Hello, another complete newbie/constant lurker since Miasmaw/occasional voter chiming in :) I've never posted here before but I've really enjoyed following this discussion and thought I'd add my opinion to the chorus, even though I don't have much to add. Sorry if my post is bad or stupid lol, please don't be too mean if it is :P
As for Corrosion, I've been fairly neutral and even tentatively for it since it gives a clear, focused direction and has a unique, interesting, and un-optimized niche (very pro-concept like Korski said). It would be really cool to see how Corrosion on a good Pokemon would effect the meta. But ultimately the recovery argument makes me disprefer Corrosion. In randbats (the only format I have time to really get into lmao) Salazzle's SubToxic set is practically the most annoying and unfun thing to exist, I know we won't be exactly like Salazzle and a real meta is not AT ALL like randbats, but good golly that thing is obnoxious enough without recovery. Imagine us doing that and presumably we'd have good stats. 30i can't use boots and that plus its most spammable STAB having recoil means it will really want recovery, which would push Corrosion over the edge unless our stats are crap imo, or reliable wish/defog support to be good. Sqookblooks mentioned some other utility wallbreakers, like Melmetal, Defog Scarf Kartana, and Heatran, that are good despite no recovery, but Melm and Kart resist SR, Heatran's neutral to it, and none of those use recoil moves. We take 25% from rocks and, if we are physical (seems likely as BB is stronger and more spammable than Hurricane), take recoil when we break (yes, Gunk Shot exists but TL Flying is a stronger offensive type). We don't even get Lefties; a reliable way to avoid SR damage and heal from chip are essential for 30i in my opinion.
Though I disfavor Corrosion now, I strongly agree that it should be on the slate. A lot of the more experienced CAPpers seem more opposed to it, but we spent so much discussion on it, and there's not a clear consensus, that I think we ought to send it to the polls and see what the people decide.
I think Water Bubble and Stamina are the funnest and most interesting options on the slate. Stamina completely fits the bill as an unoptimized ability, for the reason that Mudsdale doesn't get recovery and dies very easily to the strong special attacks targeting its weaknesses. The optimized Stamina user desperately needs recovery-- it wants to come in and sit in, getting fatter every time you try to smack it, and without recovery it just gets whittled down and eventually forced out. There are also several good routes we could take to fulfill Stamina. A special wall that, with a smart switch-in, instantly becomes a physical wall too. A physical wall that becomes unassailable then slaps back with Body Press (good way to deal with the Steels that naturally check us). And this is just two examples I thought of in 2 seconds-- practically any defensive mon would enjoy Stamina (if they don't have a better option; I guess this really only applies to defensive mons that get by *mainly* with naturally strong stats and typing), so this ability provides a couple clear focus ideas while also leaving a lot open.
Water Bubble just seems really cool and fun. A third STAB is a unique and interesting option, gives us something to prevent Steels from completely walling us, and the defensive value is amazing. Also Araquinid is very unoptimized and Water Bubble is such a defining ability that does so much, that I feel we could learn a lot from it. And the power ceiling is completely manageable, we can just not give out water moves of too strong base power. (But honestly? I think it'd be pretty neat to at least think about stronger moves too. Having a defensive mon that could just drop an (attack uninvested, we can give it a horrible attack stat and choose defensive stats that demand investment instead to get important benchmarks) 150 BP 100% accurate drawback free water move would open up interesting interactions. For example, how would a wall with the ability to have a strong attack of an unexpected typing change its natural checks as determined by typing and stats? This is a pretty basic question but I'm sure this setup has more interesting things to ask too.) It is not as direction-ful as Stamina is, so I think I slightly prefer Stamina (which gives more specific direction for what defensive things we ought to be doing, water bubble is way more general).
Thick Fat doesn't seem bad, just... boring. Others have explained how it doesn't actually help a whole against Ice types, and as for targeting Fire types I say we just use Water Bubble in that case. Gives similar but in practice better defensive benefits (Poison AND Burn immune for the win) and is so much more interesting and fun, while still being easy to make balanced.
Fluffy and No Guard definitely have potential for interesting interactions, with Fluffy making us weak to Fire. No Guard does seem intriguing, but I hope we don't go down the route of "I just use strong inaccurate attacks" because that's boring. Also, everyone seems to be assuming 30b will be defensive (are we allowed to assume this?), isn't No Guard bad in some ways for a defensive Poison-Flying mon? For example, now Stone Edge, a pretty common inaccurate move we're weak to, or Heatran Magma Storm that we absolutely hate, always hits us. Looking at the trade-offs of No Guard could definitely be interesting, and I think it'd be pretty neat to have a defensive mon that spreads status and damage simultaneously with like Inferno or Zap Cannon, so No Guard has potential. Also No Guard opens up status-spreading shenanigans with mainly sleep moves (but I am deeply wary of 100% accurate sleep and we would need to devote a lot to making that not overwhelming).
I'm also throwing my support behind Gale Wings. Initially I thought it was a bad option since it will be difficult to keep us at full health. But like quziel pointed out, if we do manage to stay at full health, we have potential to get massive value. It still seems very situational, but especially since HDB exists I'm more intrigued by this option. I think this one should be added to the slate.