Besides agreeing with you, I also meant (and maybe failed) to bring out what it is about 3E social skills that makes me agree with you - namely, not that they can/cannot affect an NPC's actions (in my view, insofar as a social skill affects attitudes, it will affect actions, given that - crudely but not too inaccurately - attitude + belief yields action). Rather, it is the failure of the maths.
Okay, I see where you're coming from more clearly now. Thanks for clarifying that for me.
Not so much in the context of a skill challenge. As I said in the post you replied to, multiple rolls are required - which reduces the odds of success - and the other PCs are also going to be involved (at least in typical party play), which means that a single diplomancer isn't going to resolve the scene.
Which is still odd to me. I mean, yes, sometimes the entire party needs to help out, because the guy they're talking to makes it his business to talk to everyone. Other times, though, the diplomat says "stay close, shut up, and follow my lead" before seeing the king. In those situations, I dislike the skill challenge requirement (if there is one... as I said, my 4e knowledge is lacking) of having everyone be forced to contribute. As a game mechanic, it would help with balance, though.
What have you got in mind?
In the abstract, turning a major villain into a friend who will expend resources to help you shouldn't be an more or less abusive than killing the same villain and stealing all his/her resources. Although it many be less exciting.
But I have a feeling this isn't what you're talking about.
It's definitely a much bigger swing than killing him and looting him.
Let's say there are five Bad Guys. I kill one, and now there are four Bad Guys.
If we compare that to convincing him to help me, there are now four Bad Guys, and one Good Guy. That's a dramatically different situation, in my mind. My players would deal with that variable completely differently than just "four Bad Guys left."
And, the more it happens, the bigger the swing. Having "three Bad Guys left" is very different from having "three Bad Guys left, and two Good Guys on our side."
But, as far as what I'd suggest, like I implied, is getting rid of anything that affects long term disposition. The GitP link I provided is a good start, in my mind, as it completely changes the nature of Diplomacy from long term outlook on a member of the party to whether or not somebody agrees to a very specific deal. Diplomacy checks no longer affect any NPC's long term feelings about the party, and that prevents a lot of abuse.
This means that you can convince a Bad Guy not to come after you, but you'll have to overcome the higher DC. It's 15 base, plus 5, 7, or 10 higher, for his relationship with you, plus his hit die or level, plus the risk vs. reward.
So, a level 5 Bad Guy that's a basic enemy (no personal relationship with you) has a base DC of 25 to affect. That's still easy to accomplish by level 3 (as I pointed out earlier), but if the deal isn't good, then the DC goes up by 5 or 10. This at least makes it more difficult. However, if you make him a very good offer, then he'll agree to it, as it's very favorable to him.
In this scenario, if you want to get him for sure, you might have to give something up (just to make the Risk vs. Reward break even). So, if you succeed on your negotiation, you lose something, and you have four Bad Guys left (and no Good Guys on your side). If you kill him, you get his stuff, and only have four Bad Guys left, but now you're taking a risk (with combat). If you make him a good enough deal, he'll switch sides (but the Risk vs. Reward will be hard), but it's not a long term thing, inherently. If something comes up, there's nothing stopping him from changing his mind again.
This sort of short term attitude changing is what I think makes for a much better balanced set of skills. Of course, in my opinion, you still need to bring skill checks lower (I don't use skill synergies, magic items as D&D knows them, and my max ranks are 3 lower than normal, amongst other changes) to work with this change, so that things don't always become "let me make a short term deal that has no chance of failure (unless you're really high level)" past a certain point.
And, of course, I changed Intimidate to work with the new "Diplomacy" (I call it by a different name, and modified the GitP skill). It now affects how they see you on the Risk vs. Reward section, and scales the more you beat the DC by.
At any rate, I'm not sure if this is the kind of thing you're asking for. Hope it gives you some idea of my preferences. Again, though, it's not for everyone, and I'm not advocating anyone change their style or anything.
As always, play what you like
