Cofree comonad

[edmundnoble]

Should I add it? Would anyone be interested in both lazy and strict versions of Cofree? I find the strict version very useful in creating trees.

[adelbertc]

Paging @sellout and @tpolecat our Cofree extraordinnaires

[sellout]

Well, I think if you’re going to have Free, you should have Cofree! I’m still mostly a Cats spectator at this point, though.

[adelbertc]

I think it makes sense to have corresponding Co- versions of the Free stuff we have. 👍 from me.

[edmundnoble]

I'm on it. My current plan is to trampoline the tail with Eval, if anyone is interested in a strict Cofree (or, perhaps, a tagged "newtype" with strict instances on a Cofree) please do speak up.

[adelbertc]

Awesome, thanks!

[tpolecat]

Yep we definitely need this. Scalaz only has the lazy one and nobody ever complained as far as I know so that's probably sufficient.

[adelbertc]

Do you think it would also make sense to add Fix-y things, or does that belong in a (hypothetical) matryoshka-cats ?

[sellout]

@adelbertc I vote the latter – I really want to make Matryoshka FP-lib-agnostic, but don’t really have the bandwidth for it at the moment.

Also, because I think there’s another lib of higher-order stuff (which I am tentatively calling Griffins) that Matryoshka will want, that we probably(?) don’t want to lump into Cats.

[tpolecat]

Yeah I think Free and Cofree are probably good enough for Cats proper.

[KadekM]

Yes please, Cofree is useful and I'd feel much better using standardised cats one than reinventing it :)

[kailuowang]

closed by #1446