Ross A. Baker

This is basically [Bayou](https://github.com/armanbilge/bayou) with some deduplication because it lives in the same repo.

Perhaps this achieves subtleties I don't appreciate. I'm not precisely sure what this means: > pattern of creating untraced dependencies whose operations need to be traced a lot more usable....

Okay, I think I now see what you thought you saw. I think the entry method could return `Resource[F, Unit]`, but also would need a kernel, and have continueOrRoot-like semantics....

The entry point is supportable on the stateless instances, but only works with a signature that passes the continuation, and would not work particularly well for `Resource`. It's necessary only...

> in my mind Trace + Kleisli is more "stateful" I've been using the opposite terminology. `Trace[IO]` captures a root span as state when it is acquired. The statefulness is...

/cc @alexandru for sanity check of implementation My results: ``` [info] Benchmark Mode Cnt Score Error Units [info] InMemoryBenchmark.sumInts0II thrpt 20 11818.266 ± 209.216 ops/s [info] InMemoryBenchmark.sumInts1IM thrpt 20 11321.185...

Neither am I. I got the idea from https://github.com/monixio/monix/issues/185. I can't put my finger on how it might leak, but mutability without conplete understanding makes me nervous. Redefining fs2's runLog...

I imagine the collection version is also using a `VectorBuilder`. I tried and failed to prove this by reasoning about `CanBuildFrom`. I think we're measuring that Monix beats scala.Stream in...

I tried adding `spanResource` to `Trace`, and the `KleisliTrace` has sprained my brain. 🤕

You can put yourself in a dark place with this if you use allocated and don't release in reverse order. It's totally fine to create a child span from a...