"Roleplaying" skill vs Character Skill

[Garbados]

Do you limit the appropriate skills? Or do you just assume that whatever they learn is always appropriate to the situation? I rather the latter, based on the straightforwardness of The Shadow of Yesterday's basic conflict system.

Depends on what the player wants to know. It can help to state an objective, and then ask questions that work toward that objective, and roll appropriate skills for each question. More than once someone would use a completely unrelated skill like Weapons to see if the NPC knew his way around a sword. If his Weapons exceeded the target's Bluff (or whatever your system calls that sort of thing), he'd get the answer. Bizarre or obscure inquiries incurred difficulty mods to the check, while easy questions like "Is that a fedora?" earned bonuses. if the target won the contest, he was free to give off whatever impression he wanted. ("No, it's actually a bowler." *shifty eyes*)

On the downside, it looses the "power of dice" effect when dealing with an uncertain or combative GM, and requires a bit more logic/umpiring from them, but if you include the old standby "roll-off and make the result a house rule" (for disputes about whether something should be reasonable/expected) then you should be fine!

It doesn't so much lose the power of the dice as much as it naturalizes the process. You can still show the GM "my Empathy check exceeded his Bluff" and he can't argue with those dice, but if the NPC simply won't believe you no matter what you say, well that's the NPC's problem. In this way, if your GM makes every NPC a jerk when it comes to social interaction, then yeah they can ruin the fun, but only a few systems have any real means of fighting this, e.g. by combative (read: bad) GMs less so. Uncertain GMs will have to learn the invaluable GM trait of improvisation, which they can luckily learn on the job. Any aspiring GM has to learn it sooner or later, regardless of system.

But you're right that it does hamper especially heroic PCs from simply brute-forcing the opinions of lesser mortals. So if your game does a lot of that, this might not be the best solution.

[JoyWriter]

But you're right that it does hamper especially heroic PCs from simply brute-forcing the opinions of lesser mortals. So if your game does a lot of that, this might not be the best solution.

Most of the time I'd say that's an advantage, Kant's categorical imperative and all that.

On the other hand, really playing with the fantasy of being able to "bend everyone to your will" could be pretty awesome, especially if it is treated in the "genies wish" style. For example, what if the daft logic suggested at the start of the thread "sell cheap before you loose it" starts to take hold in the village, and you get a hysterical selling of anything and everything until the town is emptied. The "reductio ad absurdum" fable potential of that kind of thing sounds awesome to me, though likely pretty incompatible with the mood of that game!

Bizarre or obscure inquiries incurred difficulty mods to the check, while easy questions like "Is that a fedora?" earned bonuses. if the target won the contest, he was free to give off whatever impression he wanted. ("No, it's actually a bowler." *shifty eyes*)

I think there's a lot of sense to that, provided the mods are agreed pre-roll in a nice transparent way, but I was actually referring to when you abstract a conflict to a single roll:

There is a difference between difficulty mods to answer questions, and difficulty mods to have the question be useful in your intent. I was thinking that after rolling for a piece of information, from hair colour to opinion on cardamom, (which presumably would be the only roll) the degree of success of this roll would be compared to that of the other negotiator, and the highest success be used to say who wins. Obviously there is a certain amount of absurdity to this, and finding appropriate limits to that absurdity is quite a trick, but the advantage is that you can for example, divert the trader into a discussion of spices so as to trick them into giving it to you for cheap, or something equally leftfield. Flexibility.

Alternatives to pure "all info is good" is to implement soft limits on what gathering can be done, shift the criteria for difficulty mods, or require a short explanation of the strategy to use the info which is ok'ed by the table (if no-one cares about this, just get on with it!). That's all I've got for now!

One thing you could include to make things more interesting is to give the same freedom to both participants of the negotiation; the extra freedom being given simultaneously to be used against their character may restrain the contexts in which this kind of system is used. Not sure if that works in your system, or whether it would instead produce 4 dice rolls (for/against on each side).

The point about uncooperative GMs is a good one, and picks up what is often the essence of this problem; mindblank on their part! Suggests to me that the solution to that problem is not in hammer mechanics, but in tools to help well intentioned but stuck GMs flow with what they're players are up to.