Mimblewimble: The 2016 Protocol Proposal

Author Tom Elvis Jedusor
Year 2016
Project Mimblewimble protocol design
License Not stated in the original note
Official Source Original Mimblewimble note (plain text, dated July 19, 2016)

This page is an educational summary and analysis, not a verbatim reproduction. CryptoGloss does not claim authorship of the original work. The original note does not state a license; rights remain with the author unless the author or a rights holder says otherwise.

Mimblewimble is a blockchain protocol design, not a single coin or network. Tom Elvis Jedusor published the original note under that pseudonym on July 19, 2016. It proposed a different way to represent transactions and chain state: amounts are hidden with cryptographic commitments, transactions do not use reusable on-chain payment addresses, and spent transaction data can sometimes be removed through cut-through. Later work by Andrew Poelstra made the design more precise and explored additional scaling techniques. Projects such as Grin and Beam implement related ideas, but their networks and wallet features are not identical to the 2016 proposal.

What the original proposal set out to change

In a conventional UTXO blockchain, transactions reveal a public graph of which outputs are spent to create later outputs, even when the owners are represented by pseudonymous keys. Jedusor’s note argued that this history can expose patterns and that much of the stored transaction detail no longer describes the current set of spendable coins. The proposal combines confidential transactions with a way to validate the net state without keeping every intermediate transaction in the same form forever.

The note itself is dated July 19, 2016. Later accounts differ on when the text appeared in a research chat: Andrew Poelstra’s follow-up describes an August 2 UTC appearance, while Grin’s history gives an August 1 date. Those are accounts of circulation, not the date printed on the original document, so they should not be collapsed into one date.

How amounts can be hidden and still checked

Mimblewimble uses Pedersen commitments to represent values without publishing the values themselves. In simplified terms, a commitment lets a verifier check that inputs and outputs balance while keeping their amounts confidential. Range proofs provide a separate check that committed values are in an allowed range, preventing invalid values from being used to create value outside the rules.

This does not make every part of a transaction secret. Nodes still validate the commitments, proofs, signatures, and other consensus data. What an outside observer can infer depends on the implementation, the transaction’s context, and what the observer saw before transactions were combined.

No on-chain address does not mean no wallet communication

Mimblewimble does not put a reusable recipient address into each on-chain output. The parties instead exchange data while constructing a transaction. In Grin, for example, a wallet may use an off-chain Slatepack address or send a Slatepack message. That wallet-to-wallet contact information is not an address recorded in the ledger.

Transaction construction is interactive, but both people do not have to remain online at the same time. Grin documents Tor-based wallet communication as one option and Slatepack as a way to exchange the required transaction data through messages such as a file, email, or chat. This adds coordination compared with simply pasting a conventional on-chain address, but it can work asynchronously. Wallet-specific workflows differ, and Beam’s current site describes both interactive wallet communication and simple send-and-forget transactions.

What cut-through removes, and what it does not

When an output created by one transaction is spent by another, an aggregated block or transaction can remove the matching intermediate output and input. The combined result can preserve the net change while omitting that link from the compacted record. This process is called cut-through. It can reduce the amount of historical transaction data that needs to remain in the same form.

Cut-through is not a guarantee that every trace disappears. Nodes may have observed transactions before aggregation, and implementation rules still retain data needed for current outputs, range proofs, kernels, headers, or other consensus checks. The original proposal, Poelstra’s later analysis, and deployed networks make different choices about compact-chain verification and retained state. Avoid describing Mimblewimble as requiring only block headers and a current UTXO list.

Privacy is a set of protections with limits

Hidden amounts and the absence of reusable on-chain addresses make some forms of public transaction analysis harder. They do not prove that transaction relationships or network metadata are always unobservable. Grin’s own privacy documentation discusses transaction-graph linkability and Dandelion, a network relay method intended to make it harder to identify the IP address that first broadcast a transaction. The protection depends on network conditions and how much activity an observer can monitor.

A 2019 Dragonfly Research post reported linking the flow of 96% of Grin transactions in one network-monitoring experiment. That is the researcher’s reported result under a particular setup; it does not mean that 96% of real-world identities were identified. Grin community members disputed the inference that observed input-output links revealed who paid whom, and emphasized that graph linkage alone did not identify people. Read the figure as a bounded, debated experiment—not a universal privacy score. Usage, timing, wallet behavior, and an observer’s network access all matter.

Range proofs are not the same thing as Bulletproofs

The original note calls for range proofs in general; it does not specify Bulletproofs. Bulletproofs are a later range-proof construction. Monero’s documentation records their adoption on its main network in the October 2018 upgrade. That history is separate from the July 2016 Mimblewimble note and does not support calling Bulletproofs part of the original proposal or claiming Grin was their first mainnet deployment.

Grin and Beam are separate implementations

Grin and Beam both draw on Mimblewimble, but a protocol proposal does not determine every choice made by a production network. The Grin project identifies its network as live and its release page lists v5.5.2 as the latest release on the date of this review. Beam’s current site describes a broader product with confidential assets, DeFi features, and a Beam Virtual Machine. Those are implementation-specific project claims, not capabilities guaranteed by the 2016 note.

The design also has trade-offs. Interactive transaction construction can make payments less familiar to users, and transaction-graph or network-level observations can reduce privacy under some conditions. The 2016 proposal is useful for understanding the goals—confidential amounts, compactable history, and a different transaction model—but it is not a security audit, a guarantee of anonymity, or a description of every current Grin or Beam feature.

Social Media Sentiment

In a March 2025 Reddit discussion asking whether Grin could become a widely used privacy coin, some participants praised Mimblewimble’s compactness and its different approach to transaction data. Others questioned how its privacy compares with Monero, pointed to transaction-graph and network-observation risks, and raised the extra coordination needed to construct transactions. This was a small, mixed discussion in a privacy-focused community, not a representative survey or a measure of current adoption.

Last updated: 2026-10

Related Terms

See Also

Sources