Intellectual insularity is because we don’t en masse read other sources. So we can’t discuss them. Sure, good books are mentioned in the post, but that didn’t create a collective action. What could?
Proposal: At the beginning of the month, let’s choose and announce a “book of the month”. At the end of the month, we will discuss the book. (During the month, discussing the book should probably be forbidden, to avoid spoilers and discouraging people who haven’t read it yet.)
Have we grown as a website? I don’t know—what metric do you use? I guess the number of members / comments / articles is growing, but that’s not exactly what we want. So, what exactly do we want? First step could be to specify the goal. Maybe it could be the articles—we could try to create more high-quality articles that would be very relevant to science and rationality, but also accessible for a random visitor. Seems like the “Main” part of the site is here for this goal, except that it also contains things like “Meetups” and “Rationality Quotes”.
Proposal: Refactor LW into more categories. I am not sure how exactly, but the current “Main” and “Discussion” categories feel rather unnatural. (Are they supposed to simply mean: higher importance / lower importance?) A quick idea: Announcements for information about SIAI and upcoming meetups; Forum for repeating topics (open discussion, rationality quotes, media thread, group diary); Top Articles for high-voted articles, and Articles for the remaining articles. In this view, our metric could be to have enough “Top Articles”, though of course having more meetups is also great.
Also, why are Eliezer’s articles so good? He chose one topic and gradually developed it. It was not “hit and run” blogging, but more like teaching lessons at school. Only later, another topic. That’s why his articles literally make sequences; most other articles don’t.
Proposal: We could choose one topic to educate other people about, such as mathematics or statistics or programming, and write a series of articles on this topic. (This can be also done by one person.) It is important to have more articles in sequence, a smooth learning curve, so they don’t overwhelm the layman immediately.
The common factor to all three proposals is: some coordinated action is necessary. When LW was Eliezer’s blog, he did not need to coordinate with himself, but he was making some strategic decisions. To continue LW less chaotically, we would need either a “second Eliezer” (for example Luke wrote a sequence), or a method to make group decisions. Group coordination is generally a difficult problem—it can be done, but we shouldn’t expect it to happen automatically. (One possible solution could be to pay someone to write another sequence.)
Upvoted. I think that refactoring LW is a strong move, but it’s also one which has been discussed for a while and hasn’t happened. I think that’s because there’s never been a well-presented case for new sections, but the site admins are the ones to talk to about that.
Proposal: At the beginning of the month, let’s choose and announce a “book of the month”. At the end of the month, we will discuss the book.
I like this idea but it seems like it’s on the wrong side of the 80⁄20 value/effort split. badger’s summary of EPHJ is one twentieth of the length of the book it summarizes, but contains at least half of the value one gets from reading that book.
Kaufman’s Personal MBA comes to mind as another thing to model off of. He’s read hundreds of business books, and has distilled them down to create a mostly complete business education in 400 pages. The book reads like the blog- an explanation of a part in a few pages, and then on to the next part, with the parts fitting together to make a lean system.
Perhaps a summary contest? Identify some book as a valuable addition to LW, and announce a contest with a prize and deadline for posts that summarize the book or possibly posts that turn the book into a sequence. (The candidate posts might get their own section, with the best one or a hybrid of the best ones being pushed to main, so that people don’t have to see three or four of the same thing if they don’t want to.)
Why not post a list of such valuable or potentially valuable books and see if anyone has already read them and is willing to do a quick skim and summarise?
I should probably add that I’m opposed to the idea of a summary contest because it will cost a relatively large number of people a lot of time and gain them very little.
Summaries aren’t too useful. On the other hand, commentaries and in-depth discussion might be useful. For example, I’ve occasionally thought of doing a chapter by chapter discussion of Good and Real, with additional material like a Haskell implementation of his Quantish universe (since I don’t really understand it).
I should probably add that I’m opposed to the idea of a summary contest because it will cost a relatively large number of people a lot of time and gain them very little.
Mmm. Active reading of quality books is its own reward- the prize is for sharing the notes, and to raise the option to attention. It seems fine compared to a book club, but I agree that it’s generally an economic model that favors the buyer over the producers.
I think we could rewrite Eliezer’s articles. I would disagree with the statement that they are “so good”. The material is great of course, but the way he goes about conveying it is not for everyone. I can’t really see a whole cohesive structure as I am going through and frequently, I am not so sure what point he is making. His use of parable just obfuscates the point for me; his constant referral to his story “The Simple Truth” in Map and Territory really bothered me because that story was difficult for me to get through and I just wanted to see his point in plain text. I still have trouble organizing LW material into an easy-to-think-about structure. What I am looking for is something more resembling a textbook. Very structured, somewhat dry writing (yes, I actually prefer that), maybe some diagrams. I’d do it, but I am not sure I have a strong enough understanding of the material to do so.
I think this is a great analysis, and I like the specific proposals. I got involved in LW only about a year ago, and while I read through pretty much all of the Sequences, I felt a bit left out that I couldn’t participate while they were originally unfolding. A “book of the month” program, or else some kind of coordinated discussion of specific topics, could go a long way toward allowing that kind of ongoing participation.
I also really like the proposed re-categorization of posts. I’m never quite sure what’s supposed to go where, and it seems like a lot of the most important stuff (like this post) end up in Discussion. To state the problem more generally, there seems to be a natural divide in thing-space between “procedural posts” (announcements, meet-ups, quotes, etc.) and “substantive posts” (basically everything else). But we presently group “high-level” substantive posts along with procedural posts, rather than with other substantive posts, which seems awkward. Seems better to first make the basic distinction between procedure and substance, and then find a way to identify the high-level substantive posts within that category.
On the subject of coordinated action, who actually has the authority to make this kind of change, and what sort of process would that entail? I don’t really know much about the current LW org chart, but these sorts of concerns seem to be coming up with increasing frequency, so I’d like to figure out how we can actually do something about it.
Intellectual insularity is because we don’t en masse read other sources. So we can’t discuss them. Sure, good books are mentioned in the post, but that didn’t create a collective action. What could?
Proposal: At the beginning of the month, let’s choose and announce a “book of the month”. At the end of the month, we will discuss the book. (During the month, discussing the book should probably be forbidden, to avoid spoilers and discouraging people who haven’t read it yet.)
Have we grown as a website? I don’t know—what metric do you use? I guess the number of members / comments / articles is growing, but that’s not exactly what we want. So, what exactly do we want? First step could be to specify the goal. Maybe it could be the articles—we could try to create more high-quality articles that would be very relevant to science and rationality, but also accessible for a random visitor. Seems like the “Main” part of the site is here for this goal, except that it also contains things like “Meetups” and “Rationality Quotes”.
Proposal: Refactor LW into more categories. I am not sure how exactly, but the current “Main” and “Discussion” categories feel rather unnatural. (Are they supposed to simply mean: higher importance / lower importance?) A quick idea: Announcements for information about SIAI and upcoming meetups; Forum for repeating topics (open discussion, rationality quotes, media thread, group diary); Top Articles for high-voted articles, and Articles for the remaining articles. In this view, our metric could be to have enough “Top Articles”, though of course having more meetups is also great.
Also, why are Eliezer’s articles so good? He chose one topic and gradually developed it. It was not “hit and run” blogging, but more like teaching lessons at school. Only later, another topic. That’s why his articles literally make sequences; most other articles don’t.
Proposal: We could choose one topic to educate other people about, such as mathematics or statistics or programming, and write a series of articles on this topic. (This can be also done by one person.) It is important to have more articles in sequence, a smooth learning curve, so they don’t overwhelm the layman immediately.
The common factor to all three proposals is: some coordinated action is necessary. When LW was Eliezer’s blog, he did not need to coordinate with himself, but he was making some strategic decisions. To continue LW less chaotically, we would need either a “second Eliezer” (for example Luke wrote a sequence), or a method to make group decisions. Group coordination is generally a difficult problem—it can be done, but we shouldn’t expect it to happen automatically. (One possible solution could be to pay someone to write another sequence.)
Upvoted. I think that refactoring LW is a strong move, but it’s also one which has been discussed for a while and hasn’t happened. I think that’s because there’s never been a well-presented case for new sections, but the site admins are the ones to talk to about that.
I like this idea but it seems like it’s on the wrong side of the 80⁄20 value/effort split. badger’s summary of EPHJ is one twentieth of the length of the book it summarizes, but contains at least half of the value one gets from reading that book.
Kaufman’s Personal MBA comes to mind as another thing to model off of. He’s read hundreds of business books, and has distilled them down to create a mostly complete business education in 400 pages. The book reads like the blog- an explanation of a part in a few pages, and then on to the next part, with the parts fitting together to make a lean system.
Perhaps a summary contest? Identify some book as a valuable addition to LW, and announce a contest with a prize and deadline for posts that summarize the book or possibly posts that turn the book into a sequence. (The candidate posts might get their own section, with the best one or a hybrid of the best ones being pushed to main, so that people don’t have to see three or four of the same thing if they don’t want to.)
Why not post a list of such valuable or potentially valuable books and see if anyone has already read them and is willing to do a quick skim and summarise?
I should probably add that I’m opposed to the idea of a summary contest because it will cost a relatively large number of people a lot of time and gain them very little.
Summaries aren’t too useful. On the other hand, commentaries and in-depth discussion might be useful. For example, I’ve occasionally thought of doing a chapter by chapter discussion of Good and Real, with additional material like a Haskell implementation of his Quantish universe (since I don’t really understand it).
Please do this. I’m finding it impenetrable.
Mmm. Active reading of quality books is its own reward- the prize is for sharing the notes, and to raise the option to attention. It seems fine compared to a book club, but I agree that it’s generally an economic model that favors the buyer over the producers.
Unless you’re talking about fiction I’m not sure why spoilers matter. Better to encourage people to discuss parts they don’t understand.
I think we could rewrite Eliezer’s articles. I would disagree with the statement that they are “so good”. The material is great of course, but the way he goes about conveying it is not for everyone. I can’t really see a whole cohesive structure as I am going through and frequently, I am not so sure what point he is making. His use of parable just obfuscates the point for me; his constant referral to his story “The Simple Truth” in Map and Territory really bothered me because that story was difficult for me to get through and I just wanted to see his point in plain text. I still have trouble organizing LW material into an easy-to-think-about structure. What I am looking for is something more resembling a textbook. Very structured, somewhat dry writing (yes, I actually prefer that), maybe some diagrams. I’d do it, but I am not sure I have a strong enough understanding of the material to do so.
Isn’t that precisely the end goal of SIAI?
(#EliezerYudkowskyFacts)
I think this is a great analysis, and I like the specific proposals. I got involved in LW only about a year ago, and while I read through pretty much all of the Sequences, I felt a bit left out that I couldn’t participate while they were originally unfolding. A “book of the month” program, or else some kind of coordinated discussion of specific topics, could go a long way toward allowing that kind of ongoing participation.
I also really like the proposed re-categorization of posts. I’m never quite sure what’s supposed to go where, and it seems like a lot of the most important stuff (like this post) end up in Discussion. To state the problem more generally, there seems to be a natural divide in thing-space between “procedural posts” (announcements, meet-ups, quotes, etc.) and “substantive posts” (basically everything else). But we presently group “high-level” substantive posts along with procedural posts, rather than with other substantive posts, which seems awkward. Seems better to first make the basic distinction between procedure and substance, and then find a way to identify the high-level substantive posts within that category.
On the subject of coordinated action, who actually has the authority to make this kind of change, and what sort of process would that entail? I don’t really know much about the current LW org chart, but these sorts of concerns seem to be coming up with increasing frequency, so I’d like to figure out how we can actually do something about it.