If you're using innervate on yourself as standard (i.e. every time, instead of the times you mess up), you need a slap and a wake up call especially whilst you're gemming 23 sp in sockets that could net you 17 sp and 10 spr or 17 sp and 10 int. As I also stated above (which again, you seem to be good at not reading, making me question why I said them in the first place, let alone why I feel the need to repeat it to you), if you're not slow-stacking LB on tanks as part of your necessary HoTs then you also need a slap and a wake up call, I'm sorry but you can't convince me that when you've got 1 tank and nothing else to heal that you're spending 100% of your time keeping LB(slowstack),RJ and RG on one guy. If you're keeping 8 melee HoT'd because of healing in phase2 Northrend Beasts, you're wasting at least 6 lots of potential Nourishes every 18 seconds (and the damage could not come at all during those 18 seconds) to deal with a bit of damage that comes once in a while that is much more effectively (and quickly) covered by a shaman using Chain Heal. Yes, they switch off the tank to do that (or just chain through tank), but they don't take 8 GCDs to do it and they can give the healing when it's required, as can a holy priest, rather than trying to pre-empt it.
If they're any kind of half decent healer, they know when you're healing the target and can see the difference it makes. You can leave mana management to the tank healers, if you feel so inclined, but personally I'd rather help than sit around twiddling my thumbs or doing something as unproductive as spamming Rejuv on raid members not taking consistent damage. WG is good for 6 people, in a given area, it has nothing and I mean nothing to do with what group they are in. It is a convenient way to get people to max health that have other incoming healing whilst also giving a Revitalise effect that is very useful and I agree it is a very bad pre-emptive heal, but since the damage on the melee in phase2 Northrend Beasts is random and not consistent, ergo not pre-empt-able, it being a bad pre-emptive heal has nothing to do with it being a good heal for the job. So when you're healing to the best of your ability (and not just mindlessly spamming RJ on everyone you can think of) you're going to have a lot more time for Nourish. Hence why I told you that you should be using it more.

It may happen that your raid only has resto druids in it, or perhaps a lot of people that could put Innervate to good use (arcane mages, holy priests, etc). It is not difficult to make small gearing changes to vastly improve your efficiency. If I put on the Naxx relic and Spark of Hope, I can easily give out my Innervate without overly adverse effects on my throughput in many fights.

As far as my generally nonplussed reaction to Rapid Rejuvention, here's a somewhat in-depth outline of how I see its interaction with the main strengths of our healing:

Think about the basic reasons we like HoT's in the first place, as compared to doing all healing instantly.
1) Automatic healing of small damage. If a raid member takes damage from an RST ability, and they have a Rejuv on them, they get topped off while all the other healers are doing other things, without ever having to divert attention to that particular event.
2) Faster healing of medium damage. If a raid member takes significant damage and needs direct heals quickly to avoid a risk of death, the HoT a) does some healing before the direct heal arrives, and b) allows the direct healer to leave them short of full HP and go heal someone else more quickly.
3) Implicit AoE effect. When many raid members take damage at once (Ground Tremor/Black Hole Explosion), you want AoE heals. People love their AoE heals that cover 4 (Chain Heal) or 6 (Circle of Healing) targets, but they don't hold a candle to Rejuv. Rejuv heals up to 17 people simultaneously.
4) Easy healing of damage over time. A Rejuv significantly cancels out any DoT on the target, allowing them to take few or no direct heals and be less susceptible to a spike death for the duration.
5) Coverage of tank spikes. This is was the main driving force behind Lifebloom's dominance in TBC; it's much less of a factor currently, due to comparably weaker HoT's and Rejuv's slow tick rate compared to Lifebloom.
6) Swiftmend enablement. Don't forget about this--Rejuv provides a secondary effect to people who take sudden them by allowing us to use our instant heal on them.

The tradeoff being made by Rapid Rejuv is that the HoT gets to do higher HPS while ticking, at the expense of being able to affect fewer targets at once. But looking over the purposes of HoT's outlined above, I think most of them are better served by having more simultaneous targets than they are by having the HoT do higher HPS.

1) Here, having the Rejuv on the person at all is more important than having it tick quickly. While there's a tiny chance that a killing spike will come before the HoT tops them off, the primary value of the HoT here is obviating the need for a direct heal. It's better to increase the probability of HoT's being on targets in the first place.
2) This is more balanced, as we're trying to heal someone up quickly, but it's still not clear that focused healing is better than spreading. The situation is very similar to (3).
3) This is really the heart of the issue. Faster healing up of targets, or more targets? Here's why I like more targets: people will die if they take another hit before receiving "enough" healing (i.e. to survive whatever it is). Direct healers are going to be moving around the raid topping people off one by one. The HoT's are there to protect all those who aren't the first ones topped off. Having a HoT tick slowly on someone carries a small risk: maybe a hit will arrive right before the tick that would have saved them. Having someone entirely unhotted carries a much bigger risk: they're in danger all the way until the direct heal arrives. You're going to allow fewer deaths to random events (as an average matter, individual situations may vary) when you spread your healing across all low-HP targets.
4) Similar to last one. Being unhotted is bad--the target drifts downwards between direct heals and is in danger of dying to a spike (think Twins--surge/orb). Being imperfectly hotted isn't a big deal--the target loses net HP, but stably and consistently and probably without significant danger of death.
5) Faster HoT's are better here, obviously, as there's only one target. When you're try to focus tank healing, you'll use the Glyph. The tank HoT's still will never play as large a role as they did in the past, however.
6) Faster HoT's are a strict loss here. It just reduces the chance you'll be able to Swiftmend someone when you want to, for no benefit.

Wound up being long, but there you go. In the current content, I'd use Rapid Rejuv on:
1) Beasts, probably Glyph. The most dangerous part of the fight is Gormok, so I may as well take the faster tank HoT. It's hard to judge because nobody else is seriously threatened by anything.
2) Jaraxxus. We're actually not very strong here; I couldn't say either way. Can't think of an important advantage to either.
3) FC. No glyph. A faster Rejuv isn't likely to save a focus target, where the action happens within a second or two. It's here to top off assorted random damage and enable Swiftmend.
4) Twins. No glyph. This is the fight where Rejuv is very strong, and the Glyph just cuts against that. We want to dull the effect of the damage aura on as many people as possible.
5) Anub. No glyph. 18s duration, 3s tick, perfect against Penetrating Cold. Not going to mess with that. And for residual heals against Leeching Swarm, Rejuv is already overkill and the last thing you want to do is speed it up.

1) Rejuv is not very good for this sort of healing anyway. This is better handled by the AoE heals like WG or CoH, alternatively by JoL.
2) If this is a RST ability, generally the glyph will be useful here. Grip on Kologarn, pot on Ignis, and PC-style debuffs (although not PC itself). Even if the duration is longer than 12s, you can just cast it again. That's what I like about the glyph - it's a potent HPS increase per target. That's a strong effect, no matter how you toss it.
3) That is where the glyph is most questionable, yes. As I mentioned our raids usually have 2 trees. Glyphed each druid can cover 10 targets, so it's not bad and does distribute the healing more evenly - with 2 druids blanketing you have some overlap.
4) That really depends on the fight. I'd argue that a glyphed rejuv is better here just because the HPS of rejuv on a single target is increased by 50%.
5) This is not trivial at all. A rejuv tick hits every 2 seconds or faster. Especially if y'ou combine it with the normal rejuv glyph, a 5K tick will save your tank here and there. If only LB was affected too...
6) Not much to say here Again with 2 trees less of a concern.

My thoughts on the new Rejuv glyph:

Obviously you are trading off 'spread' for 'focus.' The question is whether Regrowth is a good enough 'spread' HOT for Rejuv to become our 'focus' HOT.

It seems to me that if Regrowth casts at 1.1-1.2 seconds with NG + new GotEM + Icecrown haste gear, then it essentially acts like a more powerful Rejuv + 4 pc t8. And as we remember, Rejuv + 4 pc t8 was very powerful, since you could control where you land it and make sure the initial instant tick covered a health deficit. Furthermore, even if you overheal the direct heal portion of Regrowth, it is likely to leave Living Seed behind, which acts as a free large tick for when they DO take damage. There are three problems with Regrowth, the first is that it's relatively expensive, the second is that it has weaker HPS ('focus') than even unglyphed Rejuv (to compensate for Regrowth having better 'spread,') and finally it's not instant.

Regrowth expense I don't see as a big issue with Icecrown gear. In high raid damage situations like Twins where most of Regrowth will not go to waste I think it is efficient enough for general use. In situations where there isn't uniform high raid damage you would want to use direct heals or the new 'focused' Rejuv anyways.

The weaker HPS ('focus') issue isn't really an issue since Regrowth gains 'spread' over even an unglyphed Rejuv in exchange, so overall HPS is almost the same.

Having to stand and turret is a limitation, but druids have options of instant hots as a temporary stopgap while they move.

The other thing about Regrowth is that if it starts seeing general use for raid healing, you can boost its throughput by 10-15% with the glyph (you won't see 20% gains due to the way the glyph works in practice). Rejuv does not have a pure throughput gain glyph.

Conclusions:

I think the new Rejuv glyph is something I might end up carrying full time, and using Regrowth as a 'spread' hot for high raid damage fights instead, if I can sustain it.

Regrowth will be a nicer spell in 3.3 for sure, and I can see using it as a main supplement to Rejuv for raid healing. I think you're overplaying its ability to really perform the same role as Rejuv though:
1) The hot is far weaker. Half as strong. It's misleading to think of Regrowth as simply providing more spread due to a weaker hot and longer duration (as I've said above, I usually like having better spread), because the total is simply so much lower. You're increasing spread, but very, very inefficiently. Switching from a Glyphed to an unglyphed Rejuv increases spread with perfect efficiency--0 total throughput lost in the process.

Rejuv can cover 17 people at haste cap. Regrowth, at the same amount of haste and with NG, will cover about 22 people (going by mean cast time in my spreadsheet). Small gain for gutting the size of the hot.

2) The direct healing portion is of dubious. Sometimes there's someone at low HP, and now we can both heal them and leave a hot, which is nice--it's the reason the spell will see raid use in 3.3. But that's a small percentage of the time overall. This is not analogous to the 4T8. The 4T8 was a small heal at the beginning of a strong HoT. Useful for something if the target had taken even a little damage, but wasting it (which was common, since we often HoT people at full HP) was no big deal. Regrowth is a large direct heal that leaves a weak HoT--casting it for the HoT only is a heavy waste. Moreover, the 4T8 was instant, which was a big deal.

On the topic of innervate, I personally usually never have to use it on myself. The only fight where I can really justify using it on myself is heroic twin valks, because we use the door strat where you barely ever get enough orbs hitting you for the buff.

I flat out avoid using nourish, the way it is right now for 25 mans. Spamming it is just so ineffective because of its mana cost. Usually our direct healers can handle the their roles well, and I barely have to do more than keep a few hots on the main tank because our pally and disc priest take care of them well. This leave me to focus almost entirely on the raid, in which my job is to blanket and predict incoming damage.

Since we're using NRB as an example, I'll go with that.
I mostly just keep hots on the mt (rejuv, full lb stack, regrowth), keep the melee rejuved, and help handle fires and a rejuv on someone with a snowbold. I wg the melee right as stomp hits, and the rejuvs help keep them up as well.
P2 I keep the melee rejuved because they are the most likely to take a sudden spike of damage, especially if a group of them get burning bile or toxin. I help with hots on the tanks, and wg when a sudden aoe from bile or toxin hits (if people aren't spread out enough...).
P3 I usually keep at least the melee rejuved, since they always get knocked back, and I like keeping rejuvs on the healers for the frost aoe. Wg on that aoe, then rejuv as you see a crash coming with everyone topped off.
In heroic NRB I always use my innervate on our pally mt healer, who can nearly keep the mt up by himself in p1 with HL spam.
Maybe it's just my healing style for 25, but I find that other class' direct heals atm are much more efficient, and watching the raid and predicting damage is my forte. Swiftmend for emergencies is another tool I use to make it so I need not cast.

Rapid Rejuv: This is the 3.3 change I am most looking forward to. A major increase to throughput for my most effective (and favorite) healing spell is welcome to compensate for other changes. As I see it, there are very few fights this won't be beneficial. I'm not sure about the rest of you, but most of the time I'm casting rejuv its reactive. Also, I rarely ever need to cover more than 6-8 targets with rejuv at a time. If there are 8 people in the raid who are spread out and have taken significant damage, rejuving them all as fast as possible seems like the best solution, and this situation seems pretty common. Also, this glyph will be excellent for tank "buffer" healing. Even in fights such as Twins, I have another druid assisting me pretty much every time we raid. Between the two of us, we can cover almost the entire raid even with this glyph...and our soakers and tanks will get a much-appreciated boost.

Innervate: I use my own. In fights I don't need it nobody else does, either, so that is hardly worth mentioning. As someone a few posts up said, I could take Spark and Duck Idol so I could spare innervate in nearly every fight, but I would be losing something like 430 SP which makes up a significant chunk of my healing. This sacrifice's rewards do not justify the cost IMO.

Regrowth: I'll have to play with it a little, and it seems like it might cover the few instances where the entire raid is taking damage all at once. However, as close as I cut it with mana, and it having such a small tick, I doubt it will see much use. It will probably remain something I just use on tanks as a buffer or for niche purposes like napalm shell on Mimiron.

Gearing for haste: Getting to 850 haste isn't going to be as easy as some people claim. Gear drops are imperfect, and we still have to share the loot with the rest of the raid. For a while, we will still have some slots that it will come down to keeping crit or going with a lower iLevel haste piece (and losing out on all the other important stats). Also, looking at the new "orb" craftable gear, its clear that Blizzard isn't going to do us any favors as 3 out of 4 pieces (including cloth) we would be interested in have crit instead of haste. I think its safe to assume that a lot of us will be under the haste cap for a while after 3.3 hits.

I use my Innervate on myself. The way I see it, either I will be gearing to rely on it or someone else will be, and the latter doesn't make sense from a raid efficiency/reliability standpoint. I will be the first to admit that there is a certain amount of downside to this approach, mostly because of Rebirth (either having to Innervate someone I Rebirth, or dying shortly after I use my Innervate and getting BR'd just to OOM right away). But those are exceptional cases and the alternative is, IMO, worse... i.e. using Idol of Awakening and Spark of Hope over Flourishing Growth and Illustration is a very large sacrifice of throughput for an uncertain return.

Glyph of Rapid Rejuv: we'll just have to wait and see. (Likewise with the 4T10 bonus, which could be anywhere between awesome and worthless, depending on the fine print.) I'm skeptical of GoRR, or at least I'm skeptical that it will be a benefit in every fight; we might end up carrying around stacks of glyphs and swapping them out as needed, which would kinda suck. Much depends on what the Icecrown bosses throw at us.

Although I have 4T9 now, I still use 4T8 quite a bit, especially in 10-mans. Even post-nerf, the 4T8 bonus is extremely powerful. Blizzard screwed up there and I'm surprised it took them as long as it did to "fix" it.

I've been meaning to bring this up for a while. Nobody corrected this post, but I was curious about it. It basically refers to the latest version of GoTEM:

To me this is worded as it the meaning was (10% haste) + (reduction of GCD to 0.9 seconds). Personally, I'm really hoping my interpretation is wrong and we only need 442 haste to hit the 1 second GCD. If I'm right, the second part of the talent is gonna be hard to make useful. With only 10% haste from GotEM its gonna be hard enough to hit a 1 second GCD. To get a 0.9 second GCD, its gonna be downright ridiculous.

I think that's just a bad misprint, and what it's supposed to be is, "increases spell haste by 10% and further reduces the GCD of Lifebloom by 10%." We'll see though.

How does it currently work on the PTR? I haven't checked in a few days; maybe I'll try tomorrow.