Thomas Eizinger

Just confirming what you said above: Yes, not awaiting the response of a message should be used very cautiously because you are breaking the back-pressure chain. It is fine to...

Perhaps we should at least make the struct opaque such that we can work on this feature without making a breaking change?

@Restioson Do you have any thoughts on the [above](https://github.com/Restioson/xtra/issues/121#issuecomment-1183538016)? If we agree on this, then we can simplify the implementation by removing the separate queue for default priority messages.

> With the preliminary support for async traits in latest stable, is this a necessary addition? *nightly. I think they are in latest nightly. I haven't experimented with it yet,...

> The explicit handler impls are much more... explicit, and easily searchable. I do feel this would be a bit black magic. You never invoke handlers directly anyway so really,...

> > Spaad was always a little bit of an experiment with it and I was not totally satisfied with how it turned out. > > @Restioson would you elaborate...

> In my view a major pain with taking advantage of the actor model is the boilerplate that comes with the message passing. Ideally it would be hidden as an...

> As the number of methods an Actor has increases, a macro really does cut down a huge amount of boilerplate. Even just writing out the code block in this...

> > In fact, that is what this issue is about :) > > My most recent thoughts are mostly towards the boilerplate of a "Controller" (sorry for drifting off-topic...

I don't think I understand enough about spans to help here but if somehow possible, I'd be great to have a failing test case that we can work towards fixing.