Are we there yet? You can hear the Ethereum community asking again and again. The destination is the promised land, where Ethereum can match VISA’s 45,000 transactions-per-second (TPS). And in
Before Devcon 4, the annual developers’ conference which was held in Prague last week, Buterin tweeted that the spec for Casper, the scaling solution, is “roughly finished” and that a VISA-like destination is “not really so far away.” Casper is designed to increase the network’s capacity from 15 transactions per second to 100,000–through techniques such as one called “sharding,” where blockchain data is split across multiple locations.
At the conference meanwhile, he outlined the vision for Casper and other upcoming changes under one banner, Serenity.
But why are some members of the community impatient?
Sid Coelho-Prabhu, product lead at Coinbase Wallet, tells Decrypt that Serenity is, “probably a couple of years out at the minimum.” The wallet, designed to be a holder for ERC-20 tokens and a Dapp browser, is being held back because any attempt at boosting numbers is undone by the slowness of Ethereum.
One member of a project working on Ethereum, who wishes to remain anonymous, recalls early Devcon events
Others say the change of pace is justifiable.
“We’re getting our noses to the ground. We’ve got infrastructure and usable apps but we want to get it world class,” says Nick Johnson, core developer and creator of the Ethereum Name Service. “We need to polish the hell out of it.”
Nabil Naghdy, COO at Status, points out that taking the time to get everything right is important to achieve true decentralization: “The moment you start compromising on certain decentralized elements, you start recreating the web. You can’t take shortcuts.”
While the debate continues, there are other solutions that people feel can get Ethereum off the ground sooner rather than later—”state channels” and “sidechains,” for instance.
The mysterious layer two
Fixes to Ethereum’s scaling issues are often grouped into layers. Layer 1 solutions affect the base level of the protocol, a typical example of this would be Sharding. Layer 2 solutions meanwhile are things that are built on top of the blockchain–sometimes referred to as ‘off-chain’–rather than changing the nature of how it works.
State channels and sidechains are considered layer two solutions. Both of these have the potential to drastically scale the number of transactions Ethereum can process. State channels solve the issue by keeping track of the payments offline–while remaining cryptographically secure–and then uploading the final bill, or settlement, to the blockchain. Sidechains, meanwhile, involve the use of separate blockchains that can handle more transactions.
Ameen Soleimani, CEO at SpankChain, explains that with state channels, “You get instant payments that are off-chain and don’t require any gas. The user experience of instant is the best. You can’t beat instant.” One of the most used decentralized applications, SpankChain ought to know since it already uses state channels so penny-pinching users can tip its adult performers with small amounts without incurring large fees.
Yet any fast off-chain solution is usually met with
Josh Stark, co-founder at web3 development firm L4 Ventures said sidechains, in theory, are perfectly safe. “As long as users are following the protocol and they can access the layer one, we are not introducing any additional risk,” he said, during a speech at Devcon 4 in Prague last week. So, as long as there’s access to the blockchain, users can upload transactions at any time and settle the tab. This means, as long as the payment records are accurately kept, they are secure.
Another element of layer two is sidechains. These are separate, more scalable, blockchains which are used to free up space on the main blockchain. Coelho-Prabhu, recommends that developers will need to either adopt state channels or
However, the alpha testing version of the Raiden Network has still not been released despite indications it would do so in Q3, 2018. Serenity appears to be far off and sidechains are still on their way, despite Buterin being broadly in