So I just read through the over 650 posts in this thread that I've missed in the past couple days. Frick you people talk a lot. Anyway, I had a ton of quotes lined up, but it seems like one by one they each became obsoleted as the thread went on. So instead of actually quoting people and responding, I'll give a general response to each of the major topics in this thread. I'll separate my post into sections, so you can skip over parts you don't care about.
---
First off, lucas zair.
Stingers made a great post on this which seemed to get largely ignored in favour of the fanboys with little to no real argument. I think he brings up some valid points, and the move
can be worked as a way to regain momentum/space in a match without breaking it. Giving it low damage, a low angle, and fairly consistent kb would turn it into a techchase starter, which as people mentioned... lucas already has several of those. The difference is that zair would trade damage for range. He would have to work harder for damage when using zair, getting only 3-4% on hit, as opposed to something more like 10 or 11% with other options. You can get away with just putting a foe in a compromised position with higher damage moves, even if you don't get any direct followups, but with zair, he'll really need to predict right to get any real reward from landing it, even though the move would be easier to land than many of his other options due to its range. It will not suddenly make him a great character, but instead give him a way to gain some footing in a match, where he still has to predict correctly after landing one.
I'd also like to note that I'm not a big fan of the idea of adding moves to the game, but I think lucas's zair is acceptable. Largely because the move is - for all practical purposes - already programmed. The input for lucas's zair results in an established animation playing, the input is not a taunt, the effect of the input is not movement oriented, and the effect of the input does not switch you to a different character. In fact, it is the only non-movement and non-taunt animation in the game to not have a hitbox that does not also switch characters. Lucario's upB doesn't have a hitbox, but it's used for movement (not only recovering, but also for offense). Peach's downB doesn't technically have a hitbox, but it spawns an item that
does have a hitbox. Independent pokemon's downBs don't have a hitbox, but they don't even have an animation to begin with; it's not that independent pokemon don't have a hitbox on their downB, it's that they don't have a downB... period, and if we were to give them a downB that would truly be
adding a move in its most literal definition. This is the major difference, here.
Lucas's zair already has an established animation. Yes, it's a substantial edit, but it is just that - an edit. It is not
adding a move, but rather editing an already existing one. Since this is technically merely editing an existing move rather than adding a whole new one (again, the animation that plays as a result of this input is the same after the edit as it was before it), people trying to push for completely reinventing a move
by giving it a new animation will have no precedent to justify their change. So I say... why not? Give it a try in a nightly. If it gets turned down, no harm done. If it gets accepted,
also no harm done.
---
Next, there's the topic of kirby.
Cam made a great post on this, I think, which does help put things into perspective. I've largely been of the opinion that kirby is fine as he is ever since the fsmash nerf was put in. Perhaps a small nerf on bair, but other than that, I figured just let him be. This post made it a lot clearer what the nerf should actually be. A damage nerf would not wreck the move, and it would still be extremely important to kirby's playstyle. It would still kill well, it would still gimp, and it would still combo. It just wouldn't rack damage as fast while doing it. It would make the move less safe on shields, as well.
---
Buffer texture hack.
What can I say? It's awesome. It should definitely be made part of the official texture package, along with independent pokemon portraits.
---
Now, for the metaknight discussion.
MK is an extremely difficult character to do anything about. His whole moveset is basically
designed to be good. He has fast, disjointed moves, his movement is fast (except horizontal air speed), his recovery is solid, his moves combo, and he can chase people offstage extremely well to gimp them. The problem is that these are attributes that
define MK. We took away his outright kill power. We took away his ability to dominate matchups with a single move (although he can arguably do the same now with about three... it's an improvement, right?), which as a result even hurt his recovery somewhat. But he's still metaknight, which means he still has good speed, disjointed and fast moves, gimp power, and a good recovery. There isn't a lot more we can do without making MK... no longer MK. Now, what we
can do is raise the angle on some of his moves (ie. dthrow) to prevent them from setting up for edgeguards as easily. I suggested giving his dthrow a 65 degree angle, for example, which would hopefully prevent him from techchasing, comboing, or gimping from it. The upB sweetspot could also be given a less nasty angle to prevent gimps off of uair combos. There
are things we can do, but they're limited, and MK is simply naturally too good of a character to really rebuild in the way we did to marth. We've been trying to force him to focus on gimps to keep him specialized, but he's still simply too good even with that specialization to try to rebuild him from here. The dthrow and upB changes are some of the last things we can do that will nerf what makes him a problem character, although this almost encourages further three move spam. MK is a touchy subject for obvious reasons, so we'd rather see counters evolve naturally, but if it becomes necessary, we
will force counters by further nerfing him.
---
And now for a misunderstanding about quarter-circle and double-stick DI that never actually got resolved earlier in the thread. Note that this is mostly based on melee knowledge and may not be entirely accurate in brawl, due to the way it handles the cstick differently.
You can SDI with either stick. I think what you are thinking about is double stick DI. Where you hold one direction on one stick for survival DI and you quarter circle the other one or w/e for SDI. One of them overrides it but can't remember which. JCaesar help me out

.
for double stick DI, press towards the stage with the c stick ASAP and you will SDI every possible frame towards the stage, DI towards the corner with your stick so you can live as long as you can before being sent past the blast zone.
is my understanding of it.
Quarter-circle DI is performed by rotating the control stick in... a quarter circle... in order to obtain multiple SDIs in rapid succession when you're hit. You
cannot SDI in the same direction on two consecutive frames (even if it was humanly possible to set the control stick back to neutral and then smash that same direction again in the matter of one frame, it would still be read as "holding" the stick in that position), however, a very slight change in the control stick positioning will be read as a new direction, allowing you to get three to five SDIs off of QCDI. This is mostly useful for escaping multihit moves or for ledgeteching, ensuring you get the maximum SDI possible. It's rare that you will use QCDI as a means of survival unless you're trying to force a tech, since it's rare that you would want to SDI in the same direction as you want to perform regular DI in.
This is where double-stick DI comes in. DSDI is using the cstick to SDI in one direction while performing normal DI with the control stick. The control stick controls regular DI, while the cstick will control both SDI and ASDI, allowing you to obtain the best possible survival DI, where you rarely want to SDI or ASDI in the same direction you want to DI in.
Note that in melee, QCDI
does work on the cstick, which is probably where some of this confusion came from, but the two terms are completely separate things. Since brawl forces you to set the cstick back to neutral before it will read another input, QCDI
does not work on the cstick in brawl, however DSDI
does.
---
And I see people were talking about fox's JC shine while I was making my post.
In melee, fox could JC his shine and then shine again the frame he leaves the ground, which will end up grounding him. He is not actually canceling his jump with the new shine (unlike JC grabs and JC usmashes), since the jump's start frames would have already ended by that point. In brawl+, this is currently believed to be impossible due to the "dead frame" on jumps, although if someone could prove this wrong it would certainly be interesting.