Exploration
<snip>
Here, I don't think the game is actually lacking much. Honestly, the most appropriate resolution is probably "the players declare what they're doing; the DM describes what they find." Sure, you'll need some mechanical support, to decide whether they actually find the trap/clue, survive the wilderness, or understand the symbolism in that painting they found (or whatever).
But I don't think it's appropriate to have a massive, rules-heavy resolution method here.
<snip>
Roleplaying
<snip>
I don't think the game is actually lacking much here. Once again, the players will largely drive the action, with the DM narrating the results. And, again, there is a need for some mechanics to determine success or failure, but again I don't think this wants the same level of detail as combat scenes.
<snip>
Action Scenes
<snip>
The difference between these and the exploration/roleplaying events is that these tend to be (hopefully) exciting, fast-moving scenes with immediate life-or-death consequences. As such, it's entirely reasonable for the game to move to a much finer granularity of resolution here.
I think I look at each of these quite differently.
On exploration, I tend to agree with SabreCat:
There are other ways of dialing complexity than resolution methods. Does exploration involve graph paper and descriptions like "you proceed 50' down the corridor, then it turns to the left and goes another 20' before ending in a T-intersection"? Or is it "you make your way through the necromancer's labyrinth toward his lair. Everybody roll Dungeoneering, those who fail lose a healing surge to traps before they arrive"? Or somewhere in between?
In
a session that I GMed earlier this year, the PCs were sent 100 years into the past by some witches, and found themselves exploring an abandoned manor. The climax of the session was a combat against a swarm of necromantic spiders in a wizard's library and laboratory. After that combat I moved very quickly through the rest of the exploration - just telling the players the most interesting things they found in the rooms they hadn't yet checked out, and glossing over the details. Some sort of mechanical system to support this, rather than just my vague handwaving, might have been helpful for this.
On action scenes, I don't think it's relevant that they have life or death consequences for the PCs. Because from the point of view of the players, they're just more scenes in which the PCs may or may not achieve their goals, and any given action scene may carry no more emotional weight than a non-action scene.
Because the current design of the 4e mechanics almost inevitably causes action scenes - especially combat - to carry the single biggest load of investment in play, it means that a good 4e scenario will be one in which combat and emotional weight are tightly correlated. (A weakness of some WotC modules, in my view, is that they invovle combats which don't carry much emotional weight, and hence are just a waste of everyone's playing time.)
If the game had more flexible combat/action mechanics, then it would be possible to include combat scenes that carried little emotional weight and yet didn't needlessly bog down the game (eg wandering monsters, which I don't particularly care for but that many others do). And mechanics for non-action scenes that could be dialled up to reflect the same degree of investment as 4e's combat mechanics would open the door to scenarios in which non-combat scenes are able to carry a greater amount of the overall emotional load of the scenario.
Which brings me to social conflicts:
there's no reason to assume that a social combat system would have to produce the same results, if it were better designed than Skill Challenges are for structuring social conflicts.
[MENTION=386]LostSoul[/MENTION], not long after 4e came out, had some posts on these boards where he pointed out a key dynamic of skill challenges: the GM doesn't get to roll. Which means that to get the sort of back-and-forth, and resultant tension, in a social skill challenge that is there in [MENTION=12037]ThirdWizard[/MENTION]'s FATE example becomes tricky.
It's not enough to introduce complications or adversity only when a player fails on a check, because that limits it to two complications before the PCs fail completely. But it can be awkward to introduce complication or adversity in response to a player succeeding on a check, because a successful skill check is, intuitively, meant to bring good consequences.
The DMG2 goes some way to suggesting how skill challenge successes can be used to mark progress through a series of complications, rather than just gradually accumulating the elements of success - and I have used some of those ideas to run successful skill challenges - but its certainly an area where more advice and examples from the designers could help.