My memory's very vague - didn't we try something similar and back it off before? was it different operator? /cc @benlesh

Honestly, we talked about doing this at some point (I'm not sure if it was in an issue or a discussion). I mean, it's obvious.

However, it was decided at the time that mapTo was so popular that we should allow it to be as optimized as possible. That said, I'm not entirely sure how much better it is the way it was over the way this PR is proposed.

In fact, I'm inclined to deprecate all xMapTo operators, just because they're silly when they're just hiding a closure. mergeMapTo(x) and mergeMap(() => x) no matter how we'd reasonably write them are pretty much the same thing. Just one does the closure for you.

There was probably an advantage when our operator internals were all class-based, but now it's unlikely. Especially with closures being more optimized in modern runtimes.