Blockchain receives a lot of attention these days for some very good reasons. Some of the common attributes that are frequently used to describe Blockchain are distributed, sustainable, secure, indelible, transparent, auditable, consensus-based, transactional, orchestrated, and flexible. This is certainly a good start for a distributed ledger, but some are questioning whether this goes far enough. A recent conversation with Dr. Leemon Baird, CEO of Swirlds, shed light on some of the issues with Blockchain and the advantages of his hashgraph alternative.
This is a very important attribute and its omission from the list of common attributes describing Blockchain is troubling. A Blockchain lottery is used to award and compensate a miner for packaging and publishing a block of transactions to the community. One concern revolves around transaction packaging. Because a miner in Blockchain can exercise discretion in how transactions are assigned to and sequenced within a block, this can significantly alter the impact of settlement. Consider a situation where several people bid on the last share of stock and these bids all arrive at the same time and are for the same price. The actions of the miner in Blockchain will influence the outcome of transactions. The techniques used by the miner may or may not be fair. Baird maintains that with a hashgraph, no individual can delay or manipulate the order of transactions. In a hashgraph, transactions are ordered based on when the majority of the active community receive them. Consensus timestamping of transactions occurs when the majority of the community (as defined by the community) agrees to the validity of the block and its transactions. This approach eliminates the incentive to either bribe miners or delay a transaction from entering the system. Since the definition of fairness can vary (first in, highest bid, status, randomness, … or some combination of these) it is critical that the community select and contractually adhere to a transaction processing model that best addresses their needs.
The miner and consensus-driven approach of Blockchain mean that the time it takes for consensus is measured in minutes based. Bitcoin has a 1 MB per block limit and a settlement time that averages 10 minutes. This timing has everything to do with risk avoidance and is designed to give participants time to validate the transactions in the block. The more confirmations associated with a transaction the lower the probability of fraud. Baird’s hashgraph approach allows every community member to add block(s) at any time avoiding the Blockchain lottery. Every time a block is added, the community is made aware of the addition and once the majority of the community confirms receipt of the block and its contents, its order and addition to the ledger is finalized. This approach means that there is no governor or limit to how fast transactions can be processed. This way, if each participant has enough bandwidth to download 3,000 transactions per second, then this is the speed of the system. Hashgraph consensus times are dependent upon the time it takes the majority of the community to validate the transactions in the block.
Baird has also shown that the hashgraph algorithm is provable, which is important to ensure eventual consistency. Baird also describes the hashgraph approach as Byzantine, meaning that no single member of the community can stop the community from reaching consensus (subject to the rules established by the community for reaching consensus). The hashgraph approach also looks to be more efficient since there are no calculations necessary to slow down how blocks are mined or work that is discarded. This will ultimately make the hashgraph approach less expensive to implement.
Baird’s ability to posit a more compelling approach to Blockchain speaks to the immaturity of Blockchain. ESG believes that the outlook for the distributed ledger market is very promising and the contributions of people including Dr. Leemon Baird and vendors like Swirlds will help accelerate the capabilities and maturity of this important capability.