We're talking about the rogue/thief trading backstab, for a non-combat ability.
I was unaware you were in charge of deciding what may be discussed on any given thread. I think the rest of us, or at least some of us, may view the issue a bit wider. This is especially ironic given the next post indicates the original tweet that motivated this thread was directed at swapping sneak attack for a different
combat ability only.
So now that we've dealt with the majority of what I view as a strawman, you tell me why this rogue/thief would be useless and bored in combat without sneak attack? I played that in 3e (it was an option) and it was great.
I set that out above. If the Halfling Rogue has a slightly better attack bonus and the same damage as a human wizard in melee (ie 1d4), that seems pretty boring to me. Could the Rogue have better "no tradeoff combat" abilities to resolve that? Sure. Could it be resolved with a choice of combat-enhancing abilities, one of which is Sneak Attack? Again, sure. Must the ability add damage? No, not really (see below). But "hey, you can dump combat for noncombat or vice versa"? As a default, my preference is NO. As an optional rule, caveated for the issue of bored/frustrated players and overspecialization, sure - more options will never be a bad thing.
Well when I originally asked Mearls on twitter the question if you could be able to switch out Sneak Attack for another class feature, I meant for another combat ability. Because of twitter's character limit, I could not ask the question completely.
Thanks for that clarification!
To me, a rogue should not be locked into offensive and underhanded tactics of Sneak Attack. They could get damage reduction equal to their "sneak attack dice" if the attack doesn't have advantage (perfect for acrobats). Or they could get a henchman/bodyguard/pet with "sneak attack dice" for HP (perfect for lazy noblemen and gangleaders).
Both good ideas. Something aimed at the ranged combat rogue, who doesn't get in there and flank, would also be nice. And even a choice of three or four in the base rules gives us something to build on, rather than the possibility of some other option being in PHB V, when "sneak attack" is ingrained, and there are a host of other abilities building off it on the assumption each and every rogue will have that specific ability.
But according to Mearls' reply, a lot of people hated not getting SA on the rogues. So it would be in splatblooks if it ever came.
As a very revolutionary concept, wouldn't it be nice if the first release at least made an effort to be a complete game, not the obvious first volume in the PHB Encyclopedia? Yes, I know, bad business to make the game complete in its core rules rather than spread the rules over an encyclopedia of sourcebooks that everyone will have to buy to have a complete set, and that ship has long since sailed in the D&D world. It hasn't exactly made the Hero system publishers business take off. But it would be nice to buy the core rules without feeling the Money Siphon push its way into my wallet to wait for the next dozen volumes you need to actually have all of the truly "core" rules.
At the moment, Next lets me build a 1st level Illusionist who has no damaging spells: Light, Minor Illusion, Prestidigitation; Charm Person, Comprehend Languages, Detect Magic, Fog Cloud.
Charm Person is combat-useful, as is Fog Cloud. Let's not equate "useful in combat" with "deals significant damage", or even "deals damage". Minigant suggests an ability for an acrobat that mitigates damage taken, and a character who can distract one or more of an enemy group for a few rounds while taking limited or no damage in return is also "useful in combat". Lots of other options exist.
As I gain my two spells per level I have plenty more non-damaging options, too: Alarm, Grease; Darkness, Invisbility, Levitate, Spider Climb; Blink, Dispel Magic, Fly, Water Breathing; Arcane Eye, Dimension Door, Polymorph, Stoneskin; Contact Other Plane, Passwall, Scrying, Seeming; etc.
Most of these have definite combat advantages. I suggest that the wizard who has no combat-useful spells, or who has no non-combat spells (like Arcane Eye or Alarm) is pretty rare. Again, "non-damaging" is not "useless in combat". I can't count the number of times a spellcaster Dimension Door'd himself and another character to a better location (to flank and sneak attack; to get at the spellcaster behind his line of bodyguards; to get past the distraction to the real target faster). If we have no combat, how useful is Stoneskin?
Is that a build that the game shouldn't allow?
No. What I don't want is a spell list restricted to, say, scrying spells and other spells with no, or virtually no, chance of being useful to cast in combat so the wizard player is reduced to stabbing with his dagger or firing his crossbow at 10th level, resulting in him breaking out the tablet to surf the web whenever combat starts, with a postit on the back that says "at my initiative, duck and cover".
This is not the case for 4e, which does not adjudicate social encounters as "one roll, succeed or fail".
Again, I'm not very familiar with the 4e mechanics, so I can't comment on specifics. I'd like to see options for adjudication methods, actually. There's nothing wrong with "one roll, succeed or fail" for tasks that have limited game impact, and a more detailed, granular resolution system for more important challenges on which the specific game focuses. In a game of political intrigue at the King's Court, maybe social conflict should have a deeply granular resolution system playing out over many "attacks" and "counterattacks", and a duel with the King's Champion should be a one roll resolution.