ProgPoW Wars: Ethereum dev call places a spotlight on the ProgPoW debate

Setting the stage for the ProgPoW debate

Earlier today, Ethereum core developers and contributors convened to discuss the highly controversial topic of EIP-1057, otherwise known as ProgPoW. The proposed change to Ethereum’s hashing algorithm has the community divided with tensions running high on either side. Supporters want Ethereum to remain friendly to GPU miners, while opponents fear the change could disrupt existing applications and lessen network security. Some believe moving forward with the proposal could set the foundation for another DAO-like network split.

Core developers hoped today’s call would quell the community infighting, which peaked over the last two weeks after discussions from the previous dev call fanned the flames on the ProgPoW debate.

Today’s discussion pinned pro-ProgPoW representatives Kristy-Leigh Minehan (one of the creators of ProgPoW) and Michael Carter (cryptocurrency educator) against dissenters Martin Köppelmann (Founder of Gnosis) and Matt Luongo (CEO of Thesis and Fold).

Pro-ProgPoW

The general sentiment from Kristy and Michael is that ProgPoW was designed to keep Ethereum ASIC-resistant, thereby making the network available to a broad range of mining devices. Their goal is also to prevent a future situation where a small number of ASIC chips control the majority of Ethereum’s hash rate. ProgPoW’s well-intentioned mission is, therefore, to preserve network decentralization.

Kristy also mentioned the rumor that Bitmain E3 ASIC miners will stop mining Ethereum next month due to the mounting storage requirements for Ethereum miners. Her team’s recent research concludes about 40% of Ethereum’s hash rate comes from these ASICs. If the storage constraints turn out to be as problematic as advertised, Kristy predicts all of the hash power E3 ASICs dedicate to Ethereum will gradually drop off of the network.

Michael argues this free fall in ASIC support would mark a “good time” to implement a change to the mining algorithm. While there’s a chance more powerful ASIC miners launch to fill the void left by E3 miners, it’s hard not to see the irony in using an upcoming event that could rid the network of ASIC miners of implementing a controversial protocol change to rid the network of ASIC miners. The storage overload for E3 miners could, in theory, play the role of ProgPoW without the contentious fallout.

Anti-ProgPoW

Martin, on the other hand, questions the perceived benefits from forcing ASICs off of the network when compared to the potential costs. ASICs can boost a network’s level of security by increasing the hash rate and, therefore, the cost of a 51% attack. There's also no guarantee a new hashing algorithm is resistant to specialized equipment, as any ASICs manufacturer can produce a new chip if there’s the potential for a profitable return.

Martin notes his issues with ProgPoW aren't technical (even though a developer under the handle @kikx recently disclosed a vulnerability in ProgPoW’s design). In his mind, it’s a coordination problem, since convincing numerous independent miners and node operators to execute a hard fork is already a difficult social task. This level of difficulty is compounded if there isn't an objective understanding of how a proposal's benefits outweigh its costs.

Matt took aim at the governance process, echoing the sentiment of many pro-status quo community members by labeling it a “governance misstep.” He mentioned how many contributors and projects, including himself, were taken aback to find out ProgPow was still under discussion. The surprise felt by those not directly involved in core dev calls points to a disconnect and the need for a better line of communication between core developers and the community.

As a final note, Matt adds that a fork on networks supporting numerous stateful projects is far from ideal. Developers nor users would have the final say in the event of a network split. Instead, the most popular and valuable Ethereum applications like Maker and Compound would ultimately determine which chain would become the “main chain.”

A settlement on the rise?

Both camps agreed on two key topics: (1) No protocol change is worth risking a network split, and (2) ProgProW would have benefited from a better marketing and user education strategy. Most projects and community members want to minimize platform risk, and a proposal vying to change the underlying security design of a $26 billion network can send the opposite signal. In response, some of the call’s participants discussed the possibility of rebranding ProgPoW (e.g., as “Ethash 2.0”) before pitching it again as a solution to ASIC domination.

Ben DiFrancesco, the founder of ScopeLift, also chimed in with what he deemed as a compromise proposal. At a high level, Ben suggests clients should implement EIP-1057 (once the code is ready) but not activate the code change. Why? ProgPoW would act as a failsafe against a potential fork initiated by ASIC miners to resist Ethereum’s eventual switch to Proof-of-Stake (PoS). It would otherwise remain dormant unless a network switch is socially accepted.

Ben’s proposal seems to appeal to both sides. For ProgPoW advocates, the new mining algorithm could play a central role in protecting the network and facilitating the transition to ETH 2.0. For opponents, it doesn’t subject projects and community members to an unwanted protocol change clouded with potential risks.

What’s next?

The comments regarding Ben’s failsafe proposal were relatively positive. Even Ameen Soleimani, whose distaste for ProgPoW is well-documented, supports the compromise model. It could represent the least divisive path forward in the short-term.

But some members remain displeased with how core devs handled the governance process for EIP-1057. Ameen didn’t hold back any of his hot takes during the discussion, saying ProgPoW’s existence is the result of profit-seeking GPU miners lobbying core devs to shift the mining dynamics in their favor. Other contributors have echoed Ameen’s thoughts since ProgPoW benefits a specific mining group over others; thus, its potential adoption “is being captured by special interests.” In their minds, the whole ProgPoW saga gives the appearance that Ethereum core developers are pushing their own agenda without considering the thoughts and concerns of the community.

The infighting showed Ethereum’s off-chain governance process could be a bottleneck on the path towards decentralization. But where there's an obstacle lies an opportunity, and the recent dilemma could result in a more effective method of communication between core developers and the community, as well as a piqued interest in Ethereum’s governance process from otherwise passive users. Crypto governance should be an ever-evolving model to maximize its effectiveness over time.

As for EIP-1057, the call left the fate of it somewhat open-ended. ProgPoW is not officially dead and could even be reborn under a new name and EIP. But the call allowed the opposing sides to find some common ground, namely that no protocol changes are worth triggering a network split. Based on this outcome, the community’s outspoken response will likely prevent core developers from including EIP-1057 into one of Ethereum’s upcoming hard forks.