4 ms·

Folks here might be interested in Chia's smart contract language: https://chialisp.com/ https://chialisp.com/ Basically the designers saw how easily many Ether

by Glench 5y ago

Folks here might be interested in Chia's smart contract language: https://chialisp.com/ https://chialisp.com/ Basically the designers saw how easily many Ethereum/Solidity contracts were hacked and saw that stateful side effects were the main culprit so they made their language have no side effects! Pretty cool.

hanniabu 5y ago
> they made their language have no side effects Yeah, I doubt that
- cossovich 5y ago
  It's not that they "made" their language have no side-effects, it's more that they made a choice to use a lisp/functional design for the language which gives the advantages of that family of languages languages... advantages like limited side-effects and formal approaches for proving "correctness" of programs. It's a really clever design choice IMO for a smart-contracts language.
qwtel 5y ago
I've been wondering if non-turning complete languages might be the way to go for many smart contracts, which often resemble vending machines more than what general purpose languages are used for. Those should be captured by pushdown automaton or even a finate state machine and the corresponding languages (but except from regex, I'm not really aware of any..). My CS knowledge got really rusty, but I think it should be possible to put tight bounds on the computational cost of these, as well as statically guarantee that certain states cannot be reached, both of which would be very valuable in this case.
ketamine__ 5y ago
Tezos has a similar pitch (having a better language that is less bug prone) and is a ghost chain now.
SuchAnonMuchWow 5y ago
But what you really want for smart contract is a language with integrated formal verification. The ability to formally prove some properties of your smart contract is much valuable than functional vs imperative paradigm. Functional programming might makes it easier to detect flaws in your programs, but when there is billions at stakes and no recourse in case of a bug, you really want your program to be flawless.

by Glench 5y ago

hanniabu 5y ago

> they made their language have no side effects Yeah, I doubt that

cossovich 5y ago
It's not that they "made" their language have no side-effects, it's more that they made a choice to use a lisp/functional design for the language which gives the advantages of that family of languages languages... advantages like limited side-effects and formal approaches for proving "correctness" of programs. It's a really clever design choice IMO for a smart-contracts language.

qwtel 5y ago

I've been wondering if non-turning complete languages might be the way to go for many smart contracts, which often resemble vending machines more than what general purpose languages are used for. Those should be captured by pushdown automaton or even a finate state machine and the corresponding languages (but except from regex, I'm not really aware of any..). My CS knowledge got really rusty, but I think it should be possible to put tight bounds on the computational cost of these, as well as statically guarantee that certain states cannot be reached, both of which would be very valuable in this case.

ketamine__ 5y ago

Tezos has a similar pitch (having a better language that is less bug prone) and is a ghost chain now.

SuchAnonMuchWow 5y ago

But what you really want for smart contract is a language with integrated formal verification. The ability to formally prove some properties of your smart contract is much valuable than functional vs imperative paradigm. Functional programming might makes it easier to detect flaws in your programs, but when there is billions at stakes and no recourse in case of a bug, you really want your program to be flawless.