My deconversion from Christianity had a large positive impact on my life. I suspect it had a small positive impact on the world, too. (For example, I no longer condemn gays or waste time and money on a relationship with an imaginary friend.) And my deconversion did not happen because I came to understand the Bayesian concept of evidence or Kolmogorov complexity or Solomonoff induction. I deconverted because I encountered some very basic arguments for non-belief, for example those in Dan Barker's Losing Faith in Faith.
Less Wrong has at least two goals. One goal is to raise the sanity waterline. If most people understood just the basics Occam's razor, what constitutes evidence and why, general trends of science, reductionism, and cognitive biases, the world would be greatly improved. Yudkowsky's upcoming books are aimed at this first goal of raising the sanity waterline. So are most of the sequences. So are learning-friendly posts like References & Resources for LessWrong.
A second goal is to attract some of the best human brains on the planet and make progress on issues related to the Friendly AI problem, the problem with the greatest leverage in the universe. I have suggested that Less Wrong would make faster progress toward this goal if it worked more directly with the community of scholars already tackling the exact same problems. I don't personally work toward this goal because I'm not mathematically sophisticated enough to do so, but I'm glad others are!
Still, I think the first goal could be more explicitly pursued. There are many people like myself and jwhendy who can be massively impacted for the better not by coming to a realization about algorithmic learning theory, but by coming to understand the basics of rationality like probability and the proper role of belief and reductionism.
Reasons for Less Wrong to devote more energy to the basics
- Such efforts to spread the basics will have a short-term impact on more people than will efforts toward Friendly AI, and these impacted people will in turn impact others, hopefully for the better.
- Some LWers may feel they have little to contribute because they aren't masters of Solomonoff induction or algorithmic learning theory. But they will be able to contribute to raising the sanity waterline by spreading the basics of rationality.
- Providing more basic resources will attract a wider base of readers to Less Wrong, leading to (1) more new rationalists and (2) more donations to SIAI, for solving the Friendly AI problem.
- Even for experienced rationalists, it can be easy to forget the basics at times. Humans are not naturally rational, and revert to pre-rationality rather quickly without ongoing training and practice.
How to do it
Let me put some meat on this. What does more focus on the basics look like? Here are some ideas:
- The sequences are great, but some people are too busy or lazy to read even those. Some of the sequences could be summarized into single posts crafted so as to have no prerequisites. These posts could be linked widely, and entered in relevant blog carnivals.
- There is another huge community who will watch a 10-minute video, but will not read a short post. So the YouTube lectures are a great idea. But they could be improved. As of today, three of the videos show a presenter against a whiteboard. To make this work well requires lots of resources: (1) a good camera, (2) a shotgun or lavavlier microphone, (3) a teleprompter, and (4) an experienced and enthusiastic presenter. That's hard to do! But videos in the familiar PowerPoint style or the Khan Academy style are easier to do well. All it requires is some free presentation software, free screen capture software and a $70 high-quality USB mic like the Blue Snowball. This approach would also allow more people to participate in making videos on the basics of rationality.
- Sometimes, a basic concept of rationality will only "click" with somebody if presented in a certain way. Some will need a story that illustrates the concept. There are some of these on Less Wrong already, or in something like Harry Potter and the Methods of Rationality, but there could be more. For others, perhaps a Cartoon Guide to Bayes' Theorem or a Cartoon Guide to Reductionism would make it "click." Those who are more ambitious might attempt to create an animation explaining some core rationalist concept, ala this visualization of special relativity.
- Write "Introduction to X" or "How to Use X" posts.
- Keep developing the wiki, obviously.
- Develop a rationality workbook.
I like how pragmatic you're being. I am new here, but one of the things that attracted me to this site was the fact that much of the material is simply above my head. That's hard to find in informal public online communities outside of academia, and I feel that the very challenge of trying to wrap my head difficult material is an absolute necessity for keeping my math and statistics skills sharp. However, different people have different bars that they want to reach, and I do agree that more accessible material is a great idea. As for me, I have a voice for radio and a knack for stating difficult theories in an accessible way, so I think a good microphone will be my next purchase for my computer. Making a Youtube video or two on rationality would be a great way for me to contribute to this goal.
I know you can't see me because this is only a text comment, but I am right now really giving you a double thumbs up for this idea.
Here we see the author breaking a goal into small pieces, which tends to be an excellent move in many kinds of conversations. Nice post.
Thank you. I truly appreciate this.
I think that it would be far better if you combined the tutorials with spaced repetition decks.
I'm not sure this is in general a very good approach. Many people won't want to read something that is 50 pages long. But if one splits the same content up into 4 or 5 posts it will appear to be much more readable.
Perhaps, though many people prefer to scroll and search through a single page rather than having to jump across many pages. I use both formats on my site; I'm not sure which is better for my goals.
Do you have significant traffic? Can you do some sort of A/B test using some analytics software?
Multiple pages have the advantage of being easier to remember where one finished if one is forced to stop reading mid way.
This is one of the significant advantages of an ebook reader over a web browser (at least the current crop of both products). Firefox is supposed to keep my place in long web pages, but darned if it doesn't forget half the time.
This is the post that finally prompted me to make an account on the site; I've been lurking and reading for quite some time, but while I haven't felt (most of) the topics to be over my head, I also did not think my (sometimes sparse) background in the subjects gave me much to add.
In the past I've been struck by how useful some of the articles here tend to be, and the simultaneous regret that they would not be likely to reach the people they would assist the most. I think that this post provides an excellent breakdown of ways to reach more people more effectively, and I hope that in the near future I'll be able to contribute to that goal. (And in the spirit of naming specifics, I can hopefully provide some biology- and evolution-centered tutorials, and assuming I can keep my other creative writing projects moving as well, I might very well try my hand at the story approach. I do ADORE HPatMoR.)
I look forward to what you put together!
This is a great post, thanks! I want to vote it up 2^8 times.
Another good route for rationalist education is podcasts. To share some anecdata, I personally got into basic rationalism, and gave up my fairly vague and non-commital beliefs in ESP and various other forms of woo, after getting into The Skeptic's Guide to the Universe.
I think it would be amazing if there were a Less Wrong-ish parallel to such a podcast, that focuses less on traditional skeptical topics like anti-woo and atheism (since there are already plenty of podcasts about those), and more on formal rationalism and abstract topics like fun theory and Bayesian reasoning.
Also, seconded on the notion of Khan Academy style videos. Those are just amazing; the fact that they're drawn on the fly actually makes them easier to understand for me than pre-done PowerPoint-style slides. I think this is because I get a glimpse into the creator's thought processes as they explain, and also it normalizes the amount of information delivered per unit of time down to a nice even stream, instead of sudden jumps at slide transitions. It also has the advantage of also reducing viewer expectations for shiny graphics.
I love this idea, though unfortunately doing a podcast requires a fair amount of work. I have lots of experience in doing a podcast, but I am way overburdened as it is. I do plan, however, to be doing some interviews on these subjects for my own podcast, and in fact Eliezer is an upcoming guest.
This might be on-topic. In a recent discussion Emile asked what topics people would like to see more of on LessWrong. Comments with suggestions were voted up accordingly and the winner so far is statistics.
Thanks. I put in another vote for statistics. :)
This post is focused on a reasonable and discussion-worthy question: the question of whether to devote more time and energy to spreading the basics of rationality.
Still, I wish it were in the Discussion section rather than the main area, because it would be nice if the main area could contain a high ratio of direct discussion of how to be rational to meta discussion about the community.
I understand this, but want to add some comments/questions. I'm newer and am not exactly sure what differentiates top-level vs. discussion-area-appropriate posts. About only says this about the discussion area:
But the common understanding I find is that discussion = "meta" (perhaps as well as weaker/less-developed posts). Should the About section be clarified to reflect this? It seems that there are unofficially defined prescriptions floating around.
Would you clarify meta vs. non-meta. Is "meta" just concerned with suggestions about the LW site and the participants? If a post on raising the sanity waterline isn't meta, would this post, which suggests ways to do this, be considered meta? In other words, if Luke has presented some arguments for the "best rationally decided methods to help others become more rational at a basic level"... is that meta?
Lastly, for something like this topic which might imply action for those capable of writing content here and elsewhere to help noobs, I would consider the more experienced users to be the target audience. In other words, the post may be viewed as looking for teacher-level individuals to propagate LW content into several other formats in order to make rationality more accessible.
Given this, will a post like this receive adequate feedback/response from the "teacher-level" members if it is posted in the discussion area? If the simple answer is that most of those able to contribute to such an effort read the discussion area regularly, this is all the answer that is needed.
If that's not the case, however, could the discussion area be a black hole of sorts for a post like this?
To propose a possible solution for some of these points: define clear guidelines for top-level/discussion areas such that this post would have fallen under the discussion area definition. Then perhaps it could be moved to the top-level with enough voiced comments to do so?
My understanding is that Discussion is simply an area that can house a larger set of materials than can the main area of Less Wrong. It is in no way limited to meta-level discussions, but meta-level discussions are welcome there.
There's been a general request to keep meta-level discussions in the main area to a minimum, though not to zero. This request seems sensible to me. It would be nice to keep the main site full of posts that can actually help readers improve their rationality, with high signal to noise. And the Discussion area allows us to have most of the benefits from meta-level discussions without diluting the main site.
But I like the energy in this post and in some of the comments, and I might regret my original complaint in this case.
It would be nice to get some more explicit guidelines about appropriate content for the front page. In a way, meetup posts are 'meta' posts, and less substantial than what I've written here. But then, I understand what you mean about now wanting to clog the Posts section with meta-level posts. But maybe it's okay if it's in the Posts section but not promoted to the front page? Hard to say. Clearer guidelines would be nice...
Too late I guess; this post got promoted to the front page.
Still, here's me voicing my request for more explicit guidelines on this subject from the Less Wrong moderators.
Have any materialized since this post was made?
Thanks for the link -- that makes sense re. a newcomer learning a bit of the back story to LW content aims. I could also see the dividing line being as you put it: those directly focused on improving rationality vs. those about other things. Thanks for the reply.
My impression is that the discussion area has a higher percentage of regular readers, possibly only because it's more difficult to find.
I'm not clear on what you mean. You're saying that because it's more difficult to locate, readership regularity is increased? Do you think that "teacher-level" members fall into that class of regular readers?
I think what he meant was that those who read the discussion section tend to be higher level, because having to go beyond LW's front page screens out many of the more casual readers.
Ah - got it.
I'm attempting to work on a few short youtube videos based on Lesswrong posts, but it's my first time making such videos and I admit I'm a little intimidated by people who are plainly better. Not sure if I'm right for the job.
I wonder--what is it exactly you mean by "the basics"? I don't find Kolmogorov complexity more difficult than Bayes' theorem, but I found it absolutely essential to understanding the technical approach to Occam's Razor, which I found essential to understanding the advantages of Bayesian reasoning over traditional rationality.
If "the basics" means breaking down some of the technical points of Lesswrong posts, I agree wholeheartedly. I tried to take a friend through A Technical Explanation of Technical Explanation, my personal favorite, and found it was pretty difficult to explain to someone who didn't already know about some probability theory, expected value, and so on. Many potential Lesswrong contributors are intimidated by the background work they have to do to get into the site. You can see this in the RationalWiki discussion of this site; people find it absurd that you should have to read the entire sequences before questions can be answered. I'm not sure if writing even more will make it less intimidating, but I imagine that finding shorter (and still accurate) ways to say the same things will.
If "the basics" means ignoring some of the important technical contributions (ie Occam's Razor's technical formulation, the probabilistic structure of the Second Law of Thermodynamics, etc.), then I'm not sure I agree, partially because my deconversion story was a bit different from yours. Personally, I needed to see that many standard arguments for belief had precise refutations, almost completely ignored by any religious apologist working the scene. In my experience, people are more likely to change their beliefs when confronted with surprising arguments. Strong formalisms of arguments that people think apologists have already "defeated" are often the key.
One gets right for the job by starting off wrong for the job, and overcoming shortcomings.
I am sure you can ask questions along the way if you get stuck somewhere. If you read the sequences you can comment on the current post you are reading. Although you might be told to just read on ;-)
I am in the same boat as VirtualAdept. I've been following luke's blog forever which naturally lead me to LW, and whilst I'm still very in my Rational Infancy, I am hugely grateful for the gradual steps toward maturity that this site affords me, so I think I was due to register. Problem is it also makes me more and more frustrated at the lack of rationality surrounding me, especially as an exchristian with a very devout evangelical family and friend network. I mean I'm no Eliezer, but some of these people....
to have content that is accesible, interesting and easily sharable would be fantastic, because the type of people who need these lessons the most are the type of people who will give up at the slightest difficulty.
So my first step is attempting to be more active in this site, then start 'evangelizing' the more accesible parts on mediums like twitter and facebook. after that....the world!
but we'll see....
Mega kudos to Lukeprog and www.lesswrong.com!!
Off-topic but maybe of interest to some readers. While following the link to Luke's post on cognitive science I looked for more books by the author of the discussed book and saw that Jose Luis Bermudez wrote another one called Decision Theory and Rationality. It sounds very LWish:
That might be outdated given that Yudkowsky et al. are developing a new decision theory. But it might be a good introductory work? If someone could point me to other basic works on decision theory I'd be grateful.
Also, there is a professional review of the book by Lara Buchak (Rutgers) at NDPR.
BTW, that book is online if you know where to look.
Also, it helps to have someone with a good voice.
I think it may be useful to look at infrastructure changes that could promote these ends. Would it be worth having a video section (not sure if the karma value per upvote should be less or more than 10)? A basic and advanced section? It would be nice to have the site be useful for both beginning rationalists and people who are interested in Solomonoff induction without having them trip over each other.
The other benefit of having an explicit 'basic' section might be that restatements of other posts (ideally linked in the new post) could be encouraged- one basic educational fact is different people approach similar concepts in different ways, and so two posts with nearly identical content but different presentations can be rather valuable, and help things click for a larger group of people. It would be cool (but I don't know how easy it would be to make) to have a concept index, where I could type in "belief in belief" and get a list of all the basic posts explaining belief in belief, sorted by upvotes. (A less explicit method of doing this is the tag system we have already, but I doubt it should be used for this / would do this as well as it could. Actually, if the tags were specific to the basic section, that might work out well.)
Good thoughts. Would you do us a favor, thinking about this in some more detail, and write a discussion post?
Finally finished it.
The linked post looks well-written, but it is long for a top-level post. On the other hand, if many of these are being done, then it might be better to reduce the number of appearances in the feed rather than the length of individual posts.
For popularizing rationality (and hence raising the high water mark), online video is a solid choice of medium. In addition to the beat-it-into-the-ground simplicity of the Khan Academy lectures, there are some subjects where a less intensive, more conversational approach may help. People have been made to learn certain principles by rote (e.g. Bayes) but may have never grokked the full implications; I would love to see an animated visual in the style used by The RSA in Britain, fitted to a lecture on Bayes.
As creator of the power-point style video, I'd appreciate some help making future videos. Email me at firstname.lastname@example.org if interested.
Maybe not so obvious. The wiki is stagnating. Looking at the recent changes page, very few edits are being made and most the ones that are made seem to be on a small number of popular pages.
I'm not sure, but isn't taking the ideas of an author and giving it away for free in this way illegal?
No. Copying large portions of something directly would qualify as copyright infringement, summarizing in your own words and quoting small excerpts does not. Ideas cannot be copyrighted.
Only if he quotes large portions rather than summarizing them or rewriting them himself. Copyright applies to the presentation, not to the underlying ideas.
What jimrandomh and Kaj_Sotala said. :)