I disagree. The problem is not that we are promoting people starting with the old classics, which tend to contain more core insights than whatever happens to have been posted most recently. The problem is that there is no way to collapse it.
You’re right that attention incentivises content creation, but we are also custodians of this attention and we want to direct it towards whatever will most benefit our audience. We shouldn’t just blindly follow the corporate model. Corporations want to maximise activity by getting people commenting on the new stuff, rather than maximise utility.
Besides, we already get a steady flow of submissions.
FYI, we’re planning to make it so logged in users see a much shorter section, which is something like “sequences you are currently reading or maybe should read”, which is more like 1.5 inches tall rather than a full page. I think this will solve most of the problem.
Our current mindset is something like “be more like youtube”, in the sense that new content is very visible, but a lot of attention is directed in various ways to the oldest, best content.
This makes sense and I think I know what you mean now, but it wasn’t as obvious in the original post.
The underlying goal Habryka is working towards is incentivizingthe building of important intellectual knowledge, which can be used concretely in the external world. This is similar to “improve quality of posts”, but subtly different. The four sub-things that need incentivizing are:
Newcomers reading and learning foundational build blocks.
Veterans coming to check on important new updates to the frontiers of our knowledge that are relevant to them.
People contributing posts and sequences that build on that knowledge, that we’ll want to refer back to and build on yet-more in the coming years.
People who either improving at thinking and writing, or refining ideas together, until they produce #2.
I probably have more to say, but for just noting that the frontpage has multiple jobs it’s trying to accomplish. It could probably improve at all of them. I read your post as saying (at least when translated-into-my-models) something like “Our front page incentivizes #1 at the expense of #4, and #4 is a necessary step to get to #3, which is necessary to get #2.”
If this seems radically different than the underlying goals you were pointing at, apologies, and could you explain differently?
There’s a small caveat though. You say the current front page incentivizes #1, but actually it just recommends people to do #1, there’s no incentivizing. On the other hand, if you had a front page giving tasty attention to the best attempts to build on the sequences, that might incentivize all of #1-#4 better than our current front page. And focusing discussion on a few posts would make comments get more votes on average, increasing incentives for commenters as well. I think pre-2011 LW got all of those right, except the curation for Main wasn’t responsive enough.
Agreed that incentivizing isn’t the right word for what’s happening with #1, but there’s a related thing (maybe a superset that includes incentivizing, in some frames?) that also includes “nudges”, or “making certain things convenient” which seems like an important piece of the puzzle.
I disagree. The problem is not that we are promoting people starting with the old classics, which tend to contain more core insights than whatever happens to have been posted most recently. The problem is that there is no way to collapse it.
You’re right that attention incentivises content creation, but we are also custodians of this attention and we want to direct it towards whatever will most benefit our audience. We shouldn’t just blindly follow the corporate model. Corporations want to maximise activity by getting people commenting on the new stuff, rather than maximise utility.
Besides, we already get a steady flow of submissions.
This framing of the problem seems right to me.
FYI, we’re planning to make it so logged in users see a much shorter section, which is something like “sequences you are currently reading or maybe should read”, which is more like 1.5 inches tall rather than a full page. I think this will solve most of the problem.
Our current mindset is something like “be more like youtube”, in the sense that new content is very visible, but a lot of attention is directed in various ways to the oldest, best content.
Yes! It’s a good thing my proposal is trying to improve quality of posts, not maximize activity.
This makes sense and I think I know what you mean now, but it wasn’t as obvious in the original post.
The underlying goal Habryka is working towards is incentivizing the building of important intellectual knowledge, which can be used concretely in the external world. This is similar to “improve quality of posts”, but subtly different. The four sub-things that need incentivizing are:
Newcomers reading and learning foundational build blocks.
Veterans coming to check on important new updates to the frontiers of our knowledge that are relevant to them.
People contributing posts and sequences that build on that knowledge, that we’ll want to refer back to and build on yet-more in the coming years.
People who either improving at thinking and writing, or refining ideas together, until they produce #2.
I probably have more to say, but for just noting that the frontpage has multiple jobs it’s trying to accomplish. It could probably improve at all of them. I read your post as saying (at least when translated-into-my-models) something like “Our front page incentivizes #1 at the expense of #4, and #4 is a necessary step to get to #3, which is necessary to get #2.”
If this seems radically different than the underlying goals you were pointing at, apologies, and could you explain differently?
Yes, that’s a great explanation.
There’s a small caveat though. You say the current front page incentivizes #1, but actually it just recommends people to do #1, there’s no incentivizing. On the other hand, if you had a front page giving tasty attention to the best attempts to build on the sequences, that might incentivize all of #1-#4 better than our current front page. And focusing discussion on a few posts would make comments get more votes on average, increasing incentives for commenters as well. I think pre-2011 LW got all of those right, except the curation for Main wasn’t responsive enough.
Agreed that incentivizing isn’t the right word for what’s happening with #1, but there’s a related thing (maybe a superset that includes incentivizing, in some frames?) that also includes “nudges”, or “making certain things convenient” which seems like an important piece of the puzzle.