Conversation

i had always hoped that interoperability of chains wouldn't devolve into these kinds of "bridges". always thought it would be cool if the keys lived on chain2 so a simple action on chain2 would move BTC instead we just get tBTC and variations relying on "signers" and bonds...

Quote Tweet

toly

@aeyakovenko

Oct 9, 2020

Brain dump on bridge security: Bridges are nothing special, they are a form of decentralized custody. There are two main attacks: 1. Collusion 2. Intrusion (attackers hijack the private key) And these attacks are implementation agnostic. It makes no difference (cont)

Show this thread

SBF

@SBF_FTX

Replying to

@austerity_sucks

yeah you can do it that way -- it's just *hard* though I think not as hard as people make it seem? guessing if we worked on nothing but that for a month we'd have something

10:05 AM · Oct 11, 2020Twitter Web App

Replying to

and

is there a high level whitepaper which actually describes how this is feasible with no handwaving? ive only seen atomic swaps doable, not actual controlling assets on 1 chain from another chain (specifically, btc). guess theres complexities with key generation/network interaction

Replying to

hm so BTC is way harder than others--would you accept solana <> ethereum? with BTC you have to structure everything as micro-swaps which just makes it massively clunkier for custody like things.

Show replies

Replying to

and

which makes me wonder why the others are so slow etc. but ya it's unfortunately hard enough that I wouldn't get around to making fully on-chain bridges for another 6-12 months

Replying to

and

There is no free lunch :). BTC is unfortunately a pia. My preference is for atomic swaps, which move the settlement risk until after the trade is executed, since it sandboxes the settlement risk to just the trade.

Show replies