The discussion of Holy Smite DPS is still quite relevant for PVP, and there are many class mechanics threads on PVP. I still find this theorycrafting useful, even if it is not necessarily applicable for endgame raiding.

Shadow: SW:P, DP, VT, MB, MF.
Holy: SW:P, DP, HF, MB, Smite.

They differ only in the name of the filler and one of the dot/cooldown abilities - ignoring the fact that Shadow has much better DPS scaling talents. The playstyles are actually quite similar. Holy can switch back and forth between DPS and healing more easily, Shadow provides raid utility, 25-30% more DPS, and can sustain that DPS indefinitely.

There's really not much theorycrafting to do. SimulationCraft will tell you how to optimize Holy DPS just as it does for Shadow. Dual-spec has eliminated the need for "hybrid" specs almost entirely. There's no point discussing how Blizzard could make Holy DPS viable, when it is obviously their intent that the ability to put out adequate DPS be mutually exclusive with the ability to heal effectively.

What's left to discuss?

Smite builds are in the same boat as Protection Warrior, Holy/Protection Paladin wearing DPS gear : unsupported by Blizzard for PvE (when not explicitly killed like dual-wield Devastate builds), however fun to theory-craft.

Damaging/healing spells are a flavor of Holy school (Penitence, Holy shock, Holy nova), they have no equivalent in Shaman and Druid Nature toolboxes. WotLK reintroduced Judgment in Holy paladin gameplay, maybe a place could be done to Holy fire and Smite in PvE.

Casting a holy fire here and a smite there while healing in a BG or arena is not Holy Smite DPS. What theory crafting needs to be done? When you get a chance, you cast a damage spell and hope that your partner doesn't get burst down (or you get locked out of your holy spells) during the cast.

If you want to spec in to a damage build that sometimes casts heals, why not just go shadow and sometimes throw PoMs and shields? If you want to spec in to a healing build that sometimes casts offensive spells, why not just go Disc or Holy and sometimes throw Holy Fires and Smites?

Smite DPS will never be viable in an end game sense. Period. Last I checked, EJ does not really care about min/maxing the leveling experience on a spec by spec basis. Can this discussion please die, once and for all?

I used smite dps for end game content. In 10-man heroic Anub'arak we were having a problem meeting the dps requirement to get him to 30% before he burrowed a second time, but we couldn't go down to two healers because they couldn't handle the healing in phase 3. Specifically, they could put out enough hps to get through the amount of damage, but since they chain cast, they would often be on GCD when penetrating cold appeared and making sure both people got heals in time was presenting as a problem. So, I respecc'd to do bad dps for the first half of the fight and bad healing for the second. Now that we are wearing mostly 245s instead of mostly 10-man ulduar gear I don't know if this strategy is really necessary anymore (we are winning the dps check by a wide margin) but it got us our first kill, and our second, and we haven't found a need to swap over yet.

As a poster above said, having someone do bad dps for half the fight and bad healing for half the fight is not some "ace in the hole." Quite right: it makes more sense to say the shaman putting out 11k hps is an ace in the hole than the priest putting out 3k over the same span, but whether I was the ace or not this was the raid configuration that won the fight. Numerically I underperformed terribly on our first Anub kill, but we killed him, and for the forty or so attempts before that we did not. I also know this is more for 10-man guilds than 25-man ones, as with 25-people you have more granularity in how many healers they can bring into a fight, and usually have to worry less about raid composition (e.g. you probably have a holy paladin in your guild, you don't have to drop your only member who can provide +10% AP to make room for +13% spell damage and replenishment, etc.). But 10-man content is not trivial for 10-man guilds (some fights have actually been harder, though I admit that was primarily pre-ulduar), and a discussion of strategies relevant to beating that content is within the definition of end game raiding content as I understand it.

You do realize though that shadow priests heal perfectly fine, albeit a bit inefficient, if they manage to drop off their shadow form, right? You don't even need to misspend a single talent point either, if the healing requirement is basically to be a GCD reliefer. And you can shield in shadowform, if that is all that is required.

Having 2 shamans with healing streams (who are or are not healers) is probably still the better solution, but in the end it depends on your exact requirements and problems.

Defending a weird unusable spec for one single gimmick in one single encounter however is quite an acomplishment in itself though, I think.

PvP discussion can be in the Priest PvP thread, and I believe there's been discussion about how to play offensively with smites and such in there.

The plain fact is Holy has and never will be intended to be a PvE DPS spec. That's what the Shadow tree is for. There is really nothing left to discuss here.