Rethinking Proof of Work: The Quest to 'Improve' Bitcoin Heats Up
In a world barreling onward in the name of efficiency, bitcoin stands out for the purposeful way it's anything but.
This 'deficiency' was the focus of many talks at day two of Stanford’s security conference, Blockchain Protocol Analysis and Security Engineering 2017, this week. For all its benefits, the system that secures bitcoin, proof of work (PoW), is estimated to require as much electricity as all of Denmark by 2020, and many researchers see this as a prize problem to solve.
But, the bad news is that, so far, it's proven difficult to find an alternative that provides as much security as bitcoin's massive collection of global computers, all brute-forcing the network in an effort to win its rewards.
At least, no alternative of has been successfully deployed on a large scale.
Nonetheless, the race to invent a blockchain security system that's less resource intensive, and that could still thwart double spending in open blockchain networks, continues, and attendees see the experimentation as a good sign.
Founder and CEO of Protocol Labs Juan Benet perhaps best summed up the sentiment that something else will eventually be discovered if enough bright minds attack the idea from all angles.
Benet told CoinDesk:
"Proof-of-work is a really good system, but there’s probably going to be something better."
Proof of blank
Perhaps the more experimental ideas presented were 'proof-of-space' and 'proof-of-time', described by BitTorrent creator Bram Cohen.
The idea is to rely on proof that you have space, (such as free hard-drive space on your laptop, for example) and conventional time, which he believes could serve as more democratic resources that could help decentralized economic systems prove ownership.
The idea is in the early stages, and Cohen walked through a lot of variations of the proposal, each with some unsolved problems, or ways it can be attacked. It’s all theoretical right now, and there isn’t any code.
Again, the reasoning behind the proposals is that it could serve as an energy-efficient alternative to proof of work.
"Blockchains fundamentally require waste," Cohen said.
Elsewhere, ethereum developer Vlad Zamfir gave an update on Casper, the proof-of-stake protocol that ethereum plans to one day switch to, and that aims to replace its existing proof-of-work protocol.
Still, his talk provided a window into the game theoretic challenges that the ethereum team is still struggling against as it strives to ensure the right balance between competition and cooperation in its network.
Ahead of science
Yet, despite experimentation with new protocols, others are taking a step back.
Cornell associate professor Elaine Shi's research, for example, has focused on gaining a deeper understanding of why PoW actually works so well.
"The protocol's success is ahead of the scientific understanding," she said.
Although it works and there has been some progress in understanding the security benefits, she believes right now people often rely on intuition to describe why it works.
She noted, as others did, that "the biggest elephant is power consumption". But, she also mentioned that throughput is relatively low, and that there might be other downsides.
Maybe a more formal understanding of how bitcoin works would make it easier to understand if another consensus protocol could take its place.
"It would be interesting to understand, are some of these constraints inherent or are there ways to do better? It would be great to understand theoretically what is possible and what is not possible," she said.
Still, not everyone believes that the work to rethink proof of work is advancing enough to offer a viable alternative.
ETH Zürich PhD student Arthur Gervais noted that most of the proposals are quite theoretical right now, and that it won’t be possible to know how successful they might be until there are deeper specifications to facilitate better analyses.
Still, Benet believes that this is only a matter of time and study.
"There’s a lot of belief in proof-of-work that is more based on agreement rather than rational, proof-oriented understanding," he said.
He noted, for example, that miners will have an incentive to push for the continued use of proof-of-work, given that the business of collecting rewards in blockchain-based systems is still lucrative.
Ultimately, he concluded:
"The reliance on proof of work will continue for a while."
Featured bitcoin image via Shutterstock
Review: Putting Ledger's New Bitcoin Hardware to the Test
Zcash + Ethereum = ♥: Why Two Blockchains Are Evolving...