Just to parrot back what (it sounds like) you're saying: using voxel-based assets in a game with community-contributed assets that extends itself using procedural generation is that the difficult of integration tends to be lower. This is because voxel shapes allow "artists" (which don't necessarily have to be highly-skilled) to build game-appropriate assets fairly easily, and for the generative process to more easily combine those elements into a cohesive world model than might be possible with polygon-based assets. Does that sound about right?
So basically + Cohesive world + Ease for artists + Doesn't need to look so real
I think this is the basic principle as to why Nintendo was able to get away with underpowered consoles for as long as they did. At least in comparison against other beefier consoles out during each releases duration.
So long as the aesthetics and quality are consistent, people are willing to accept a lot.
There are comments from some developers over the matter from a while back. I think one of them was Todd Howard, back from when 'Todd rays' still weren't a thing.
Erosion - https://veloren.net/devblog-43/ Erosion + rivers - https://veloren.net/devblog-36/ Towns - https://veloren.net/devblog-31/
There are lots of other blog posts that discuss these topics, I should really aggregate some of these sections in the future. Those links go into the most detail I think, future stuff is more about tweaks than the core systems (we've been putting out a blog post each week for 160+ weeks now!)
Currently Veloren has water and lava, but neither will flow (e.g. if water blocks are manually placed on top of a hill, they won't flow downhill, fortunately this is rare due to worldgen). A simple way to fix this would be to have an ECS system that uses CA-style rules (e.g. a water block with an adjacent-and-below air block swaps places with that air block), though naively doing this would be expensive (iterating over all loaded terrain chunks every tick). A possible optimization would be to detect such blocks, remove them from the main terrain grid, put them in a separate entity-with-terrain-collider (the same kind we use for airships), and only iterate the (sparsely populated) AABB of fluids-that-have-recently-moved. This is similar to what Noita does with keeping track of recently-updated terrain rectangles. Once this is in place, it can be straightforwardly extended with more fluid interactions (e.g. Minecraft-style water + lava = obsidian, more Noita-inspired gases rising).
I think one benefit this gives us is on the artist contribution side. Being an art contributor on the project has a lot less overhead of being a formal "artist", since you can quite easily make fauna or creatures or just concept art. We've seen lots of this in our blog posts!
But also, in terms of terrain itself, I think it allows us to really crank up how wild and wacky our procedural generation is, and still have a great looking world (unlike something higher-fidelity like No Man's Sky). We do a lot of cool things under the hood; we have a full erosion system that creates the mountains out of the sea, a site system that places villages in places that make sense, and the ability to mash together shapes that then get "rasterized" into voxel houses. I don't know how much of this would be possible and still look alright if we weren't taking a voxel approach.