Well, delaying one HS one second is not that big a deal, no, at least in the grand scheme of things - but if you delay it one second each time, you essentially lose 16% of your Holy Shocks (or more accurately replace it with something else). You're quite correct in that the analysis is not easy due to all the possible haste effects you can have going at any given time - most or all of which stack with each other - and the procs we now have to deal with that change our rotation.

However, the big qualitative observation is that haste does not work as a flat healing improvement because it interacts with our rotation if we're using Holy Shock. It's probably true that the haste benefit from DF is never zero, though, assuming it's used properly (that is, you don't end up picking a bad rotation due to the added haste). For non-TOR though I believe the haste component is always a HpM loss unless it saves you a DL(reasoning: number of WOG/HS is constant while HL count increases); for TOR it should be HpM gain as well (number of WOG scales with HL).

I'm not necessarily suggesting that we -must- work around the HS CD with our haste values - just that the HS w/ CD is the only consistent heal we have that doesn't scale with haste (HR/LoD being the other major ones, but more situational). In particular if we didn't have the HS CD, haste really would be a straight HpS increase in single target situations. Since it's the only outlier, it's natural to attract a lot of attention (everything else is more easily understood).

But Shock costs less and heals for more than a Holy Light. So at least in early content, when mana is most of our concerns, casting Shock on CD will be a benefit even for a ToR style. Depending on the increase in mana-regen when gearing up in T11 and further and the amount of Divine Light we can/must cast, Shock might take a step back. But as long as Holy Light is our primary heal, Shock seems more efficient in all circumstances.

You are absolutely correct in the sense that by itself Holy Shock is better then Holy Light and mana will never be a consideration because even in pre-raid gear we have enough spirit for both of them to be virtually Free (In case of Holy Light - by the time you finished casting the 2nd HL the cost of the first one has been refunded), but the timing of its use should be dictated by situation more then our "rotation". If you have 2 HoPo and there is no raid damage to be healed, would it not make sense to just throw out another HL on our beacon target to generate that 3 HoPo instead of using HS and saving the CD if its going to become needed or if we have to move?

ToR style healing as I understand it is - HL* on beacon target, HS and WoGo on raid as needed.

* Substitute with DL when required.

But haste greatly benefits our Holy Radiance output. DF looks to be made for HR.

In the last couple of weeks things have changed pretty seriously for me away from the game, my company went bust and it has lead to me now moving from the UK to Australia; I've discussed it with Malthrinn and I think that it would be best if I gave the project of writing/maintaining the Holy Paladin compendium to someone else with the ability to do it justice. If you are interested please send a pm to Malthrinn; any details of the post you'd be interested in (such as the editing) can be found by attempting to quote the original post.

If BoL mana cost is still same as earlier then new version of glyph don't make much sense too. It is practically free now.

Since the new beta build our Light of Dawn AE has been changed back to using holy power.

Glyph of Light of Dawn: Light of Dawn can now affect 1 additional target.

Since we'll be stacking Spirit to a fairly large amount, wouldn't there be a point where we would only need one point in Enlightened Judgements? Or will 50% not be enough, even with the larger amount of Spirit we'll get, plus using some hybrid gems or enchants?

If we don't take 2/2 Enlightened Judgements then we have to put a point in Paragon of Virtue to get to 30 (all other available talents being fundamentally useless to a PVE healer). The jury's still out on this one, but according to Noules 1/2 Paragon is only a 0.5% throughput increase. I would rather have the flexibility in positioning you gain from having an additional 5 yards range on Judgement - which for my money is the main reason to be taking points in the talent to begin with.

Indeed, i was only considering the hit part of the talent, considering that the question was only about the necessity or not to take 2/2 to be capped for our Judgments.

In regard of Paragon of Virtue, you have to consider that now it lowers the cd on Hand of Sacrifice too, making it a stronger tank cd. In my testing with normal modes in beta (i didn't have the chance to try heroic modes, not yet) i didn't see a lot of tank damage, however if this is different in heroic it could be a very strong choice.

I want to clear something up there about the overall 0.5% throughput increase, because I think it is a huge misconception to be thinking about having a shorter cooldown on a healing increase spell as having an "overall" effect. Dare I say we should not be bothering to do the maths on such a thing? Personally, I'd say we should be relating the time of cooldown to what kind of damage that is incoming during a particular time in a fight.
To give a crude example - Does a 2 minute avenging wrath cooldown line up with the fact that the boss does a huge aoe every 2 minutes - "YES", Does a 3 minute avenging wrath cooldown line up with the fact that the boss does a huge aoe every 2 minutes - "NO".

As holy we are healers, not dps. DPS classes fundementally burn their damage/throughput cooldowns as many times as they can during a fight in order to increase the "overall" effect of their DPS over the entire duration of the boss fight. It does not particularly matter how fast damage is being done for the overall duration of the fight provided they can sustain it at a reasonably high level.
(This is of course excluding the fact there may be adds in the fight, but you'll still want a DPS class being able to sustain his dps at a certain level throughout the fight regardless)

As healers, the amount of healing we put out is situational to the amount of damage incoming from a boss. There are times where damage is being taken only by the tank, and there are times where the raid is taking extremely high amounts of damage. You can't put an overall number on cooldowns such as avenging wrath because that is totally opposite to the way that healers heal. We don't burn things like avenging wrath on cooldown just to get an overall 0.5% healing increase throughout the fight because there may be almost no damage coming in at that particular time, in which case the extra 20% would be completely wasted. What you CAN put a number on is being able to say, will that cooldown enable me to keep up with the amount of damage that is incoming at that particular time (say if there's a huge amount of damage in those 20 seconds). In a vast amount of cases, having a shorter cooldown on such a spell as avenging wrath gives us a greater opportunity to increase our throughput when it's needed at that particular point in time.

~

As for your point on enlightened judgements, i'm beginning to think that having extra range on judgement will be more valued when cataclysm comes, since it looks like we will probably be the most position-dependant healers out of the 4 healing classes. Holy Radiance heals from a 20 yard radius from where we are standing. Light of dawn is a cone effect heal and needs to be carefully positioned. Other healers have the ability to place their aoe radius heal whereever they like, although saying that, theirs are static, so in that regard we have the advantage ;>

~

And finally going back to Noules point on Divine Plea; I'm sure you understand that the concept of having to use divine plea alongside avenging wrath was down to the fact that we had to heal at maximum throughput for long periods at a time. Anything less and the tank was extremely likely to drop dead within the space of a matter of seconds. The developers have stated that there won't be situations in cata where a tank could get globalled within the space of a heal, it seems more likely that the periods in which we use divine plea at low periods of damage where nobody is at risk of dying will be separated from the periods in which we'll have to push healing using avenging wrath.

I'd like to end (ironically) with the fact that using avenging wrath in conjunction with divine plea would give us a lower overall throughput than by using both separately! ;D

Something to note about Avenging Wrath is it can be considered a mana-saving cooldown. The extra 20% might be enough to let us use Holy Light instead of Divine Light for instance.