Joar would you be able to answer my question about how rain would work in subterranean regions? Would those areas progressively flood, forcing the player to seek shelter as the water is rising, all the while they're constantly facing the risk of drowning? I'm really curious now as to how the mechanic will be implemented for these unconventional level types.
I think thats basically the idea, though any "flooding" would be somewhat superficial. There are some strict limits to how dynamic we can make the terrain using the current engine and toolset, so were just going to have to make it work as best we can. I think with some time spent, the effects will be pretty cool though.
Joar and I have talked about it, and it seems like there are a couple of ways we can do "rain kills you" underground, flooding being probably the most compelling of the bunch. Since we'll be doing a few non-standard environments, it stands to reason that each will probably have their own slightly different environment-specific way the rain kills you: perhaps flooding for underground regions, some combination of deadly wind and rain for aerial regions, crashing waves and surf for watery regions. Regardless, they will basically behave like the "rain killscreen", just with a different flavor of buildup. I think thats something that we'll want to do once the world is more stitched together, as it will probably mostly effects, shaders, dynamic lighting, screen shaking, etc.
On NeoGAF someone commented:
I really hope these guys put as much time into designing the levels and objectives as they have into the animation and game engine.
Perhaps Joar/James, you could talk a bit about how you go about designing the gameplay elements and level design? How objectives might work?
Hah, thats a pretty dang huge question! I guess one of the most important things for me when trying to conceptualize the world / level design is trying to keep the balance between the open world "terrarium" aspects and the need for, uh, actual gameplay. Like, sure there are platforming challenges and puzzle-esque rooms, but its not "stage 1 complete! now go to stage 2!" sort of thing. What we'd LIKE to do is make the game playable however, rewarding whatever way you can survive within the game mechanics. Like, you could run around the map and follow the various trails of breadcrumbs, or you could just get really good at hunting bats and finding shelters and both would be totally valid ways of playing, each with their own narrative path.
Another thing we are trying to do is keep "fetch" style gameplay to an absolute minimum. With platformers, if you want to move a character around an open map there has to be a touch of that, but we want to keep it very organic. So, ideally what we'd like to do is have the progression be mostly skill level based. I dont mean "level 21 slugcat", just that as your knowledge of the world and your skill with the character improves, more areas open up to you, you know this or that trick to getting to a higher ledge, or have learned enough about how some challenging creature blocking a path behaves to be able to navigate through it. So an experienced player starting a new game from scratch would have access to a much wider area of the world from the get-go.
BUT... honestly this is just game design philosophy rambling at this point. We have it mapped out on paper, but until its actually implemented and plays well, its all subject to change. Right now we are just finishing hacking together the terra firma of region 2! Our idea right now is to get 4 regions assembled into a nice large chunk, see whats working and what isnt, then plan it from there.
Hope that sort of answers your question!
— 
Im sure he'll talk about that palette a bit more when he's back in action (Feel better Joar!!)
, but we need the game content to catch up first! Seems like we are getting there though. Im putting together what might be the last few rooms for the second region this weekend. Then only 10 more regions to go! 
Right now my idea is to get the first few regions together (maybe the first 4) and see how it all feels connected, how much time it takes to get from place to place, etc., then take stock and use that info to tighten up the ideas for scope of the greater world. Then after I get the world all stapled together, of course Joar is going to do another pass to add some REAL ART CRED, hopefully some dynamic set pieces, etc etc. 
