Three sequences on AI Safety were recently released. I'd suggest that in general it is probably better when new material isn't released that fast. Those sequences are quite technical so each post requires a reasonable amount of time to be invested. This means:
- There is more psychological pressure to read each post quickly rather than taking the time to understand and internalise the ideas
- There are less comments because keeping up with the posts reduces the time people have to write comments and if you're behind, by the time you catch up there's less value writing a comment as less people will see it
This isn't a particularly important post, just feedback for the future.
On the comments not getting engagement: I agree that the new posts might fall from frontpage soon. I want to point out they will be at the top of the AI Alignment Forum for quite a long time, and likely be pointed to as reference works commonly, so I expect the comments to be seen then a fair bit too.
Re: reference works. If LW worked a bit more like a mailing list, where reading a post caused you to "subscribe" to its comment thread by default, then people would feel less pressure to comment on things quickly after they were released so their comments would be read. The old LW 1.0 had a subscribe feature, but I'm not sure how many people used it. It was opt-in rather than opt-out.
Right now, my feeling is that checking LW and participating in discussions every day frazzles my brain in subtle ways. So I mostly don't try to keep up with posts day to d
... (read more)