Obviously, at some point being polite in our arguments is silly.
I think you seldom convince someone to change his opinion by name calling.
I once went to a talk about the implications of neurology on economics. Unfortunately for the professor who gave the talk he had a badly dressed conspiracy theorist in his audience who was upset about the professor providing a new way to justify the economic status quo. That talk would have benefited from throwing out the conspiracy theorist instead of being nice to him. The reason isn't that the conspiracy theorist is dangerous but because wasted valuable time of a talk about neurology on economics.
Most situations however aren't like that. While I would have wanted the person thrown out of that talk I would happily talk with someone who holds the same position face to face or at a forum like LW.
I personally don't attempt to manipulate people against their will but if I would want to do so, dishonesty isn't the most straightforward way. If you interact with a person in that way you bring up resistance.
If you create rapport you can ask them whether they are happy with how their life is going and hit much deeper. That means both being able to do real psychical damage that goes deeper than a cheap insult and having a lever that's strong enough to change opinions.
You probably won't convince anyone, but you can probably discourage uncommitted/future people from taking the scorned position.
Scott, known on LessWrong as Yvain, recently wrote a post complaining about an inaccurate rape statistic.
Arthur Chu, who is notable for winning money on Jeopardy recently, argued against Scott's stance that we should be honest in arguments in a comment thread on Jeff Kaufman's Facebook profile, which can be read here.
Scott just responded here, with a number of points relevant to the topic of rationalist communities.
I am interested in what LW thinks of this.
Obviously, at some point being polite in our arguments is silly. I'd be interested in people's opinions of how dire the real world consequences have to be before it's worthwhile debating dishonestly.