitcoin's protocol does not recognize ordinal numbers or inscriptions. A wallet that does not implement ordinal tracking may inadvertently spend an inscribed satoshi as part of a regular transaction, effectively destroying the association between the inscription and its intended owner. Specialized ordinal-aware wallets are required to safely hold and transfer inscribed satoshis.

Relationship to Taproot

Inscriptions depend on the Taproot upgrade and its witness discount. The Taproot soft fork was designed to improve privacy and smart contract flexibility, not to enable data storage. Some argue that the witness discount being exploited for data storage was an unintended consequence, while others view it as a legitimate use of available protocol features.

This glossary entry is for informational purposes only and does not constitute financial or investment advice. Always do your own research before using any protocol or technology.