Introducing Rlay, a Decentralized Protocol for Blockchain’s External Data Problem

Michael Hirn
Rlay Official
Published in
3 min readJun 19, 2018


We are excited to announce Rlay, a decentralized infrastructure protocol for solving DAPPs’ oracle problem to bring people (and their computers) together to validate information via Proof-of-Coherence in the pursuit of establishing the first executable state of Reality. Its whitepaper (v0.3.0-draft) is now available for review and feedback; its testnet (v0.1.0) follows in the coming days.

Blockchain’s Reality Problem

If everyone agrees on the state of a system, e.g. a blockchain, we can use PoW, PoS, Proof-of-Storage, and co. to prove to others that some information about it is true.

But many DAPPs also use information about the real world, e.g. information about the creator of the Mona Lisa, where people have often dramatically different views of what the state of Reality is. The lack of a fully decentralized consensus algorithm that works for information about partially-observable systems like the real world has since then led to the infamous external data aka. “the oracle problem” for smart contracts.

How can we prove to others that we are telling the truth? How can we discover the actual state of the world? How can we know anything?

We understand that Oracles, Prediction Markets, and Trust&Reputation protocols try to tackle the same problem, but we don’t believe that their proposed architecture is satisfying to most applications. We aim to radically improve in the following areas:

  • Expressibility: the range of statements that can be proposed
  • Cost-of-Validation: the cost and time incurred to validate a statement
  • Attack prevention: the guarantees that a statement that is false can’t be proven to be true

The Rlay Protocol

We are proposing the Rlay protocol with a new type of architecture that allows to express almost any statement* (not just statements about future events/statements in an accessible database), validate those statements at the cost and speed of executing code on your machine, with attack prevention guarantees, similar to other blockchain consensus protocols, i.e. no collusion controls more than 50% of the voting power (the RLAY token).

We achieve this via,

  • a content-addressable ontology (derived from OWL) to express almost any statement*
  • Proof-of-Coherence to verify statements by running the Rlay client on your computer (instead of requiring human time and input)
  • a deterministic consensus protocol, developed in recent years in the field of belief aggregation and voting theory to guarantee truth-convergence when at least 50% of the voting power is in the hands of competent participants; secured via Proof-of-Coherence.

See our whitepaper to learn more.

Why are we building Rlay?

At Rlay, we believe that access to the worlds (verified) information should be freely accessible to everyone and can help us build a world where democracy and inclusion are stronger enforced and more widespread.

Our Backers

Thank you so much to the team at KR1 and OST for providing seed funding and advice to help us during the early days of our journey. Your support has helped us in the last couple months to develop Rlay from an idea into the testnet and we are very proud to build Rlay with you.

Building Rlay would also not have been possible without the larger Ethereum and crypto community, who provided many parts that inspired and make up Rlay. We are truly building on the shoulders of a giant.

Michael, Max, Erik, and Sam

We will be releasing more information in the coming days and weeks, including our testnet on OST & Ethereum, private sale contributors, and more. For now, check out our website and join our Telegram. Updates will be posted on reddit and Twitter.

If you are a developer and Rlay excites you, we’re hiring.

*any non-self-referential statement expressible in OWL



Michael Hirn
Rlay Official

Author of Rlay and Autumn; prev. Sunstone Capital