Simon Warta

Thank you. This is a nice approach for sure. I tested the full removal again and it turns out we don't need the extra call anymore. See #1772 which will...

See also https://github.com/cosmos/cosmjs/issues/1897. I think the core issue here is that `Decimal` from CosmJS was never created to be a JS representation of `github.com/cosmos/cosmos-sdk/types.Dec`

There seems to be a long history of tweeking the decimal type to make it work for Agoric and the Cosmos SDK Dec type. Looking at https://github.com/hyperweb-io/telescope/blob/0e4847cf8facfbbc8c35f55996925871bfcf4312/packages/telescope/src/helpers/decimals.ts we see a...

The chain ID value is not exposed here. You'd need your own clone of `signDirect` to be able to access the value. However, what's the use case for this? The...

This is pretty straight forward to implement with CosmWasm's `MsgExecuteContract`. You can send funds along with an execute message. A contract on the Hub then receives those and internally emits...

By the way, the other way around is annoying as well. I can't send ATOM from Osmosis to a CEX directly because I can never trust those guys to be...

> seems we would need the CEX to issue a `MsgExecuteContract`, if I got that right? Right. However UX wise it would even be better if CEXs integrate it that...

Thank you for bringing this up. Turns our that this depends on Zero-Installs vs. non-Zero-Installs. See https://yarnpkg.com/getting-started/qa#which-files-should-be-gitignored and https://yarnpkg.com/features/caching#zero-installs. I'm not necessarily convinced we need Zero-Installs but for now we...

Okay thanks. In that case I lean towards removing @cosmjs/cli. Too much effort has to be put in maintaining it with questionable value for developers.

We'll indeed remove @cosmjs/cli in the next release