Hacker News Reader

Comment by drcode | Hacker News Reader

well... Vitalik so far seems pretty confident a sharding system is feasible, and I would be loath to bet against Vitalik. Certainly he scares me a bit at this point because he's already basically stated that he believes it's impossible to implement blockchain sharding without erasure coding, so it certainly doesn't look like sharding is going to be a simple undertaking. https://en.wikipedia.org/wiki/Erasure_code

View on HN

Vitalik's been talking about a sharding system around the corner for years now. The team working on it seems to be more interested in the theory than the practice of sharding. Additionally, any sharding design will probably have to break some existing assumptions and contracts, making it a really hard fork. I wouldn't be surprised if something else comes along offering a somewhat ethereum-compatible system with sharding for those who are working on real applications and need it.

> The team working on it seems to be more interested in the theory than the practice of sharding

As it should be for something that will need to be unbreakable and ultra-reliable. I detect a tone of distaste for theory in your phrasing, which isn't promising.

Agreed. Both PoS and Sharding are difficult problems that are still very experimental. The fact that they need to support backwards compatibility just compounds the difficulty. Then add the difficulty of deploying these new concepts on a system worth tens of billions of dollars...

I do not envy them.

> something else comes along offering a somewhat ethereum-compatible system

Dfinity comes to mind (Solidity contracts run on Dfinity) https://dfinity.network

Reply on news.ycombinator.com