TrustNote: the Minable, Directed Acyclic Graph Based Distributed Ledger and its Development Platform
A recent study from Deloitte appears to be the first empirical attempt to understand the trend of blockchain development using metadata available on GitHub, the largest software collaboration platform hosting the most important projects for the global blockchain community.
By analyzing over 86,000 blockchain projects, the report revealed that among the TOP 5 cities with most of the blockchain developments going on, there is no surprise to see projects from New York are geared towards financial services where scalability appears to be an exceptionally high priority; and there is a high level of activity in Beijing and Shanghai, where most of the projects pertain to cryptocurrencies and cryptocurrency exchanges, with an strong emphasis on scalability as well.
Why there is such an emphasis on blockchain scalability?
The current performance of blockchain applications remains far behind when comparing to mainstream applications.
For example, the current Bitcoin block size has gone from 50GB in late 2015 to just over 155GB in February 2018, Visa can process over 50K transactions per second but Bitcoin and Ethereum can only handle about 3~7 transactions and 15 transactions per second respectively.
In a Bitcoin-like world, the limit of block size and consensus mechanism have created a bottleneck, resulting in increasing transaction fees and delayed processing of transactions that cannot be fit into a block. When the number of concurrent transactions increase, the situation only gets worse. In addition to this, if each transaction is 1K byte in size and if every transaction must enter the blockchain, a Visa-like application would double the Bitcoin blockchain size in less than one hour.
To address these issues, the Bitcoin developer community has come up with solutions such as block size increases, segregated witness, and lightning networks, but unfortunately none of them is perfect and none of them have reached full agreement within the community. This means if you are developing a blockchain application interact and transact with multiple and diffuse stakeholders, you will have to address the scalability issue somehow.
The blockchain alternative: Directed Acyclic Graph (DAG) based ledger technology
In classical Bitcoin-like blockchain technology, at any given time, new transactions are grouped into blocks and are verified by miners before being subsequently added to the ledger. These only come after the first miner solves a math puzzle. In such a system, the throughput and transaction times are limited by the block size and the time required to generate each new block.
In the DAG based ledger technology, each new transaction confirms at least one previous transaction, transactions are represented as “units” in the graph and are not grouped into blocks. Because there are no blocks, there is no block size issue. In theory, by decoupling branch selection and double-spending detection from transaction verification, and allowing nodes to verify transactions in parallel, in theory, a DAG-ledger can achieve unlimited scalability.
Since 2016, new projects started to adopt the DAG-ledger instead of the classical Bitcoin-like ledger system, to accommodate the highly anticipated high-frequency trading scenarios. Projects including IOTA and Byteball have achieved some success.
However, the downside is that although DAG-ledger supports high-frequency trading, in the case of low-frequency trading, the old transaction cannot get enough new transactions to verify and reference, resulting in the old transaction cannot be confirmed in time, in extreme cases the transaction may never get confirmed.
IOTA proposed a temporary centralized actor called coordinator, which is used to protect the network when the volume of transactions is low, however IOTA does not disclose the design details of such coordinator. Byteball introduced the witnesses, implementing transaction confirmation via witness attestation (operated by ByteBall’s chief developer), although Byteball claims its users have the right to choose their own witness, but the transaction quoting rules make it very difficult for users to change witnesses if they choose to do so.
TrustNote’s New Approach: PoW + DAG
Most of today’s DAG-ledgers including IOTA and Byteball have no incentives like mining. Without proof-of-work based incentives, it’s much harder to achieve decentralized consensus and often weakening the consensus on which it is based.
Inspired by DASH’s two-tier network architecture, TrustNote takes the commonly used DAG consensus model a bit further and on the top of it, a proof-of-work based consensus scheme (TrustME consensus) is proposed to encourage fair and trustable witness node selection through mining.
Such mechanism comprising base consensus and TrustME consensus. The base consensus, also known as DAG consensus, requires nodes to verify new transaction unit and reference previous transaction units. The TrustME consensus is a unique feature of the TrustNote protocol, it requires that the sequences of all units are rigorously determined by attestor units (special units generated by miners who are elected from the super nodes periodically in each consensus round). Since the DAG ledger does not have a secure timestamp (or block number), the peers need a reliable source of transactions that are guaranteed to be generated in a defined order. This prevents transactions originating from a fraudulent fork from being confirmed and prevents double spending (a problem unique to digital currency in which the same single digital token can be spent more than once).
Everything Connected
Taking lessons learned from the team’s experiences on IoT development, TrustNote believes the transaction fees equal to the size of the data being stored is the right approach, a versatile architecture of wallet design and multiple types of node protocols are developed to support Internet-of-Things and as many device types as possible:
- Super node: super nodes run on PC, server, Cloud host or mining systems, have better resources, hold a certain number of tokens, and have a well-behaved history on the record. Super nodes can host micro nodes. Miners are elected from the super nodes periodically in each consensus round.
- Full node: full nodes store a full copy of the ledger, fully enforce all the rules except don’t participate the mining operations.
- Light node: light nodes store a light ledger instead of a full copy, light nodes are easy to maintain and with support to smart phones or tablets in mind.
- Micro node: no ledger stored on micro nodes, transactions are delegated to super nodes, micro node is designed to support embedded devices or IoT devices.
Verifiable Smart Contract
Inspired by Byteball’s declarative smart contract, TrustNote uses advanced declarative non-Turing complete smart contract not only supports complex application scenarios but also drastically reduces the difficulty of secure contract development. TrustNote believes this is one of the largest barriers to blockchain’s mainstream adoption. In addition, aiming to reduce the number of vulnerabilities residing with today’s smart contract systems, TrustNote wants to take a new approach of making some restrictions on the runtime system, instead of limiting what languages can be used.
What’s Next?
TrustNote has completed its private token sale on January 8, 2018. TrustNote’s token (TTT) is available for trading on Bit-Z exchange since April 26, 2018. The team is working on get TTT listed on more cryptocurrency exchanges because it helps TTT to get adopted by the masses.
Without question, there needs to be a better technology solution in the core of blockchain protocols. What TrustNote did has certainly answers the underlying issues that face many legacy blockchain systems, as blockchain applications continue to be adopted in the mass market, and these so-called third generation platforms of blockchain technology are tussling to become the next Ethereum and, among those companies in the race, TrustNote’s approach would certainly lead the way to more scalable, faster and cost effective transactions!
