wnoise comments on "Life Experience" as a Conversation-Halter - Less Wrong

11 Post author: Seth_Goldin 18 March 2010 07:39PM

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

Comments (65)

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

Comment author: SilasBarta 19 March 2010 02:27:12AM 7 points [-]

Well, turnabout is fair play. I'm not an old fart, but I've been in a position known for pleading inability to convey their knowledge to the unwashed masses until they they get roughly the same experience. Specifically, that of a graduate student (in control systems), working on a problem at the boundary of current knowledge.

I was very interested in learning what it would be like to get to a state where I literally could not explain the problem I'm working on to people far outside my field (though of course that was not why I went to grad school).

And you know what? It never happened. To an intelligent person, I was always capable of bringing them up to speed on the problem I was on and the related mathematical tools and formalisms. It certainly took some time as I had to fill in the gaps in their mathematical background, but absolutely not on the order of years. Maybe an hour or two instead.

I might one day find strong enough evidence to reverse my position, but for now, as best I can tell, the excuse of "you have to gain years of experience to understand my position" has been so overused, that it is extremely weak evidence whenever someone offers such a self-assessment.

I think what's happening is a combination of the "unwilling" and "unable" factors:

Unwilling: You take a deep hit to your status anytime you provide others with enough knowledge to obviate your sage wisdom.

Unable: If you haven't gotten into the programmer's mindset, you're all too quick to assume any problem has to be done manually and can't be converted into steps so simple that a machine could do it. "Nah nah, to play good chess requires special intuition, there's no way you can just break it down into a rulebook."

Comment author: wnoise 19 March 2010 03:27:26AM 1 point [-]

Maybe an hour or two instead.

That's often too long to be reasonable, of course.

Comment author: SilasBarta 20 March 2010 11:32:06PM *  0 points [-]

In some contexts, yes; in others -- like where they claim it's extremely important for you to believe, and took them years to get there -- it's quite a high rate of return to convey that information to you in an hour, and eminently reasonable to do so.

Also, I strongly suspect that most people in research positions haven't truly made their knowledge part of themselves, and so they couldn't ground it in its ultimate purpose (i.e. show how it relates to the rest of the world and show relevance to a layperson) even if they were given infinite time.

(Yes, I know I link that article a lot, because it's good.)

If I were wrong, you wouldn't see people so often fumbling through their explanation of how to use calculus and statistics properly in their fields, and you'd see researchers more often breaking down their problem into a purely mathematical one and hand it off to the experts at that. I remember reading an article recently that showed how ecologists have just now gotten around to using the method of adjacency matrix eigenvectors (i.e. PageRank) to identify crucial species in an ecosystem.