dpc

At some point `crossbeam-channel` will maybe become `std::sync::mpsc`, and we can switch back. AFAIK `crossbeam-channel` is considerably better in terms of performance, and if anyone is using `slog-async`, I think...

Is still a problem? In my use case we use these a lot so this failing would be a complete blocker.

> I like this proposal because compromising arvo would only reduce the security to the level of the existing best practices. Exactly. As long as it's being used to supplement...

Urbit would not just "host the encrypted data". It would actually generate on-demand site-specific password (or rather deterministic entropy), derived from the master-password it stores. But it would be **second**...

> Maybe I can add a `#[builder(setter(explicit_default))]` option, that will make the field mandatory even if it has a default (I'll force it to have a default), but make the...

Not sure what exactly happened there: might have been something to do with optimalization causing this to point to slightly different function, but I don't think it's very relevant. I've...

@PvdBerg1998 `slog-envlogger` is a fork of `env_logger` so it's filtering performance is going to be similiar, but other than this `slog` is supposed to be very fast, especially with `slog-async`...

IMO. `env_logger` (and generally the whole `log` ecosystem) should become more modular and `env_logger` itself should be split into a logical part (decide if something should get logged or not,...

Definitely possible. I thought about `_OS_` if `nh` was to handle more stuff. I don't know if that's the plan. One might have "NixOS configuration flake" and then some other...

It needs to be added as a new "target", possibly with matching "toolchain".