Balancing blockchain’s impossible trinity
In the context of public blockchains the trilemma or impossible trinity is the idea that a blockchain can have decentralization, security and scalability to varying degrees, but it cannot have all three to sufficient extents at the same time. As the most significant example of a public blockchain, Bitcoin is widely considered to succeed in security, having never been hacked, and decentralization, both in governance and development, but insufficient scalability due to its limited block size and comparatively long block time. One block is mined every ten minutes, and each block is limited to 1MB. When Bitcoin experiences spikes in volume, this has led to delayed transactions and increased recommended fees as an incentive to have a transaction processed faster.
The obvious solution to scalability issue is to increase the block size or the block time, but both of these have important repercussions to the other two corners of the trinity. More expensive machines would be needed to mine bigger blocks competitively, leading to increased centralization of mining pools. A reduced block time, say for example reducing the time between blocks from ten minutes to one minute, seems on the face of it a far more practical and beneficial solution, but this will result in more stale blocks and again the increased demand for processing power will drive centralization, along with a host of other considerations like adjusting mining rewards. Until one of these limitations are addressed, periods of increased volume will cause transactions to be stuck in waiting queues unless they set a very high Satoshi/byte miner fee.
Other projects can boast much faster transaction times and greater block sizes, leading to fewer delayed transactions. These advantages have not been tested to the same extent, however. Andreas Antonopoulus discussed this with relevance to the idea of Bitcoin becoming obsolete, whether due to hard forks or competing currencies. When other projects reach the volume and value of Bitcoin at its peak, then they will be able to prove their readiness to achieve supremacy. Accordingly, it is easy to claim lower transaction costs — Bitcoin also had near-zero fees in its early history. It was not until volume increased and blocks became full that the fees started to rise. Miners have no reason not to accept low fees when blocks are less than half-full.
More importantly, those larger blocks and reduced block times come with significant trade-offs. EOS, Stellar, Neo and Ripple are all blindingly fast at processing transactions, but are hamstrung by a hard cap on nodes and are all insufficiently decentralized. Not so for Ethereum, but Eth has suffered similar scaling issues (the Cryptokitties case being cited frequently) and this will surely worsen, given its use in smart contracts.
Bitcoin’s failure with scalability, as it were, has facilitated its unprecedented security and relative decentralization. The idea that a completely transparent store of immense value has not been victim to hacking in almost ten years is remarkable, and makes understandable the hesitancy to ‘rebalance’ the triangle, adding weight to scalability, instead relying on second layer solutions like the Lightning Network. Scaling debates and further hard forks may be inevitable, but it is clear the majority of participants in the Bitcoin network will not lightly trade much for increased scalability on the first-layer level.
Article by Byron Murphy, Editor at Viewnodes. For information on some of the services provided by Viewnodes, including our Tezos delegate, click here.