I’d say column A column B.
Turning these down to 1 pointers (without just reducing them by half effectiveness - which would make them terrible) and putting a bunch more throughput points in the tree would just be strictly a buff to Holy priest healing. That’s something they’ve just been reluctant to do in general, lol.
I don’t know how you’d reduce all of these to 1 point talents and then create several more 1 point talents without increasing overall HPS by a significant margin. Regardless of if they’re imaginative enough to come up with ‘talent ideas.’ Which I’d guess they are - It’s very easy to be an idea man, it’s not easy to be an idea man while keeping playstyle and balance in check and then implementing.
And the only talent trees that do this are very much gating off playstyles, in a way. In Paladin almost the whole tree is 1 pointers, but half of it supports “caster” pally, which nobody’s really taking unless they want to opt into that. Monk’s got the whole jadefire side split up with a SooMi build that nobody plays (and worked to fistweaver’s detriment in raid).
But you look at the classes that aren’t doing that (or the playstyles are more differentiating raid from dungeon), the trees are very much intentionally gating off healing boosts from too many being taken at once.
Again, I think holy should have more pathing options, and I especially think it’s weird that we have choice capstones - they should be split and allowed to take both on each side - but I just wholly disagree that it should look anything like Paladin’s. We don’t have a weird split toolkit like they do.
I’d expect our tree to look more like shaman - which is a spec I think talents feel really good for still - paths where you can take a single point down to a capstone, but that single point is worth less than the 2 pointers to make up for the extra capstone you’re taking, etc. And ideally, just move options around so the single target and AoE talents aren’t all clustered around the spec in a weird way.