My guess would be instant cast (going from precedent), but will probably have a cooldown, most likely on a par with CS/Silence type cooldowns (20-45 seconds). And you can't trinket any silence so this should be no different.

Regardless of that aspect, it does have the potential for great synergy because as aceofsween points out it cannot be dispelled per se, you need to move out of it. For it to be usable at all the "spot" will have to be a bit larger than a singular point and so take a few steps to get out of, and there are a lot of ways you can leverage other classes' snares, slows and stuns to extend the silence to a dangerous duration. When there are several players on top of you in most cases it's pretty hard to move without some snare/slow immunity CD.

Of course this is sort of viewing it in a WotLK PvP environment where people can be 100->0'd very quickly. With much larger health pools I can't see this silence being capable of killing anyone in a short combination of moves, but it will likely be a highly offensive ability all the same.

I like the concept a lot so far. It's what Balance druids have needed for a long time and it's unique, not just a "let's give moonkins wind shear" type solution.

I sort of disagree. Part of the fun of getting upgrades in this game is playing "upgrade tetris." I think if every stat gave a linear improvement, upgrading gear would be sort of boring. (I recognize the unreliability of haste buffs, etc. in raids as an annoying problem for gearing.) Although for Cata healers one big tradeoff that seems to be coming back is throughput vs longevity. We haven't had to worry about longevity for a long time.

It's an interesting exercise to think about how haste on hots/dots would possibly work without introducing thresholding effects. The only reasonable way I see would be to effectively convert haste into a spellpower buff, which is a bit silly and counterintuitive.

Haste breakpoints

I think Druids have HoT/DoT/Channeled spells that run from 3 to 10 ticks in LK. With Haste plateaus, here are your possible Haste breakpoints, below 100% Haste:

10.0% 11/10 <- 10 tick spell turns into 11 ticks
11.1% 10/9
12.5% 9/8
14.3% 8/7
16.7% 7/6
20.0% 6/5 12/10
22.2% 11/9
25.0% 5/4 10/8
28.6% 9/7
30.0% 13/10
33.3% 4/3 8/6 12/9
37.5% 11/8
40.0% 7/5 14/10
42.9% 10/7
44.4% 13/9
50.0% 6/4 9/6 12/8 15/10
55.6% 14/9
57.1% 11/7
60.0% 5/8 16/10
62.5% 13/8
66.7% 5/3 10/6 15/9
70.0% 17/10
71.4% 12/7
75.0% 7/4
77.8% 16/9
80.0% 9/5 18/10
83.3% 11/6
85.7% 13/7
88.9% 17/9
90.0% 19/10

Just to get a feel for what these mean, Bob the Moonkin has 700 Haste on gear. He's got 100% uptime (talents or raid buffs) haste bonuses of 3%, 3%, 5%. He has about 80% uptime on Nature's Grace (20% Haste). I'll ignore his occasional Heroism or speed potion.

His most common DoTs are nominally 7 ticks (talented IS), or 8 ticks (talented MF, fully extended with SF glyph). He occasionally has DoTs that are nominally 5, 6, 7 (MF not fully extended), or 10 ticks (Hurricane, and perhaps Starfall).

His haste, without NG is 35.2%, and with NG is 62.2%. His next couple of haste-from-gear breakpoints would be:

708=62.5%(NG) -> NG Moonfire (extended) goes from 12 to 13 ticks.
769=37.5% -> non-NG Moonfire (extended) goes from 10 to 11 ticks.

Obviously I'm using LK durations and ratings, and making assumptions (is extended MF an 8-tick HoT, or is it a 5-tick HoT followed by three 1-tick HoTs?) that could easily be wrong.

re: the haste breakpoints thing, I sort of saw it coming, and I'll be shocked if it doesn't apply to DoTs/HoTs too. There was just never a clean way to keep the duration the same and make haste scale linearly. It's still an overall improvement - it just means all casters (the serious ones, anyway) will need to be aware of the breakpoints and gear accordingly. One point I disagree on:
With the duration fluctuating, being just below a haste breakpoint could potentially cause issues with dps rotations. For resto druids, this would affect both "rejuv blanketing", if that strategy still applies, as well as keeping HoTs up on a tank in between direct heals.

The point is that to have your 100% uptime, and thus an actual increase due to haste, you may have to modify your rotation. Even if more haste is not -worse-, it's not necessarily -better-. More importantly, the value of haste can change dramatically in a few points - a very chaotic situation, and one which is not at all intuitive to the average player (or, well, any player, even if they're capable of doing the calculation).

The other things is that you don't necessarily want a shorter duration with HoTs, even if you are healing for the same total amount. This might be less of an issue in Cataclysm where players are expected to be at states besides 0 and 100% for significant portion of the time, but it may still be relevant.

On the other hand, in a rotation without any CD spells, this implementation of haste does mean the value of haste changes smoothly. In particular, if the filler spell(s) of the rotation divides evenly into a tick duration, the value of haste will be much more intuitive. As an example, if your 'rotation' consisted of a 15s HoT ticking every 3s and a 3s 'filler', you never have (enforced) dead time because the tick timer is an even multiple of your filler cast time. Once you start introducing short CD spells (and to a lesser extent, odd casting time spells), however, the value of haste becomes much less smooth, since for many values of haste you have to either wait for your CD or push back your CD ability (and if your CD ability is not one you want to hit on CD, it's generally not worth using).

Because of the odd interaction between CD abilities and haste, I suspect that Blizzard will try to reduce the amount of short CD limited abilities in Cataclysm to make haste more intuitive (this seems to be what is happening with Hunters and, reading between the lines, Ret pally, among others). As an example, Wild Growth might end up being a more powerful spell with a longer CD (15s? more?) so that there's more incentive not to use it on CD (waiting a second for a more optimal use of WG is much more palatable if it's 15s before your next one, versus 6s). Another possibility would be to reduce CD based on haste, though that seems potentially messy, too (tanks stacking haste for cooldowns!).

It should be noted though that DoTs operating in this manner doesn't seem very consistent with what Blizzard has been saying about DoTs being protected from clipping the last tick. It's not impossible for both to be the case, it's just that clip protection seems mostly pointless if DoTs worked like the way they described channelled spell operation.

I don't agree with this and thus a large part of the rest of your argument.
Hitting an ability on CD doesn't mean hitting it as soon as it's at off CD, but rather "as soon as it's off CD and i have finished what i'm currently doing". Taking Rejuv-WG rotations, latency and human reaction make it so that you don't hit WG precisely at the CD anyway but let's say a few tenths of a second after.

Many classes are in such a situation already, where having different cast time for spells and a priority rather than a rotation forces them to make some instant choices about what the best spell will be depending on how the status of their CDs, and it's just a matter of being smart with what spells you decide to use (for example a warlock choosing to Lifetap instead of Shadowbolting when near to having to refresh a dot).
The counter-example also works today already : casting Hurricane 0.5s before Starfall comes off CD isn't smart. Just change your rotation to take it into account.

Of coursing having to do that work every 5-6s could be annoying, but if we keep using 1s spells in Cataclysm then we can just ignore it, and if we get to use different spells a bit more then it'll be a choice that can be optimized, and maybe using a simple rule of "no long spell just before WG is off CD" will be enough. I rather hope they get us away from raid blanketing myself, so maybe WG on CD will be less important than it is now anyway.

I'm not sure how you're disagreeing with my quoted point here. For most values of haste and a fixed CD ability, you have either have dead time (where you are waiting for the 0.5s or whatever until the CD comes up) or you have to push back your CD ability (by filling in with something that will take longer than your CD timer). What I mean (if it wasn't clear) by 'if not hitting it on CD, you don't want to use it' is that useful, short CD abilities are typically ONLY constrained by their CD timer rather than anything else (that is, if there was no CD, it would have top priority over non-CD abilities) or other CD abilities, -not- that it's not useful if you cannot hit it exactly on CD (hence 'want to hit on CD', not 'can hit on CD').

The essence of that particular point in my post is that if you have fixed, relatively short CD abilities (unaffected by haste, and you'd want to use), the value of haste will vary in a chaotic fashion as haste increases. This would be the case whether or not DoTs and HoTs scaled as Blizzard's description of channeled spells (in particular, haste -would- scale smoothly despite channeled haste scaling if no high priority abilities are on short fixed CDs). This leads me to expect that either 1) short CDs will be affected by haste or 2) Blizzard will attempt to get rid of short CD high-priority abilities...or 3), haste will still remain a convoluted stat.

I think ultimately the most challenging encounters in Cata will still be of the "react quickly or die" variety, not "react quickly or healer will run out of mana before the fight is over" variety. I do think it is conceivable to have longevity fights at all difficulty tiers. One example would be a patchwerk type gear check fight. If you were to force healers to spam tanks in the Cata mana environment, it would automatically become a longevity check.

The big issue for Blizzard is that if mistakes merely cost healer mana and not deaths then it becomes too easy to outgear the fight and power through it, rather than have the fight serve as a check for "can your raiders dodge big telegraphed mimiron rockets". Insta-kill attacks were introduced from Ulduar on for that reason.

Really?

I remember when we started encounters in BC (particularly Kara and Gruuls) and it wasn't the fact that healers couldn't keep us up, it was that after a certain amount of time, they just ran out of mana entirely. It puts more pressure on the DPS because you couldn't fight everlastingly. There were several cases where we wiped on Prince because our healers ran out of mana.

This seems to be what they want to return to. The other side is that they don't seem to want to cap casters mana constraints so much.

Also if they manage to make mana matter for the first tier then you will have at least one more tier (be it the next or the one after) where the damage requires you to use the faster and stronger heals more frequently in lieu of the normal one you start off using which should provide some scaling mana consumption.

I don't understand how you can say that.

You can still have unforgiving DPS requirements for a fight regardless of the mana issues being brought back into WoW. If anything, they have reimplemented a dynamic timing effect where the DPS has to kill the boss in X minutes or the healers run out of gas and people start dying, but that time fluctuates constantly.

That doesn't mean they can't design a fight where if you don't kill the boss in 5 minutes, you die (think Festergut). It just means that they will also expect healers to last that long as well. In fact, if the fight is supposed to last 5 minutes and is on a strict timer, it gives you more leeway as a healer to go all out if your mana pool normally lasts 6 minutes. Basically, they can pressure healers just as much as anyone else in this manner.