I have to wonder why Blizzard is so intent on switching the roles of FoL and HL. It's almost like they're confusing casting time with time efficiency. FoL is the faster cast, but HL has always been the more time efficient spell. I'm wondering at the benefit of switching those around - what's wrong with a fast, mana efficient heal and a slow, time efficient heal? Unless the granularity is important somehow, there seems to be no benefit in making that change - unless a fast, big heal is somehow going to be necessary for Cataclysm (though that seems directly contradictory to what they have suggested).

Mixing FoL with HL to find an appropriate balance of HpS and HpM is hardly new (actually, quite old since it's rather out of date with the huge damage spikes); what is broken about it that they are trying to fix? The only thing I can think of is that there will be strong incentives to not be stopped for long periods of time (see: granularity), in which case having your small heal be the efficient heal could be a disadvantage.

Maybe the new healing design is something like this:

1 - Flash of Light - Short cast time, High HPS, Very Low HPM (emergency cast, heals fast, for a fair chunk of health, but costs a lot of mana)
2 - Holy Light - Medium cast time, Low HPS, High HPM (default cast, doesn't heal for a whole lot, but saves mana)
3 - Greater Holy Light - Long cast time, Medium HPS, Low HPM (recovery cast, heals considerably, costs a fair chunk of mana)

The same could be said about a Priest's Flash Heal, Heal and Greater Heal. So, you'd only use 1 when you needed to heal someone low on health that could die in a moment. You'd try to avoid that situation by constantly healing with 2 to save your mana, because spamming 1 like we do today (aka Grid Whack-a-Mole) would run us out of mana. If we only kept constantly healing with 2, however, we would find our target's health slowly declining; that leaves 3 as a sort of 'recovery cast', something that you would use when you had the time to do a big cast and the mana to spare, and our target's health was not dangerously low. 3 would then help us to up our target's health even while they were taking some damage (something that 2 would not accomplish), but 3 could not be cast all the time (like we do now) or we'd also risk going out of mana.

Of course there's also the rest of the healing arsenal to be interspersed, but I believe if these come to be the basics we're in for a much richer healing experience.

The other factor is:
Whirlwind is a trained skill.
Divine Storm is our 51 point Ret Talent (though this could, of course, change).

If DS is dramatically reduced then it could become the weakest single-target ability in our arsenal. If Holy Shock goes baseline and still does damage, we could wind up prioritizing a nominally Holy spell as DPS over our end talent. If that sounds lopsided, it very much is. Blizzard will balance DS in such a way that it is still desired for usage other than on trash.

Blizzard has stated an intent for haste to increase resource generation - Rogues will generate Energy faster, Hunters Focus, DK Runes. Warriors (Rage) and Paladins (Mana or GCD?) have not yet been mentioned, I'm sure it will require some careful work. If we remain GCD bound only reduction in melee GCD (unlikely) or possibly cooldowns (also unlikely) would make haste function. If we move to mana bound they need to rework the system so we have free GCD and an optional filler attack to use (spammable, low-damage, low-mana). Everything good unavailable, have enough mana - hit it. Mana an issue or better ability available - don't hit it. Think Warlock and Shadowbolt.

Otherwise we're both GCD, cooldown, and manabound. That's a straight jacket.

I think everyone is looking at the "small, fast, ineficient heal" through our current version of it. Personally, I think the "small, fast, inefficient" heal will probably heal for only 10-15% less than the "middle of the road heal" but with a ~30% higher mana cost to pay for the short cast time. The whole system ends up looking something like this (numbers are totally made up):

Holy Shock - Instant, 900 mana, heals for 7K (similar cooldown to now)
Flash of Light - 1.5 sec cast, 900 mana, heals for 7K (decent heal but not sustainable)
Holy Light - 2.5 sec cast, 600 mana, heals for 9K (spam it all day without going OOM)
Blazing Light - 3+ second cast, 900 mana, heals for 13K (high throughput and efficiency but too slow to react)
Healing Hands - X cast time, 15 sec cooldown, heals for a lot more than a totem (8k over 6 seconds?)

It appears that they're intending on keeping the damage component applied to Holy Shock. Found this (SOURCE) in the Tanking forums, but shows there may be some implications to having Holy Shock still do damage to Ret's "rotation".

Definitely could use some more single target abilities. Any fight/encounter where aoe damage is undesirable severely limits our dps.

They can tie Divine Storm via a shared CD to another ability so you are forced with the choice of AoE or single target damage depending on the situation and I'm sure there are other ways of working around it too if they really didn't want you to press it on single targets.

I agree that this is what Blizzard seems to be driving towards; personally I was wondering why. If for the most part the time interval you had to heal someone is long, the cast time is mostly irrelevant - the most important aspects of the heals are HpS and HpM. In other words, what is the functional difference between the above and:

1 - Holy Light - Medium cast time, High HPS, Very Low HPM
2 - Flash of Light - Short cast time, Low HPS, High HPM
3 - Holier Light - Long cast time, Medium HPS, Low HPM

As it happens it looks like Blizzard answered the question - apparently they want the core heals to be essentially identical between healer classes so they don't have to worry about odd interaction with any future raid mechanics (e.g. you have just enough time to get off a short cast between Massive Boss Stun, in which case pallies might be at a disadvantage). This actually sounds a bit odd to me, given that the previews seem to imply that Druid healing would not necessarily follow this model, and also HoTs do not seem to be part of the 'core' healing mechanics.

As an aside, I would also enumerate the heals differently, precisely because the cast time seems mostly irrelevant except in context of specific encounter mechanics:

1 - Flash of Light - High HPS, Low HPM
2 - Bright Light - Medium HPS, Medium HPM
3 - Holy Light - Low HPS, High HPM

Then depending on the incoming DPS, you would be in a given 'mode', where you mix up adjacent spells. In a high incoming DPS situation, you would be using 1&2 (e.g. the 'alert level' might be 1.75, which would indicate you are using 1 to 2 in a roughly 1:3 ratio) while in a low incoming DPS situation you would be using 2&3. Note that in general there should be no reason to use anything besides the two adjacent heals (if using some mixture of 1&3 was optimal in a given situation - again barring specific encounter mechanics - there would be no reason to have 2. This does imply that the 2-heal needs to be better in HpS or HpM versus any combination of 1-heal and 3-heal).

This is and has been one of Rets biggest problems in pvp. Not only do we not have a reliable interrupt, but we are the only melee class without one of the main tools that every other melee class has at least 2 of. Interrupts, MS, Slow, and a Distance-closer......without one of the above, Ret will always rely on huge burst during a HoJ, and hope they don't have a trinket.

I really expected Blizzard to give us one of the tools above, more specifically an interrupt, but hopefully it is in the unreleased plans. All in all, I think they need to get away from giving all three trees the same abilities. It makes balancing things much more difficult than they need to be.

Not sure where you hear this, much less "often." It has never been stated by Blue, anywhere.

However, it is a possibility. Tier set bonus have been used as new/reworked talents for classes in the past. Many (most?) DPS classes have some "rotation" interrupting proc. Blizzard appears to like this design. It could use tweaking if they go this direction - the other classes distinctly lose if they do not use their proc immediately - it's #1 priority when it comes up. Ret still use 2 attacks over DS (with J > CS > DS, I'm aware some do J > DS > CS), even when it magically resets from 2T10.

A better approximation of other classes would be a stacking buff that reduces Exorcism cast time (a la Maelstrom Weapon) possibly also resetting cooldown, rather than AoW making it instant on crits. Then Exorcism hits like a truck (a la Judgement). Cast time means no autoattacks, so it's a DPS loss to cast. Once it hits instant cast it's our best ability, you want to press so you can start resetting again ASAP - delay is bad.

This is merely a comparison of mechanics, not a suggestion of what they should/might do. Exorcism used because it has a cast time and could be used directly in comparison to Maelstrom Weapon's functionality, not because it's the best place for Blizzard to implement a mechanic.

Proc reset mechanics force an interruption in a rotation or for the player to suddenly change what they were about to press next in a priority system like ours. Blizzard has slowly moved many things towards this direction to encourage easy average DPS and reward great DPS from those who can do more than just mash buttons.

This is from the Twitter chat.

I am glad that they want to make Healing Hands more than just a portable Healing Stream totem.

Making relics available to more classes (like there could be an Attack power relic for DK, Pally, and Feral) is decent idea, it will be interesting to see how they execute it (like continue to put relics on the badge vendor or make them boss drops again).

The last is further confirmation that Holy Shock will still do damage (at least in the current Cataclysm, everything is subject to change).

I wholeheartedly agree, and yours was a more elegant way to put it. To further extrapolate from your example, if we were to mix 1&2 according to our target's health (gravitating towards more 1 use the lower their total health), mixing 2&3 would be the corresponding move when the goal was resource management (with more use of heal 3 the lower our mana pool). That way we have to consider granularity not only on healing size and speed but mana cost (so that both responsiveness and planning - through minimizing overhealing - would be rewarded). That all requires, of course, careful tuning to try and keep the numbers consistent (percentage wise) during later tiers, or risk phasing out one of the heals.