Developers From Parity And ChainSafe Call For Universal Ethereum (ETH) Test Network
On September 13, Parity Technologies Afri Schoedon posted online, a call for users to participate in creating a more universal testnet. The idea appears to have emerged at the recent ETHBerlin hackathon, where Schoedon worked alongside key members of the associated Chainsafe team.
It was in this proposal that he noted the existing testnets which otherwise do not adequately support all clients and aren't exactly known for being,
“robust enough to guarantee consistent availability and high reliability.”
Instead, he proposes the community come together in order to build Görli, which operates as a public Ethereum test network.
It was Schoedon and his preferred specifications are that clients choose . a proof of authority (PoA) engine, such as Aura or Clique, and implement it. And as it stands, Parity and Geth already use PoA testnets for efficiencies sake. The existing system for the proof of work net, Ropsten, is already technically compatible with both Parity and Geth, but has since resulted in a number of reliability issues.
PoA chains, in contrast with others, doesn't require as much in the way of participation or hashpower (or any hash power at all compared to other systems), making them the more ideal and reliable option.
At this moment in time, Geth and Parity use different testnets with different Proof of Authority algorithms.
However, if the two can otherwise agree to one algorithm and implement it, then the community could,
“bootstrap a new Görli proof-of-authority testnet based on the available implementations that mimics main network conditions.”
In conjunction with the protocol for Ethereum development, progress on Görli would begin with the development of an Ethereum Improvement Proposal (EIP) to otherwise specify the PoA engine clients should implement to then build the testnet around. Regardless of whichever algorithm is ultimately chosen, clients intending to use the . Görli testnet would need to implement it effectively.
Schoedon links this all to an EIP in development that currently uses Aura, which is likely due to the fact that Aura is a consensus algorithm which was pioneered by Parity and used to support the Kovan testnet.
Geth, which is the other Ethereum client, and uses Clique for its Rinkeby testnet. Schoedon is otherwise careful in specifying that this is not a political choice, however, there does seem to be some underlying tension around this statement on Twitter.
Clique already has a full speced EIP https://t.co/KZNNq5Z01J
The article goes about presenting both engines, but then pushes only Aura whilst completely omitting the fact that Clique is speced and easy to integrate. It's imho a bit dishonest. https://t.co/0pHw5KIV0W
— Péter Szilágyi (@peter_szilagyi) September 13, 2018
Aidan Hyman, the CEO and Co-founder of Chainsafe wrote earlier in a Medium post that he and his team is currently working on implementing Aura, the PoA algorithm behind Parity, in Geth. Hyman explains that they have decided on using Geth instead of Parity all because they happen to be more familiar with Go, which operates as the underlying language behind Geth, instead of Rust, the language which is used to write Parity.
Hyman does however take the time to re-state that the ‘Görli team' also plans to implement Clique in Rust, just not as expediently as Geth. In an email, Hyman did go on to clarify that the individuals from Chainsafe would be directly involved in the wider implementation of Clique-Rust, but that development is . also open to anyone that is otherwise willing and able to participate.
Political of otherwise, it's at this very early stage that there seems to be more work required in order to bring Geth over to Parity's algorithm as opposed to the other way around. Whichever engine ends up winning out of the two (if either indeed manages to do this) will hinge upon who ends up contributing to the development process more.
However, once the clients can make the first agreements on a consensus algorithm and any additional necessary client-facing specifications, Hyman and Schoeden do seem to uniformly suggest that building the testnet itself will be a fairly straightforward element.
One of the notable differences between Chainsafe's and Schoedon's framing of this initiative is that he emphasizes that this is to serve as a more universal testnet, open to any and all clients, whereas ChainSafe will only be focusing on Geth and Parity.
On the other hand, in a follow up email, Hyman did go on to clarify that while they are only focusing on these two clients at the moment, ChainSafe also intends that Görli be usable by all clients. Though, as is the case with any decentralized project, this is dependent on developer participation.