Márton Polgár

I think most of the hassle comes from the misdesign and/or miscommunication of what a Pair is and what it's meant for. [Confer the smartmatching behavior as well.](https://github.com/Raku/doc/issues/4147) Pair is,...

I don't know exactly but I would be concerned having multi candidates like [this](https://github.com/rakudo/rakudo/blob/6d8be9959ea8a4ab2c17225560fe3b4457a64fc5/src/core.c/Cool.rakumod#L320) and [this](https://github.com/rakudo/rakudo/blob/6d8be9959ea8a4ab2c17225560fe3b4457a64fc5/src/core.c/Cool.rakumod#L332) and hope they would resolve correctly.

For what it's worth, I also both prefer and come to understand the current behavior as the correct ("more correct") one, much related to the points that raiph has raised....

> Also I noticed that `start react` works fine (`react` doesn't care about sink context I guess) which seems potentially a little confusing; maybe we want to call out these...

What I'm confused about is: - what the expected return value is, eg. what would be the values of `$p1` and `$p2` in the original snippet - what is the...

> It seems fair to me that if `react` can "introduce a block of code", perhaps `start` could too. Well it can already, that's what the `$p2` snippet was about....

It seems like I managed to confuse myself somehow. At first I also wanted to say "hey, this is the fix" but then somehow my altered code snippet didn't work......

By saying that `start` should always create sink context, aren't you saying that the very example of odd behavior is ill-willed DIHWIDT code, though?

Seems like things are more complicated... ```raku my $p = start loop { if (^5).pick == 0 { last } else { "hello" } }; say $p.result; ``` The reason...

> Is that 100% equivalent to omitting the `start`? i.e. from the examples above, this doesn't actually run async. If "that" refers to the code snippet that has a value-generating...