Matthew Kay

Does `comparison = emmeans_comparison("revpairwise")` do what you're looking for?

Is that sufficient for what you need? If so I'll close. Though I suppose it couldn't hurt to have a shortcut for it either, happy to take a PR if...

I'd be happy to have it, though I don't have much time atm so the fastest route would be if someone else made a PR ;)

Hmm, in principle it seems like it should be possible to support other backing datatypes so long as they implement the necessary array operations. If we're going down that road...

@paul-buerkner an excellent point. Perhaps we should make maintainability a criterion of any large overhaul? I think we could do a shift to S7 where the goal is to make...

Doesn't seem unreasonable. We already change the output notation to show (scaled) entropy for factors and dissent for ordered factors; something like this: ```r > posterior::rvar(letters) rvar_factor[1] mode : [1]...

> Thanks for picking this up and turning it into something production-ready! Thanks for getting it started way back when! It helped a lot.

Yeah good point. This will require adjusting `density_bounded()` and `density_unbounded()` to use limits to set the evaluation points. Note to self: `density_bounded()`'s reflection method will have to be refactored a...

Heh, you know, I thought I'd written this down somewhere, but looking through the docs it's not easy to unpack --- so thanks for raising this! A version of the...

@teunbrand I wanted to circle back around here at some point and see if there's still potential for this. I think a principled inclusion of unit support would be very...