## Attendees
- amtoine
- aucacoyan
- darren
- jt
- storm
- andres
- jakub
- windsoilder

## Agenda
- [x] profiling tools (related to [Dependency: use tracing and tracing subscriber instead of simplelog and log #9519](https://github.com/nushell/nushell/pull/9519))
- [x] website CI fails ([here](https://github.com/nushell/nushell.github.io/actions/runs/5403427180/jobs/9818207681))
- [x] EVAL (kuba) https://hackmd.io/TZWT_jVfQLmU3SjucTmaQA
- [x] Typechecking improvements
- [ ] foundation and GitHub sponsors
- [x] `let blah = custom-command` vs `let blah = (custom-command)`

# Discussed Topics
## profiling tools
related to the tracing PR and the `tracing` crate.
heavier dependency than `log` :thinking_face:

we need to be clear on the problem otherwise there is no point:
- having time *spans* to know how much time a particular segment took

adding a line in PR bodies to say *this PR adds 8 dependencies* or *it removes 15 dependencies* so that it's easier to reason about whether this is worth it or not.

## website CI
let's debug this after the meeting

## eval ([Improve type hovers #9515](https://github.com/nushell/nushell/pull/9515))
the *improve hover* PR ended up being more than that :smirk:
now we properly check the types between pipelines
- pipes are checked
- give up less often on type checks

overall the type checker is more strict!

### related topic: in/out annotations

def foo [
    x: int,
    y: int,
]: string -> list<string> {
  # ...
}

def foo [
   x: int
]: [
   string -> int,
   string -> string,
] {
   # ...
}

:warning: `$in` is confusing and might be good to rename it => [#9551](https://github.com/nushell/nushell/issues/9551)