I have also wanted for a way to curate comments, separate from posts.
I think a good next step to get more in contact with this, would be for someone to write a “highlights from the comments” post for the last month. See how good they are, how many there are, how much context is needed, and whether it’s a good reading experience. I might curate it if it’s a good reading experience.
As a counterpoint, I think the rise of shortform on LessWrong has partially led to more stuff being written and partially led to more stuff that would have been posts written in a way that is less searchable, less linkable (in practice), less polished, and generally less supportive of ongoing discourse.
I have definitely published fewer posts since the shortform feed started up on the front page.
The more more we raise profile of comments the more they’ll displace posts as the medium of communication.
can you operationalize this more so we can bet on it?
Certainly there’s a bunch more shortform since we’ve made room for shortform. The question is has this displaced posts, by the same people writing it? (as opposed to turned more random Constellation lunch-room hot takes into things that were public-at-all)
But btw, part of my proposal vague ideation here is that comments can have titles. Most comments don’t, by default, but if you end up writing an effortcomment you can give it a title. And, ideally, make them more natural to link to.
can you operationalize this more so we can bet on it?
Sure!
In the years before shortforms were on the front page, Eli published more LessWrong posts (on average) than the years when shortforms were on the front page.
This is an unfair operationalization, perhaps, but it does hew directly to the evidence that makes me believe the claim!
@Eli Tyre I’m not sure when you’re counting “the quick takes started changing things” but it seems like yourposting has basically been pretty constantly except for a brief more intense period in 2019/2020.
(you don’t seem to have any writing here before 2017 although I’m a bit surprised and not sure if you had a second account)
We did have the Popular Comments section. For some reason I didn’t like the popular comments section – it often was showing comments that were kind of tribal.
FYI, I haven’t actually gotten the UI to be that great yet, but, I’ve tried out version of the site that renders comments and posts together, clustered by time period:
In practice, atm, the “Today” section has a fair number of comments (because the posts haven’t been super upvoted yet). Then they mostly shift to posts (and if you click “load more”, some upvoted quick takes).
But, this might change over time, if people got basically as much opportunity to see and upvote comments as they did posts.
Just implement transclusion, like we do on Gwern.net. I do the equivalent of transcluding tweets all the time. (In fact, I would show you how I do transclude tweets on Gwern.net but it’s a bit tricky to dig up a ‘natural’ example, especially since Twitter long ago broke Nitter which was how we were getting clean snapshots to transclude, so they’re rare than they should be.)
You provide some selector options to govern how much of a target URL to transclude: see the docs for https://github.com/gwern/gwern.net/blob/master/js/transclude.js Thus, a user can transclude the ‘annotation’ version of a URL, which would include the header like “Ben Pace | May 12, 2026 2:33PM …”, or it could just transclude the ‘body’, ‘I have also wanted...good reading experience.’, or anything inside it which has IDs etc.
I have also wanted for a way to curate comments, separate from posts.
I think a good next step to get more in contact with this, would be for someone to write a “highlights from the comments” post for the last month. See how good they are, how many there are, how much context is needed, and whether it’s a good reading experience. I might curate it if it’s a good reading experience.
As a counterpoint, I think the rise of shortform on LessWrong has partially led to more stuff being written and partially led to more stuff that would have been posts written in a way that is less searchable, less linkable (in practice), less polished, and generally less supportive of ongoing discourse.
I have definitely published fewer posts since the shortform feed started up on the front page.
The more more we raise profile of comments the more they’ll displace posts as the medium of communication.
can you operationalize this more so we can bet on it?
Certainly there’s a bunch more shortform since we’ve made room for shortform. The question is has this displaced posts, by the same people writing it? (as opposed to turned more random Constellation lunch-room hot takes into things that were public-at-all)
But btw, part of my
proposalvague ideation here is that comments can have titles. Most comments don’t, by default, but if you end up writing an effortcomment you can give it a title. And, ideally, make them more natural to link to.Sure!
In the years before shortforms were on the front page, Eli published more LessWrong posts (on average) than the years when shortforms were on the front page.
This is an unfair operationalization, perhaps, but it does hew directly to the evidence that makes me believe the claim!
Well here is your posting history:
Year
Posts
Post wordcount
Comments
Comment wordcount
Shortform
Shortform Wordcount
Total wordcount
2017
0
0.0k
2
1.7k
0
0.0k
1.7k
2018
1
1.7k
35
0.4k
0
0.0k
2.1k
2019
12
2.2k
193
7.8k
25
4.1k
14.2k
2020
28
15.6k
147
11.1k
11
7.6k
34.3k
2021
5
3.4k
96
10.1k
14
2.7k
16.2k
2022
3
4.2k
114
14.5k
2
0.4k
19.2k
2023
3
14.8k
128
20.5k
0
0.0k
35.3k
2024
1
0.4k
227
27.0k
8
8.2k
35.6k
2025
0
0.0k
307
35.2k
9
5.9k
41.1k
2026
3
4.3k
234
25.9k
6
4.1k
34.3k
@Eli Tyre I’m not sure when you’re counting “the quick takes started changing things” but it seems like yourposting has basically been pretty constantly except for a brief more intense period in 2019/2020.
(you don’t seem to have any writing here before 2017 although I’m a bit surprised and not sure if you had a second account)
You guys started a new account for me when LessWrong 2.0 started.
I do kind of think we should be applying the frontpage/personal filter reliably to the quick takes section, so that it’s less news-y.
We did have the Popular Comments section. For some reason I didn’t like the popular comments section – it often was showing comments that were kind of tribal.
FYI, I haven’t actually gotten the UI to be that great yet, but, I’ve tried out version of the site that renders comments and posts together, clustered by time period:
https://baserates-test-git-comment-and-posts-together-lesswrong.vercel.app/all
In practice, atm, the “Today” section has a fair number of comments (because the posts haven’t been super upvoted yet). Then they mostly shift to posts (and if you click “load more”, some upvoted quick takes).
But, this might change over time, if people got basically as much opportunity to see and upvote comments as they did posts.
Just implement transclusion, like we do on Gwern.net. I do the equivalent of transcluding tweets all the time. (In fact, I would show you how I do transclude tweets on Gwern.net but it’s a bit tricky to dig up a ‘natural’ example, especially since Twitter long ago broke Nitter which was how we were getting clean snapshots to transclude, so they’re rare than they should be.)
You provide some selector options to govern how much of a target URL to transclude: see the docs for https://github.com/gwern/gwern.net/blob/master/js/transclude.js Thus, a user can transclude the ‘annotation’ version of a URL, which would include the header like “Ben Pace | May 12, 2026 2:33PM …”, or it could just transclude the ‘body’, ‘I have also wanted...good reading experience.’, or anything inside it which has IDs etc.