Hacker News Reader

Comment by throw8932894 | Hacker News Reader

> I need a hero, and by that, I mean that I need a usable methodology for building scaleable decentralized apps. Yes, you've heard that right. We don't need more "Ethereum killers" that can do 10x more tx/s than Ethereum. Those are useless.

> Instead, we need an approach for the average Joe developer to create their idea within the Ethereum ecosystem without the need for hardcore unproven technologies.

So instead of fixing the actual problem, Ethereum scalability, we will do what? This technology was developed by a few guys, most of them already left. It was meant to fix Bitcoin "smart contracts" (there is a such thing).

Tech behind Ethereum is outdated. Even newer versions rehash the same broken idea with sharding. We got 10 years and billions poured into consensus research. There are actual papers and universities doing this stuff. There are github projects that do stuff like decentralized auctions, mixers...

View on HN

Hey, I think you're making a valid point and I probably should have explained more what I meant by my statement.

IMO, there's a middle ground we can go between slow PoW on L1 and scaleability. I think that it can be done by having tools that give us the same guarantees like deploying a smart contract on the main chain. It's difficult to explain: But what I want is Plasma but as a framework to develop dapps.

This project is going that route: https://github.com/hoytech/quadrable

If you want a Layer2 plasma with similar guarantees to Ethereum (account based, not zero knowledge) then you may like Gluon.network once they roll out EVM support.

We're building this, but on Bitcoin ("Plasma but as a framework for dapps" [with a focus on developer experience]) [0]. There's already robust infrastructure for L2 contracts in Bitcoin-land, and we've already heavily optimized the L1 to prepare for the load that a global, ubiquitous solution would need.

Honestly, I see no use for Ethereum (or other Turing Complete L1s) looking forward — smart contracts can be purely peer-to-peer, with the only pressure applied to L1 being dispute resolution. By pushing complex contracts up into Layer 2, we can keep the dangerous, money-destroying, theft-enabling Turing machines away from the main chain.

[0]: https://fabric.pub

You might consider taking a look at Celo. They have ultra-light client support [0], it's proof-of-stake with single-block finality, and they run a true EVM that you can program in Solidity. It's mobile-first but since it's fundamentally a fork of Ethereum it is also web3 compatible -- there's a metamask fork here that should soon be functional [1]. Gas is payable in multiple ERC20 tokens, and should remain cheap as the network scales. Not sure if this falls under your "ethereum killers that can do 10x transactions/second" so apologies for shilling if it's not interesting to you.

[0] https://docs.celo.org/celo-codebase/protocol/plumo

[1] https://github.com/dsrvlabs/celo-extension-wallet

> Tech behind Ethereum is outdated.

Well Eth 2.0 is applying the latest and the greatest in Crypto research to a real coin.

In terms of scalabliltiy Ethereum is already way past bitcoin slow transaction rates.

The idea of a new version of your currency somewhat speaks against the utility of the whole thing surely?

Why Does it speak against its utility to update to a new version? what is the problem with printing a new dollar bill, or everyone using windows 10 instead of windows seven?

There is utility in being able to evolve the protocols without giving up the integrity of the previous version.

Eth2 isn't a different currency, it's a replacement system that links with Eth1 to provide continuity. The ETH coins are not being replaced; both systems transact in ETH and are linked. That's why they are being careful with how it's evolved and how the economic links and incentives, and smart contracts and data sources, are set up in the transition.

There are many changes being implemented through this evolution. From PoW to PoS, single chain to cross-linked sharding (while managing the incentives to maintain the integrity of the network), various improvements to faster synchronisation, lower storage requirements, distributed storage of the state.

Being able to release v2, v3, v4 of software while staying compatible with all the work you did using v1 is a good thing. Same idea with evolving Ethereum.

Contrast it with the numerous new coin networks that use newer cryptographic ideas but each start their own separate networks. That's also evolution, but they don't provide continuity in the same way; instead they are a series of competing products.

Absolutely not. ETH2 is poorly named, it is an "in-place-upgrade" whose core design principles (PoS, Sharding) have been part of the ETH roadmap since 1.0.

USD has gone through many iterations in its history. Even quite fundamental ones.

If I had to choose between scalability and better (more useful, more accessible, more maintainable) apps I would pick the apps. Ways to do real-world interesting things for people, buildable without any rocket science.

For the same reason that a startup with duct-tape code, scrambling to scale is better than a startup with no genuine demand.

What are good examples of what you talk about in your last paragraph?

Reply on news.ycombinator.com