lukeprog recently wrote a top-level post exhorting LW to emphasize the basics of rationality. That strikes me as a worthwhile endeavor- if we do actually want to raise the sanity waterline, we need a lot of droplets.

The suggestion I made was to redesign the site to create more specialized areas: a basic one where people aren't expected to have read the sequences, and an advanced one where people are. That's not the division that has to be made, but it seems a natural place to make one. For ease of reading, a page that displays both sections seems like it would be a good plan (but I have no idea if that is easy to make), and it might also be desirable to have it so one article can appear in both sections (maybe there would flags or tags for basic and advanced?). Another option for a division is between video and text posts- I personally have little interest in video posts but I'm sure there are people interested in both or just video posts.

The basic section would focus on fundamental concepts and standalone / modular posts. By modular, I mean replaceable- two posts that cover the same concept from different angles, ideally in enough detail that you could read either of them and be able to go on to the next post, if there is one. There are several (mutually inclusive) ways to go about this- video explanations of old posts, entirely new material, and new restatements (or perspectives) on old material. lukeprog covers videos, so I won't; new material seems pretty self-explanatory.

The main possible focus of such a section that needs discussion is new perspectives on old concepts. Right now, one of the failures of LW according to the outside view is the sense of cultishness, and I would not be surprised if that comes from the fact that anyone who maintains their desire to read the sequences throughout the entire sequences must be in some sense on the same wavelength as EY. If we had five different explanations of Politics is the Mind-Killler rather than just EY's, it seems plausible we could reach more people. If there were more approaches to metaethics than EY's, that might convince a few critics to stick around for other things they do like, and so on.

While plausible, it's not certain. Right now, the sequences are mostly pretty good. If we invite other people to present their core concepts, the average quality will probably decline, even if the maximum quality for each individual reader will increase (as new options are either better or not). A basic educational fact is that people react differently to different approaches, and what is interesting and straightforward for one person might be boring and opaque to another. The question becomes then an empirical one about max quality minus search cost (will you read enough of the five different explanations to see which one is best for you? How long would that take? How much better are alternatives?) and an ideological one about average quality vs. maximum quality.

Warning: the ideological question can get ugly. It shares many elements with questions we've seen before- "what percentage of LW's target audience have we reached?"- but focuses on the word "target." One quickly notices that "average quality" doesn't average over all readers or potential readers, but just the readers that stick around. It's trivial to have a high average when you can pick cherries. Do we want this to be a place full of thinkers who, even if they disagree with EY, like him / the community enough to stick around despite many features that suggest it's an echo chamber? Or do we want this to be a place that grows beyond its roots and branches into many things, too large for one person to dominate? Note that smaller and larger groups each have their own benefits, and it's not clear to me which apply to the situation at hand.

Another sticking point in this idea is defining the 'advanced' section. It's one thing to say "ok, everyone who reads the sequences graduates to the advanced section," it's another thing to identify topics that only make sense once you've read the sequences. Going back through the list of the last 30 posts to the main LW page, I had trouble identifying which I would put where (except for a generic "get your AI out of my rationality" urge). My more reasonable urge was to advertise the summary break, to the right of the block quote button and to the left of the bullet points button. If you use that, only the part of your post above the summary break gets posted to the new list (and everything after requires a clickthrough).

Thoughts, on either the basic/advanced division or the more benign video section suggestion?

New to LessWrong?

New Comment
3 comments, sorted by Click to highlight new comments since: Today at 8:24 AM

Echoing wedrifid: We don't have enough content to justify any sort of further division.

I also think whether or not people have read the sequences is a poor line to divide the site along.

If it wasn't so ironic I think I would make a post "Less Meta, more direct content!" :P

[-][anonymous]13y00

We have that already: Less Meta.