It would be like looking at the upper left section of the Mona Lisa and assuming there are no people in the picture. There might be, but there might not be. There is a dearth of information still.
You didn't use a generic food analogy!
The thing is that the upper-left section of the Mona Lisa wouldn't be a very good "preview." Maybe a sketch of it before it was finished. Maybe a completely finished eye. You'd save the smile for the true release. The upper-left section of the Mona Lisa doesn't tell you anything you WANT to know. It's functionally useless.
The rogue is bigger than that. It's like the left eye. I can't say with certainty what the smile or background is going to be, but I can probably tell that we're drawing a person, perhaps a woman, and that she's caucasian, and she probably has a right eye that's just like her left, unless Leonardo wanted to pull some tricks on us. I don't know what the whole effect is going to be, I don't know the center of the painting, but I can say that there is evidence for my claims.
I could be surprised, and I don't know everything, but I do have evidence for the claim. I might even criticize it based on that evidence: "Oh, great, another brown-eyed girl, like we don't have ENOUGH of those!"
Whereas someone who came along and said "She might be hazel-eyed, or maybe she'll be different colored in her eyes depending on how you look at her, or maybe she'll have three other eyes on tentacles coming out of her head! You're just seeing what you want to see because you want to be critical!" wouldn't have any evidence for the claims. She might, sure. It is just a preview. But one position is
supported by that preview. The other is
not.
If the rogue is more like the upper-left corner than the eye, then I can say that it's a HORRIBLE preview that is, in fact, directly MISLEADING.
It might be. But I do have some faith in WotC's competence in choosing what little bits of information to release, and that these bits of information are in line with what they're telling us about philosophy means that there is PLENTY of evidence for my position.
Yes, the Mona Lisa might have tentacle-eyes, but you're going to have to show me that, or make be believe that the preview is directly misleading.
Yeah, the rogue probably has more powers, even a greater diversity of them, but that doesn't mean the rogue is suddenly Captain Flexibility and "we just don't know enough." Evidence points to it NOT being very flexible, and to the designers believing that a narrowly focused class is a GOOD thing (and probably just hope it's not 'too narrow,').
So if we can drop that little "not enough information!" canard, that'd be
great.
Cadfan said:
A lot of this boils down to archetype fiddly bits. Suffice it to say that a selection of skills doesn't back up an archetype alone. It needs to be reinforced at multiple levels, especially something that's as counter-intuitive as a medieval action-adventure movie featuring a crime-solving detective with a pipe and an overcoat. It needs abilities, it needs to combat as if it were that character, and it needs to have the 'telling marks' of the archetype.
3e helped achieve this archetype with the rogue because of the vast number of skill points and the way that skills worked to begin with. Putting 1 point in 10 different Knowledge skills (with a few left over to max out Search and Spot and Listen, and a handful of points in Gather Information maybe) worked out to be a very "Sherlocky" thing. 4e mandates some of your skills (stealth and theivery), and looks to be getting rid of knowledge (in favor of several different skills like Arcana and Nature and the like).
But here's something interesting:
Cadfan said:
What would be a particularly Sherlock Holmes-ish way of killing orcs? Seems to me that using one's powers of perception to slip a dagger right between the plates of their armor is a pretty good option. Is there a better one? It seems like this is just the usual problem of taking a character from one genre (non violent detective work) to another(medieval style fantasy).
It's perfectly fine to cede that 4e abandons the Sherlock trope entirely. 3e only gave a vague passing nod to it's possibility in the first place, and it is dissonant. In which case my criticism remains true, but becomes slightly more damning: In 4e, you can't play Sherlock Holmes-type at all (at least at launch). The game doesn't support it. Welcome to house rule and/or different game territory.
Of course, we have no evidence that 4e is wholeheartedly limiting itself in genre like this, or that it becomes impossible to play a nonviolent detective campaign or session, and I still think that D&D is "too generic" to not give a nod or three in the direction of at the very least a mystery-centric adventure or two.
So we're left with ideally making Holmes fun even when he's killing orcs, not just when he's solving crimes (but especially when he's solving crimes). Some basic seeds for abilities might include something like:
Momentum Shift: The character trips a creature who moves past him, ending that creature's movement.
Demoralizing Lecture: You point out how ignorant the target is. Intelligence attack vs. Will to gain combat advantage
Familiar Pattern: You become familiar with a creature's fighting habits and gain your Int bonus as an AC bonus against a creature that has hit you once in combat already.
Divinatory Knack: You can learn to cast 1st-level divination spells.
Trained to Subdue: You can make a Grapple check to pin an enemy and hold him immobile
...or, come to think of it, almost anything from the Archivist's Dark Knowledge class feature.