Hacker News new | ask | show | jobs
by lewi 1603 days ago
While the Facebook driven project is dead, an open fork of Libra is currently running and being built on.

https://github.com/OLSF/libra

2 comments

What would this be good for? Isn't Libra supposed to be backed by USD held by the validators in the network?
The stablecoin and the core blockchain tech are totally different things. 0L is using the blockchain tech (move language, the Rust BFT implementation, etc), and has no stablecoin -- its currency is like any other decentralized blockchain base currency.
> its currency is like any other decentralized blockchain base currency.

You did not answer the question of what is it good for. What thing is it doing that makes it important/interesting/useful/valuable?

Sorry, the gist of the question I thought was "is this not useful, _because_ it is based on a stablecoin?". I answered that question.

If you're asking what in general the Diem code, used in the 0L project, is useful for : 1. well it's technically a BFT blockchain, like Tendermint. That tech is potentially useful because it can provide a decentralized ledger like a PoW blockchain, but with significantly higher throughput and significantly lower energy use. It's arguably more advanced than Tendermint because it's written in Rust and has a native smart contract VM (move language). and perhaps 2. it's a collaborative project -- there's no sponsoring company or foundation or fund raising with a token and what not. This is attractive to some for various reasons. Bitcoin had that property and some people are interested in blockchains with new tech that share that organizational aspect.

Let me know if that doesn't adequately answer the question.

Is its performance any better? Diem was topping out at 24 tps, which is dismal for a permissioned blockchain.
0L transaction load today is quite low (see : https://0l.interblockcha.in/) so I'm not sure it's known what the current capacity is, but I'd expect this work : https://cloudflare-ipfs.com/ipfs/QmegyuTcvDJG94PaLGGErEmH6Bm... to be continued in the 0L project. Useful throughput is obviously a core feature of a BFT chain.
24 is not very good. I just want to point out that a lot of chains that report high tps are cheating in one way or another. I won’t name names but some of the “highest throughput” chains count their node synchronization messages as transactions, not transactions written to the ledger. It’s really disappointing when you understand the games they’re playing.

Real tps has to be tx written to the ledger per second. Anything else is dishonest.