Ah yeah so for BTC you can't use it for cross-chain tokenization, you have to use the SerumBTC model.
But for Solana/Ethereum OR works.
Conversation
The essence is that each chain can parse the others' history, so each knows what the other does and you can program smart contracts like "release the ETH if the SolanaETH hits the redemption address on the other side".
1
6
There are a lot of tricky things there! But you can solve them all if you're careful.
E.g. SOL is faster than ETH, so how can ETH parse SOL's history? (Basically, you pass a hash which can be challenged.)
1
6
What happens if you're asleep and so don't challenge? Well you could stay up :P
But also if _anyone_ submits the correct history it will win, and so all it takes is a single honest node to make sure the right thing happens.
1
4
Right, yes, I get the basics. But the details matter, and they’re part of the reason why L2 isn’t widely deployed yet for Ethereum. You’re essentially saying “we can use techniques championed by the ETH community to turn Solana into an ETH sidechain”, which is true, but
1
1
but the devil is in the details, and other projects seem to have made a lot of progress on those details. Serum is said to launch within weeks, would that include “trustless” ETH pegs?
1
3
Totally understood!
Guessing that'll be in ~a month or so, we'll see.
And yeah the details are really messy. It took us something like 300 hours of talking about it to understand them all (that's what happens when you have 5 math majors who haven't gotten their fix in years).
1
6
And in some situations it'll require collateral on a smart contract chain, etc.
But you can punish defectors enough that it's incorrect to lie, which means usually you don't need to go through that.
1
4
One key thing about Solana vs side-chains is that it's a decentralized blockchain with tons of independent validators etc., and so you don't need to trust that the owner isn't fucking with you.
Also it's really fast so there are 0.5s instead of minutes in which you can go back.
2
6
Really the core win here, in some sense, is:
1) actually spending the time to work out the details of OR
2) finding a decentralized chain that is also fast and so everything is cheaper and easier
I used to think (2) was just a tradeoff vs security but I was wrong.
1
5
There are actually lots of ways to make a blockchain faster without sacrificing much--just optimizing the underlying structure for performance.