I think the main downside would be that some people might read only the summary and miss out on the reasoning parts, which are often the parts that are actually the most valuable. Or, relatedly, some authors like Scott Alexander do better when people have to read his hook and get sucked into his discursive style.
However, overall I’ve been sold on this idea, but do think that the UI would need to be done well. We haven’t prioritized it yet, but I’m definitely tempted, and I think that’s true for other team members as well.
fwiw, I think a good summary can & should also summarise the reasoning, not only the conclusions. (Though of course it’ll still miss many details of both the reasoning and the conclusions.)
I’d also flag that I think this should be some sort of nudge, suggestion, optional box to fill in, or whatever, rather than a required box. So people could still decide to not have a summary if they wanted, e.g. if that would mess with their engaging discursive style in a given piece.
I think the main downside would be that some people might read only the summary and miss out on the reasoning parts, which are often the parts that are actually the most valuable. Or, relatedly, some authors like Scott Alexander do better when people have to read his hook and get sucked into his discursive style.
However, overall I’ve been sold on this idea, but do think that the UI would need to be done well. We haven’t prioritized it yet, but I’m definitely tempted, and I think that’s true for other team members as well.
Cool, glad to hear that.
fwiw, I think a good summary can & should also summarise the reasoning, not only the conclusions. (Though of course it’ll still miss many details of both the reasoning and the conclusions.)
I’d also flag that I think this should be some sort of nudge, suggestion, optional box to fill in, or whatever, rather than a required box. So people could still decide to not have a summary if they wanted, e.g. if that would mess with their engaging discursive style in a given piece.