NFTs and Tokens on Bitcoin

Bitcoin was for money. That was the rule for fourteen years. Then January 2023 broke it.

I watched the fee market change shape that winter. Today the Bitcoin chain carries tens of millions of inscriptions: images, text, audio, token definitions, all written onto individual satoshis. Marketplaces trade Bitcoin-native NFTs. Protocols issue fungible tokens on top of the base chain. You can call this progress or you can call it noise. Either way it shows up on your fee estimate, so you need to know how it works.

What are Ordinals?

Ordinals number and track individual satoshis.

Casey Rodarmor launched the protocol in January 2023. His insight was small and almost obvious in retrospect. Bitcoin has processed billions of transactions since 2009, and the order satoshis were mined in is deterministic. Ordinal theory assigns every sat a sequential number based on that mining order. Each sat becomes identifiable.

Once a sat is identifiable, you can inscribe data onto it and track who holds it. Fungibility cracks. One sat is no longer interchangeable with any other sat. That crack is the whole foundation of Bitcoin-native NFTs.

Ordinals require no soft fork, no consensus change, no permission. They use what SegWit (2017) and Taproot (2021) already shipped. SegWit gave us the witness field where inscription data lives. Taproot made storing larger payloads in that field cheap enough to matter.

Translated: a satoshi becomes a numbered collectible you can send, receive, and track like any other Bitcoin.

What are Inscriptions?

An inscription is the data written onto a satoshi. It is the content that gives an Ordinal NFT whatever value the market assigns it.

You can inscribe almost any file type: JPEG, PNG, SVG, GIF, plain text, audio, code, HTML, 3D models. The bytes sit in the witness portion of the transaction. Most Ethereum NFTs store only a pointer to a file held on a server or IPFS. Bitcoin inscriptions store the file itself, on-chain.

So the inscription is as durable as Bitcoin. No server to die, no IPFS gateway to vanish, no company to fold and take your JPEG with it.

The cost side is harsh. Block space is finite. Bigger files cost more in fees, in proportion to the bytes. A small text inscription might run a few dollars. A high-resolution image during a fee spike can cost hundreds. And when inscription activity surges, the whole fee market rises, which hits everyone moving sats that day, not only the people minting art.

Wallets matter too. A wallet that does not know about Ordinals will treat your inscribed sat like any other sat. Send it to a normal exchange deposit address or spend it in a routine payment, and the inscription is gone. The sat still exists somewhere in someone else's UTXO. The artwork it carried is overwritten.

What are Runes?

Runes is the fungible token protocol on Bitcoin. Casey Rodarmor shipped it in April 2024, timed to the fourth halving.

Before Runes, the dominant token standard was BRC-20, which piggybacked on text inscriptions to track balances. BRC-20 was a hack. Every transfer needed multiple transactions and dumped junk into blocks.

Runes is cleaner. Token data goes into a small OP_RETURN output, which Bitcoin already supports and which carries no spendable value. Balances ride along inside normal UTXOs. A Runes transfer looks like a regular Bitcoin transaction with extra metadata.

A Rune has a total supply, a minting schedule (open or fixed), a name, and a symbol. You can move it in a single transaction, including multi-party operations that BRC-20 would have needed half a dozen txes for.

The launch was loud. Around the halving block, Runes briefly accounted for the majority of on-chain activity. Fees hit historic highs. Miners booked record revenue. Anyone trying to send a normal payment that week paid for the party.

That episode is the whole story in miniature. Runes generates economic activity on Bitcoin and feeds the security budget. It also taxes every other base-layer user during the spikes.

How Bitcoin NFTs Compare to Ethereum NFTs

Both chains have NFT ecosystems. The architectures sit on opposite ends.

Storage: Bitcoin inscriptions carry the file on-chain forever. Ethereum NFTs usually store metadata and a URL pointing to a file on IPFS or a server. Bitcoin is more durable and more expensive to mint. Ethereum is cheaper at mint and dependent on whoever is hosting the asset.

Contract logic: Ethereum NFTs are smart contracts. Code on the chain can enforce royalties, transfer rules, governance. Ordinals have no contract layer. The inscription exists. Trade rules live at the marketplace level via PSBTs (Partially Signed Bitcoin Transactions), not in code.

Royalties: Ethereum contracts can enforce royalties on every secondary sale. Bitcoin has no native royalty mechanism. A creator earns a royalty on resale only if the buyer chose a marketplace that honours it voluntarily, which means in practice often not.

Market infrastructure: Ordinals trading runs through PSBTs on Magic Eden, Ordinals Wallet, and Gamma. Ethereum NFT markets are older, with deeper liquidity and more standard tooling.

Security model: Inscriptions are dumb in the good sense. No contract logic to exploit, no upgrade proxy to manipulate, no admin key for an attacker to lift. Ethereum gives you programmability at the cost of contract-level attack surface.

Neither model wins. Each reflects the chain it grew on.

Key Definitions

Satoshi (sat): the smallest unit of BTC. 1 BTC = 100,000,000 sats.

Ordinal: a satoshi assigned a sequential number by the Ordinals protocol so you can track it as a specific sat.

Inscription: data (image, text, code, file) written onto a specific satoshi in the transaction witness.

BRC-20: an early fungible token standard on Bitcoin built on text inscriptions. Less efficient than Runes.

Runes: a fungible token protocol using OP_RETURN data and standard UTXOs, launched April 2024.

PSBT: Partially Signed Bitcoin Transaction. The format Ordinals marketplaces use to coordinate trades before final signing.

Risk Note

Fees spike during inscription and token waves, and they spike for everyone on the network. NFT and rare sat markets are thin and speculative; prices fall as fast as they climbed. An Ordinals-unaware wallet can destroy an inscription with a routine spend. Some inscribed content raises IP questions under European law. The tooling is young, so expect bugs and protocol changes.

Reader Takeaway

• Ordinals make individual sats trackable and tradeable. Inscriptions write permanent data onto those sats.
• Runes is a cleaner fungible token protocol than BRC-20, with less bloat and closer behaviour to native Bitcoin transfers.
• Bitcoin NFTs store content on-chain forever. Ethereum NFTs lean on external storage to stay cheap and flexible.
• Bitcoin has no native royalty enforcement. Ethereum offers programmability and the contract risk that comes with it.
• Start small. Use an Ordinals-aware wallet. Watch fees during high-activity weeks.

Chapter Summary

• Ordinals assign sequential numbers to satoshis so each one can serve as a unique digital asset.
• Inscriptions store the file itself on-chain, in contrast to most Ethereum NFTs, which only reference externally hosted media.
• Runes launched in April 2024 as an efficient fungible token protocol using OP_RETURN, replacing BRC-20.
• Bitcoin NFTs and Ethereum NFTs differ on storage, contract logic, and royalty enforcement.
• High Ordinals or Runes activity raises fees for every Bitcoin user. Thin markets, young tools, and fee risk are the main hazards for participants.

References

• Casey Rodarmor: Ordinals documentation and specification
• Bitcoin Optech newsletters on Ordinals, Inscriptions, and Runes
• Magic Eden: market data and PSBT trading documentation
• Mempool.space: fee history and block analysis
• Ethereum.org: ERC-721 and NFT standard documentation (for comparison)