0x: An open protocol for decentralized exchange on the Ethereum blockchain

01 Claim ledger

What the paper is allowed to say

Each row is a sentence already in the study. The status is the same on every row: a model claim, not a live measurement.

1. Claim 01 · Paper model

   Orders are messages

   A maker signs an order: this much of A for that much of B, until a time, for this fee. The signature is the commitment. The chain has not seen it yet.

2. Claim 02 · Paper model

   Relayers match without custody

   A relayer keeps an order book and tells a taker about the order. The relayer does not hold the tokens. Settlement is a contract call the taker makes.

3. Claim 03 · Paper model

   The exchange contract is the settlement layer

   It checks the signature, the expiry, the fill so far, and then it pulls tokens. A relayer who disappears does not take the inventory, because they never had it.

4. Claim 04 · Paper model

   Fees pay the relayer

   The paper includes a fee the maker attaches, paid in a protocol token in the original design. That choice ties marketplace incentives to the token. Later versions loosened it.

5. Claim 05 · Paper model

   On-chain order books were the cost being avoided

   Putting every quote in storage was expensive. 0x's split is a response to that cost, and it creates a new problem: which relayer has the order, and is it still valid.

02 Three cuts

Observation, model, falsifier

A desk does not stop at the summary. Each claim is cut three ways, using only this study's own assumptions and checks. Nothing here is a new figure.

1. 01 Orders are messages

   Observation

   What the study says

   A maker signs an order: this much of A for that much of B, until a time, for this fee. The signature is the commitment. The chain has not seen it yet.

   Model

   What has to hold

   Takers can submit a transaction before the order expires. A congested chain turns a firm quote into a maybe.

   Falsifier

   What would retire it

   Where does the order book live, and who can settle an order they did not originate?

2. 02 Relayers match without custody

   Observation

   What the study says

   A relayer keeps an order book and tells a taker about the order. The relayer does not hold the tokens. Settlement is a contract call the taker makes.

   Model

   What has to hold

   Makers have approved the contract and still hold the tokens at settlement time.

   Falsifier

   What would retire it

   Where does the order book live, and who can settle an order they did not originate?

3. 03 The exchange contract is the settlement layer

   Observation

   What the study says

   It checks the signature, the expiry, the fill so far, and then it pulls tokens. A relayer who disappears does not take the inventory, because they never had it.

   Model

   What has to hold

   Makers have approved the contract and still hold the tokens at settlement time.

   Falsifier

   What would retire it

   How does cancellation become visible to takers?

4. 04 Fees pay the relayer

   Observation

   What the study says

   The paper includes a fee the maker attaches, paid in a protocol token in the original design. That choice ties marketplace incentives to the token. Later versions loosened it.

   Model

   What has to hold

   Tokens transfer in the amount the contract expects. Fee-on-transfer tokens desynchronise the trade.

   Falsifier

   What would retire it

   Is a protocol token required for this deployment's fees, or is that the 2017 design only?

5. 05 On-chain order books were the cost being avoided

   Observation

   What the study says

   Putting every quote in storage was expensive. 0x's split is a response to that cost, and it creates a new problem: which relayer has the order, and is it still valid.

   Model

   What has to hold

   Takers can submit a transaction before the order expires. A congested chain turns a firm quote into a maybe.

   Falsifier

   What would retire it

   Where does the order book live, and who can settle an order they did not originate?

03 Sequence

One action, as an operating tape

1. 01A maker approves the exchange contract to move their token, and signs an order off-chain.
2. 02A relayer indexes the order and shows it to takers.
3. 03A taker submits the order to the exchange contract, possibly filling only part of it.
4. 04The contract verifies the signature and that the remaining amount and the expiry allow this fill, then transfers both sides.
5. 05A partially filled order stays valid for the rest until it expires or the maker cancels it on-chain.

04 Load-bearing

The argument, and where a pitch drops it

1. Off-chain quotes can lie about availability

   The cut

   The maker may have spent the tokens, or cancelled, between the relayer's screen and the transaction. A taker learns the truth on-chain. The paper's design accepts failed settlement as the price of cheap quotes. Interfaces that hide failures are pretending otherwise.

   Why it carries weight

   If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.

   Where it is dropped

   Off-chain orders are only as live as the API that serves them. The paper does not remove liveness risk. It moves it.

2. Relayers compete on flow, not on custody

   The cut

   Because anyone can settle any properly signed order, a relayer does not own the liquidity. They own the attention. The paper's open protocol is a bet that liquidity wants to be quotable in more than one place. It also means a relayer's commercial moat is not the matching contract.

   Why it carries weight

   If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.

   Where it is dropped

   Off-chain orders are only as live as the API that serves them. The paper does not remove liveness risk. It moves it.

3. The protocol token is not required by the idea of signed orders

   The cut

   The original fee design uses one. The useful mechanism, signed orders plus on-chain settlement, can exist without it, and later 0x versions moved that way. Separate the matching idea from the 2017 token choice.

   Why it carries weight

   If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.

   Where it is dropped

   Off-chain orders are only as live as the API that serves them. The paper does not remove liveness risk. It moves it.

05 Register

What has to be true

• Model · Not re-measured

  Makers have approved the contract and still hold the tokens at settlement time.

• Model · Not re-measured

  Takers can submit a transaction before the order expires. A congested chain turns a firm quote into a maybe.

• Model · Not re-measured

  Cancellations are visible. A cancel that does not land leaves the order fillable.

• Network · Not re-measured

  Tokens transfer in the amount the contract expects. Fee-on-transfer tokens desynchronise the trade.

06 Divergence

What happened after the paper

0x shipped, revised its settlement contracts, and became infrastructure that aggregators and market makers quoted against. The 2017 paper is the off-chain order and on-chain settlement split. The original fee-token mechanics should not be described as if they were unchanged.

A later client, parameter set, or reward formula is a different object. Cite this paper for the mechanism. Cite a primary release for the network. This desk has not re-run the proof.

07 Pre-mortem

What to check before you use the idea

1. Kill test 1Where does the order book live, and who can settle an order they did not originate?
2. Kill test 2What happens if the maker's balance is gone by the time the transaction lands?
3. Kill test 3How does cancellation become visible to takers?
4. Kill test 4Is a protocol token required for this deployment's fees, or is that the 2017 design only?
5. 0 of 4 marked on this browser. A mark is a reading note, not a pass, a rating, or a recommendation.

08 Anatomy

The paper, in the order a builder needs

09 Lexicon

Terms, opened into the record

Maker

The party who signs an off-chain order and provides the liquidity.

Relayer

A party who broadcasts orders and may charge a fee, without holding the assets.

Settlement

The on-chain transaction that checks the signature and moves tokens.

Partial fill

Taking only some of a signed order, leaving the rest open until expiry or cancellation.

10 Repository

Every linked record on this page

Underlined words open a page that already exists: a concept, a protocol profile, a failure record, or another paper. If a word is not underlined, this desk does not have a record for it.

11 Related

The next papers, as theses

• Bancor · 2017

  Bancor Protocol: Continuous Liquidity and Asynchronous Price Discovery for Tokens through their Smart Contracts

  The 2017 Bancor paper on smart tokens that hold reserves of other tokens and quote a continuous price from a reserve ratio. It is an early automated-liquidity design, distinct from Uniswap's later constant-product pools.

• Uniswap · 2020

  Uniswap v2 Core

  The 2020 core paper for Uniswap v2: a constant-product automated market maker, with arbitrary ERC-20 pairs, price accumulators, and a flash-swap callback. It is the clearest short specification of the pool that much of later DeFi either forked or assumed.

• Augur · 2018

  Augur: a Decentralized Oracle and Prediction Market Platform

  The Augur paper: prediction markets whose outcomes are reported by token holders, with a dispute ladder that can escalate a contested result. It is both a market design and an oracle design. The historic library already holds Gnosis. Augur is the other canonical public prediction-market paper and was not in that set.