Ivo Kubjas

@yelhousni - actually, maybe we could change interfaces such that the hash functions are a property of the key, not runtime parameter to `Sign` and `Verify`. So, when we have...

I am not sure, I haven't been involved in the implementation, but I would assume the main reasoning would have been to have compatibility with alternative libraries. Length-extension attacks are...

I'm not sure I follow completely. Shouldn't the interface be instead: ```go type Signer interface { Sign(msg []byte, hFunc hash.Hash) ([]byte, error) SignNum(msg *big.Int) ([]byte, error) } ``` Otherwise I...

Following up from https://github.com/ConsenSys/gnark-crypto/pull/313#issuecomment-1408814469, I would propose instead changing the functions for generating/unmarshalling keys for making prehashing hash functions/Fiat-Shamir hash function etc. a property of the key instead. So, instead...

> @ivokub I agree with your argument for key options vs many parameters, and I'm starting to doubt the necessity of exposing a "sign custom string" function in the first...

@ThomasPiellard - I don't touch the branch currently - I'll add the documentation and examples after you have finished.

As a note - we discussed with @ThomasPiellard about how to compute the n-ths roots of the input points. But in the field implementations we have, `n` is usually a...

A few TODOs for another PR to improve the packages: * instead of using `[]fr.Element` for points in shplonk, we could define a structure a la `struct Points { Pts...

I'll start creating issues for every subtask for better organization.

cc @tmpfs, @davidsemakula, @drewstone - any issues to add/remove?