Today we’re gonna talk about Ethereum. Lately, there was many news reports about its planned updates. We have to remember that it's a long-term plan, and we have to wait 3—4 years to see all these updates deployed on the mainnet, so now we have time to quietly analyze, what features will it have, how will it work.
What are the most fundamental changes, that Vitalik and his team plan to implement?
We can count them all:
All of them are important, but what does mean each of these terms? We'll start with PoS.
Currently, Ethereum runs on Proof-of-Work algorithm. For starters, it’s the same consensus model, that is implemented in Bitcoin, where block mining depends on expensive hardware resources and requires large amounts of electricity. The biggest disadvantage — PoW systems can’t scale. Proof-of-Stake isn’t so resource-demanding, it depends only on the amount of tokens on validator’s account.
Switching to Proof-of-Stake allows to save energy and scale effectively, reducing the time between blocks and increasing the bandwidth of the Ethereum Network. After the update, Ethereum would be able to handle at least one hundred transactions per seconds, instead of 17, that we have now.
Upgrading Ethereum isn’t an easy task, because it’s actively used every day, the number of daily transactions now exceeds 500k, with the peak of 1.3 mil transactions on January 4, 2018. If something screws up, it’s not hard to imagine what would happen with the network, worth billions of dollars.
That’s why the major upgrade has to be tested carefully before the launch. This release is called Casper v2.0, it’s gonna be deployed in 2019—2021, and it’s gonna be crucial for another update, which is next in our list.
Sharding heavily depends on PoS. To understand, why it’s not possible on PoW blockchain, let’s figure out, what’s this. Sharding is the technology, allowing to divide the whole network to many small pieces and divide the block validation between them. Thus, instead of 1000 nodes validating 10 transactions you could get 10 shards, each one containing 100 nodes, and validating its 10 transactions per second, which gives us 100 tps in total. It’s the same Ethereum chain, but the nodes are more effectively used. Why it’s impossible on PoW? Since a shard is much smaller than the whole network, attacking it, and gaining most the majority of computational hashpower requires fewer resources. PoW can’t restrict miners from applying their efforts to any chosen shard, so they can attack it anytime they want. At the same time, PoS Sharding algorithm regularly reshuffles validators, so malicious actors don’t have an opportunity to choose the shard, hence they are unable to concentrate their efforts on one weak point.
Sharding is a tough nut to crack, so it also might take a while, before Ethereum development team will roll it out. It should be released together with Casper or after it, in 2019—2020.
Plasma is a Layer 2 solution. It allows infinite scalability, by creating child chains that don’t need to record every transaction on the main chain. It’s similar to already working Lightning Network on Bitcoin, which isn’t surprising, given that Joseph Poon, Lightning Network author, works on Plasma too. Plasma is capable to handle millions of transactions per second, because it doesn’t have the bottleneck problem — the state of Plasma chain can be settled on main chain once per 10 minutes, or once per hour, and it can use Sharding too, creating child chains on child chains, thus the scaling potential is pretty much infinite. The first project, using Plasma, will be OmiseGo, its developers. The Loom Network team also announced their intent to build their project on top of Plasma.
All these technologies are great, they are the foundation to build upon, and Ethereum community has many brilliant developers, so they would definitely like another technology, eWASM.
It stands for Ethereum WebAssembly, and it’s a new, improved version of Ethereum Virtual Machine. Why is it important for Ethereum development? Current Ethereum Machine is a bit limited. It supports only its own language, Solidity, forcing Ethereum developers to learn a new language, which is also limited. eWASM will allow using multiple programming languages, making it more useful for developers. Most modern blockchains allow this, for example, Ethereum’s direct competitor, EOS, already has the support of multiple languages.
The network will greatly benefit from this new virtual machine.
The future for Ethereum
Currently, Ethereum has the largest community of developers. More than 1,906 decentralized applications are built on top of it. There is no sign that it may change in the future, it’s possible that Ethereum will hold the crown. All these updates will allow running unlimited amount of decentralized application, creating a new version of the Internet, Web 3.0. Ethereum doesn’t sacrifice decentralization for it, resorting to delegating Proof-of-Stake, which is easier to implement but lacks decentralization, that’s why it takes so long to develop it. But in the end, we’ll see how great it will benefit everyone in the long run. Corporations, governments — all of them are interested in implementing Ethereum-based solutions. And by the looks of it, a future where every major corporation uses Ethereum is pretty possible — if they succeed in scaling.