PeterS comments on Rationality Quotes - September 2009 - Less Wrong

2 Post author: thomblake 01 September 2009 03:06PM

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

Comments (101)

You are viewing a single comment's thread.

Comment author: PeterS 01 September 2009 06:03:47PM 1 point [-]

A single conversation with a wise man is better than ten years of study.

Comment author: Eliezer_Yudkowsky 01 September 2009 10:02:33PM 16 points [-]

That had better be a long conversation, a very wise person, and one damned lost field you were studying for ten years.

Comment author: brian_jaress 03 September 2009 08:25:27AM 1 point [-]

This is a lesswrong quote, but I think it belongs in this discussion because it's remarkably apropos:

I remember when I finally picked up and started reading through my copy of the Feynman Lectures on Physics, even though I couldn't think of any realistic excuse for how this was going to help my AI work, because I just got fed up with not knowing physics.  And - you can guess how this story ends - it gave me a new way of looking at the world, which all my earlier reading in popular physics (including Feynman's QED) hadn't done.  Did that help inspire my AI research?  Hell yes.  (Though it's a good thing I studied neuroscience, evolutionary psychology, evolutionary biology, Bayes, and physics in that order - physics alone would have been terrible inspiration for AI research.)

-- Eliezer Yudkowsky

Comment author: PeterS 01 September 2009 11:30:56PM *  -1 points [-]

Wise in comparison. The other quantifiers are hyperbole.

The quote applies insofar as the field being studied is already somewhat mapped and investigated. Programming forums are ultimately more useful than tutorials and textbooks, talking to an author more insightful than reading the 200 page thesis, and having access to a community of intelligent synthetics is much more valuable than having access to a library.

Comment author: sketerpot 02 September 2009 10:28:25PM *  5 points [-]

talking to an author more insightful than reading the 200 page thesis,

Find the right books, and it'll probably be far more rewarding than talking to an author, simply because of the information density and better organization that you can get in written form.

An hour of reading Hennessy and Patterson's excellent book on CPU design will teach you a hell of a lot more than six hours of classes. I speak from recent quantitative experience here, which is where I got those specific numbers. The exceptions to this rule are local: particularly hard-to-understand concepts like the Tomasulo algorithm are a lot easier to wrap your head around if you have someone to walk you through them. But for the most part, a well-written textbook can teach you better than a person talking with you.

One problem is that most textbooks just aren't written that well. Often they're too concerned with signaling academic status, and they forget to make the book something that people will want to read. Just because an author can go off on a tangent about graph isomorphisms doesn't mean they should. Other times they get bogged down in obscure details up front, killing off people's interest. There are other failure modes, too depressing to list here.

By the way, I think that one reason why wikis are so easy to learn from is because you can skip past the boring stuff until you need it. This makes reading a wiki more fun, and also leads to tab explosions, keeping you hooked. I figure that this could significantly improve on the traditional textbook model, despite all those nice things I said about it earlier in the post.

(In honor of the tab explosion, I've stuck in a bunch of links to pages that might be interesting.)

Comment author: Kaj_Sotala 02 September 2009 06:11:26PM 5 points [-]

talking to an author more insightful than reading the 200 page thesis

Only if he were an exceptionally bad writer. 200 pages contains a lot more information than you can fit into most conversations. Not to mention being more logically structured.

Of course, a conversation is more interactive and lets you ask about the things that were left unclear, as well as clear up misunderstandings... but I don't think that anywhere near compensates.

What you could argue is that talking to the author is time more efficiently spent, as it gives you a better idea of whether her thesis is worth reading.

Comment author: Nominull 02 September 2009 07:30:34PM -2 points [-]

most people are exceptionally bad writers

Comment author: RolfAndreassen 02 September 2009 07:44:51PM 4 points [-]

How can that be the case? You apparently have 'exceptions' forming most of the population!

More generally, being able to talk to the author after reading the thesis is hugely valuable, because whatever was unclear in the thesis can be cleared up. But talking to the author without reading his work is fairly worthless; you won't know what questions to ask, unless of course you're already knowledgeable in the field.

Comment author: rwallace 02 September 2009 12:43:11PM 0 points [-]

But actually it also applies only insofar as you have already studied the field. Programming forums are great, but we've all seen the guy who shows up to post a tutorial question verbatim and appends "send me the code plz", and we all understand he's just wasting everybody's time. You have to read the textbooks and at least seriously attempt the tutorials yourself before you can ask the right questions on the forum.

Comment author: SilasBarta 01 September 2009 09:40:31PM *  0 points [-]

Meh. Maybe if that wise man is Eliezer Yudkowsky. But then, calling Eliezer Yudkowsky a "wise man" is like calling the Sahara Desert a "litterbox".

Or Chuck Norris a "tough guy".

ETA: Grr! I can't put underscores in people's names anymore without adding italics!

Comment author: dclayh 01 September 2009 10:35:51PM 2 points [-]

Is it perhaps time for another round of Eliezer Yudkowsky Facts?

Comment author: thomblake 02 September 2009 12:05:14AM *  0 points [-]

Just backslash escape them - type it like this: Eliezer\_Yudkowsky

ETA: this is an amusing example of "do as I say, not as I do". What I actually typed looked like this: Eliezer\\\_Yudkowsky

Comment author: SilasBarta 02 September 2009 03:30:28AM 0 points [-]

I think you misunderstand the problem. I know I can override the formatting. It's just that if I did so, it would invalidate my claim that I have software that lets me call up forum screen names with hotkeys, and replaces the spaces with underscores in forums that allow spaces in names.

In other words, people would know I was needlessly typing out their whole screen name and adding underscores, and it wasn't just some glitchy software.

Comment author: darius 02 September 2009 08:32:49PM 1 point [-]

You could see that as anthropomorphizing the power of interaction.

Comment author: kpreid 01 September 2009 07:32:03PM 0 points [-]

Of who is this a quote?

Comment author: PeterS 01 September 2009 07:58:24PM 1 point [-]

I believe it's actually a Chinese proverb.

Comment author: billswift 01 September 2009 08:47:57PM 6 points [-]

You need to be careful of supposed Chinese proverbs; I recently found that the Chinese curse "May you live in interesting times" was actually created in the fifties by Eric Frank Russell.

Comment author: SilasBarta 04 September 2009 03:41:19AM 1 point [-]

Maybe the problem is that you're focusing too much on whether the proverb is authentic Chinese rather than on whether it accurately captures reality?

Comment author: ChrisHibbert 04 September 2009 03:28:18AM 1 point [-]

It has to be "may your grandchildren live in interesting times", or the caster of the curse is as cursed as the recipient. sheesh!