Abstract
Lineage proposes Universal Turing Market Machines (UTMMs): a framework for blockchain-based markets in which market policy is a programmable, adaptive object. Transaction history is treated as an epistemic substrate, with deterministic feature extraction, evaluation of policies in a constrained space, and improvement via bounded search, all while remaining within classical computability limits.
The flagship whitepaper, Universal Turing Market Machines (UTMMs), is archived on Zenodo and anchors the reading list below. Surrounding papers develop the consensus layer that defends these adaptive algorithms, the world-state model that carries context in a UTXO setting, and the currency design that aims for low volatility without external collateral.
Research themes
The published work clusters into a few recurring questions. First, how a market can compute: how to express market policy as an object that adapts at runtime rather than being fully fixed at design time, and how to keep that adaptation deterministic and verifiable inside consensus.
Second, how that computation is defended: the Dynamic Proof of Weighted Work (DPoWW) family of papers studies a Nakamoto-style design that can move between fully distributed and semi-decentralized modes for throughput, reverting to full distribution when fault conditions require it, including parallelized Merkle-tree validation to scale miner workloads.
Third, how markets evolve: framing self-organizing market process through Hayek’s spontaneous order and Ashby’s law of requisite variety, with adaptive resource-coordinated organisms (ARCOs) combining sensory inputs, ledger signalling, and actuators. World state is carried in a UTXO setting using monad-style composition to limit side effects, while coin scripts stay non-Turing complete for value movement.