Eliza Weisman

> In the current implementation, we _always_ poll _all_ unready inner services, which leads to poor behavior as the balancer scales. Something like #283 is necessary to relax the balancer's...

@olix0r nothing in particular, I was just wondering about your thought process. I am interested in doing some benchmarking though, although I'm not sure if it would be practical...

@olix0r In the approach I described, you only poll _two_ inner services on any given `poll_ready` call on the balancer --- I definitely agree that making the balancer actually constant-time...

Ah, yeah, that makes sense now that you mention it. Thanks.

@hdevalence any interest in wrapping this up, or should we just close the PR?

> I'd be happy to wrap this up, but I don't really know what the "good" design would be -- the current PR is just what worked for my needs....

I'm not totally sure what @olix0r had in mind here, FWIW, so he can probably better answer these questions. > * Say more about switching back to enums. Do most...

I went ahead and assigned everyone who might have opinions on this, hope that's not too annoying. :)

Quick ping — how do other maintainers (especially @seanmonstar, since it's potentially relevant to `hyper` 1.0) feel about this proposal?

@davidpdrsn > One thing though: It all depends on the timeline that we agree on but I don't think we should necessarily wait for new features to stabilise in Rust....