Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Jonathan_Graehl comments on The Danger of Stories - LessWrong

9 Post author: Matt_Simpson 08 November 2009 02:53AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (103)

You are viewing a single comment's thread. Show more comments above.

Comment author: Jonathan_Graehl 09 November 2009 11:17:50PM *  4 points [-]

Quickly judging people as not worth listening to is a fabulous heuristic, especially given the Internet explosion of available alternatives.

But sharing such judgment risks offending people who didn't make the same cut.

Comment author: Vladimir_Nesov 09 November 2009 11:26:05PM 3 points [-]

Following such a heuristic doesn't at all mean making strong high-certainty judgments.

Comment author: AdeleneDawner 09 November 2009 11:33:41PM 2 points [-]

Strength of emotional response and certainty of the underlying heuristic's accuracy aren't the same thing. It may not've been clear that I was reporting the former, but I was, and one of the possible responses to that comment that I was prepared for was "yes, but he went on to make this good point...".

Comment author: Jonathan_Graehl 10 November 2009 12:09:11AM *  1 point [-]

I agree, but the fantastic thing is that you lose so little when you reject too hastily. If the ideas you ignored turn out to be useful and true, someone you're willing to listen to will advocate them eventually.

Comment author: Eliezer_Yudkowsky 10 November 2009 02:11:41AM *  3 points [-]

That works if you assiduously and diligently and without flaw, start paying attention after no more than the third time you hear the idea advocated, and without using the idea itself to judge untrustworthy those who otherwise see competent.

In practice, people usually reject the idea itself and go on rejecting it, when they claim to be acting under cover of rejecting people. Consider those who die of rejecting cryonics; consider what policy they would have to follow in order to not do that. What good is it to quickly reject bad ideas if you quickly reject good ideas as well? Discrimination is the whole trick here.

I suppose we might have no recourse but to judge people and shut our ears to most of them, in the Internet age, but to say that we "lose so little" far understates the danger of a very dangerous policy.

Comment author: Jonathan_Graehl 10 November 2009 08:48:26PM *  2 points [-]

I agree that people often don't make the necessary distinction between ideas they have evidence against, and unevaluated ideas they've been ignoring because they've only heard them advocated by kooks. As you point out, only ideas in the prior category properly discredit their advocates.

Comment author: AdeleneDawner 10 November 2009 02:36:35AM 0 points [-]

There's more than just the one non-failure mode to this kind of thing. My method involves taking the time to consider the information gathered up to the point where I decided to stop listening to the person, as if I hadn't stopped listening to them at all. Information that I would've gotten from them after that point isn't affected by my opinion of them, since I haven't heard it (where it would be, if I were distracted by thinking 'this person's an idiot' as I listened to them), and I give as fair of a trial as I'm able to to the rest.

It may also be noteworthy that I didn't judge him for an argument he was making, and I make something of a point of not doing so unless the logic being used is painfully bad. (Tangential realization: That's why activists who aren't willing to have any 101-level discussions with newbies get a (mild) negative reaction from me; discarding a whole avenues of discourse like that cuts off a valuable, if noisy, source of information.)

Comment author: AdeleneDawner 09 November 2009 11:34:26PM *  1 point [-]

But sharing such judgment risks offending people who didn't make the same cut.

I figured that out, but bringing it up seemed like it would just compound the problem.