Andreas Fackler

Clippy can now also be installed via rustup (instead of `cargo install`), so you don't have to manually match Clippy versions to the corresponding Rust nightlies anymore. And it will...

Thank you for taking on this issue! Yes, ideally batch size, threshold encryption (see #209) and subset optimization (see #243) should be dynamically configurable, and allowing a Honey Badger restart...

The threshold encryption settings are [already dynamic](https://github.com/poanetwork/hbbft/blob/059e4471cb3581489473ff29b93c233373ddf752/src/dynamic_honey_badger/change.rs#L17); other settings could easily be added to that. But I'd say it's not a high priority at the moment.

It has been replaced/renamed, but it's still [spaghetti code](https://github.com/poanetwork/hbbft/blob/059e4471cb3581489473ff29b93c233373ddf752/tests/dynamic_honey_badger.rs#L89).

Yes, that's a clippy lint. Also, `cargo clippy` doesn't check the test code. You need to use `cargo clippy --tests` or `cargo clippy --all-targets` for that.

I'll update this once https://github.com/poanetwork/threshold_crypto/pull/102 is merged. But before we do that, let's make sure this version will work with OpenEthereum.

If I remember correctly the `hbbft` crate doesn't do that _automatically_, but gives the caller the option to do so. E.g. you could imagine using `hbbft` in a blockchain context,...

Good point! I think that _all_ messages back to the proposer are redundant: We might as well have the proposer output right away. (I'm not convinced there's much to gain...

Exactly: In the optimistic case, we want to avoid the unnecessary redundancy. Of course we never want to compromise on security and asynchronicity! But I think my proposal wouldn't: It...

Yes, exactly: If there are more than _g_ faulty nodes among the _N - 2 f + g_ ones to your left, you will receive fewer than _N - 2...