Tie in Warsong Gulch
This is a new change that has been applied to live realms as well.

If the game is tied when the game ends, the winner will now be the team that most recently captured the flag, rather than the team captured the first flag. (Source)

RMP Teams in Arenas
To address the (frankly overblown) RMP "bias", we're seeing a lot more comps this season so far, which is cool. We just want to make sure those comps are doing something besides just blowing someone up. We certainly think the Arena can support more than RMP, which is why we have made so many changes for other classes. But the reason a lot of players like the RMP playstyle, and why some consider it the only legit team (though that is emphatically not our viewpoint) is because it requires a lot of coordination and layering of abilities. That's not to say RMP can't go for the blow someone up strategy on occasion, and there is nothing fundamentally wrong with that strategy existing, as long as it's not the only one. When every match feels like the same match, Arena gets a lot less interesting. (Source)

Tanking - Balancing Effective Health and Cooldowns
EH is very important. But it is not the all there is to tank balance. I suspect even the players that first came up with that way of describing survivability would admit that it gets focused on too much these days. (Source)

Race Changes ETA
A race change is only a slightly more simplistic technical process than a faction change. Given that, it was pretty important to get the conversion frame work for the faction change service developed first before offering additional services like race changes. (Source)

The end of classes changes
It will never happen because we're always going to be adding content to the game (including class content) that will require us to make adjustments to each class. Additionally, classes scale differently with gear and content, and we always have to monitor that closely and change things as needed. (Source)

PvP/PvE versions of spells
Internal consistency for belief in the game world is part of it. Just keeping complexity down is another part of it. We don't want to see every spell or talent have two tooltips and players having to keep all of that straight. It just feels like two games with one art set at that point, which isn't what we're going for. (Source)

Death Knight
Heart Strike in 3.2.2
Heart Strike is balanced around its single target damage since its a primary ability for Blood. However, since it can strike a second target for full damage (on live), Blood dps jumps up dramatically just because of the presence of a second target. The ability needs to be either balanced around its single target or dual target dps. If it's the latter, you're going to use Blood Strike on a lot of attacks instead and the core strike for the tree becomes more situational, which feels lame.

It will still keep the kit of hitting a second target, and Blood dps will still jump up when you can cleave a second target. Just not as much. (Source)

[...] We'll write a more clear tooltip, something like: Instantly strike a target, causing 50% weapon damage plus 368 and 10% additional damage for each disease. Also strikes a secondary target for half of what it strikes the original. (Source)

Shamans and Cleansing Totems
It's overpowered. Long-term we need some kind of overhaul to the dispel system so that you can free allies who get debuffed or CC'd but you aren't constantly losing important class abilities because of cheap, fast and sometimes pulsing dispels.

I'm being vague because we haven't announced any 3.2.2 or 3.3 changes yet. Those won't be the overhaul I described, but we hope to do something. (Source)

Paladin
Changes to Hammer of the Righteous and High Threat Abilities
 High threat abilities are always going to be part of the game for the simple reason that we don't want tanks to do as much damage as dps specs, but they need to generate more threat than the dps specs.

It only gets to be a problem when those abilities don't scale with gear or the tank spec is so woeful at dps that it's impossible to solo or contribute meaningfully when not tanking.

In this case, the loss of threat from switching from Holy to physical damage necessitated adding additional threat.

Having said all that, we don't have a lot of confidence that this change will actually go live, so I wouldn't freak out just yet. (Source)

Shaman
Earthen Power Nerf
We think Earthen Power and the Ghost Wolf change were trying to solve the same problem. Together they felt like overkill to us, and we thought the Ghost Wolf change was more interesting so we kept that one. Your mileage may vary and I would be surprised if many shaman agreed since they were the target of the nerf. (Source)

Flame Shock duration change in 3.2.2
We don't think the coefficient needs to change. We literally just added the duration bonus from the old glyph so that you didn't need that glyph to function. We could have a long discussion about mandatory glyphs for dps specs, but let's not have it here.

The dot coefficient rule that I think you're implying isn't necessarily something we follow anymore.

You get two more ticks on Flame Shock, which frees up GCDs to cast other spells. If the glyph was a dps increase, and it generally was held to be, then the change to Flame Shock should be too. (Source)

3.2.2 Glyph of Flame Shock
Most of the posts I've read on the last few pages seem to get the purpose of the change (to make it so the spec could function without the glyph), but there are still a few confused shaman. Reread the thread and you'll see some good explanations in there.

This was not intended to be a massive Elemental dps increase. It was intended to solve a problem that may not have been important to you personally, but was important to a lot of shaman.

One thing I will add is to fret less about scaling. In BC if you didn't scale well, you were going to be in trouble for many tiers of content. These days we adjust classes so frequently that it's much less of an issue. Worst case is you will start to slip slightly behind other classes as everyone gets gear, but then the very next tier will be a reset (and honestly we really only have one tier left).

Put another way, if you and your friend start an instance at 5000 dps and then as you get gear she goes to 7000 dps while you go to 6000 dps, then that's a scaling problem. But under out current design philosophy in WoW we will adjust you if you fall that far behind. Yes in an ideal world all the talents and glyphs would be set up in such a way that we'd never need to adjust them and all classes would scale linearly (or even exponentially) with the same slope. The reality is that we tweak things and some scaling mismatches are inevitable (if for no other reason than melee scale differently than casters). (Source)

3.2.2 Glyph of Flame Shock and T8 Set Bonus
We aren't changing the set bonus. Most shaman will outgrow it eventually. If you have the set bonus, you can look at other glyphs for the moment if that works for you.