I don’t think Dagon saying not “don’t tell stories” but “if you want to make an argument by telling stories, please at least tell us what the argument is meant to be so that we can evaluate it with system 2 as well as system 1″.
Meditations on Moloch didn’t just quote Ginsberg and say “Lo!”, it explained what Scott was calling Moloch and why, and gave explicit concrete examples.
I think it’s entirely possible that Sailor Vulcan has something interesting and/or important to say here, but at least from my perspective (as I think from Dagon’s) it still needs actually saying rather than merely gesturing towards. Until we have at least some of those “nitty gritty details” we’re promised later, it’s hard to tell what bits of the story are intended more or less literally, what bits are intended as metaphors for other things, and what bits are mere window-dressing.
Indeed—I perhaps should have been more explicit when I wrote “STARTING WITH a story...” Stories are important in presenting, reifying, and exploring details of a theory. Stories are an important part of the way humans discover truth about the world and each other (and they’re often entertaining). I heartily encourage storytelling!
But on this site specifically, we aspire to be mostly about truth-seeking. Stories don’t belong here without a pretty clear tie (preferably footnotes and links) to the theory and data you’re trying to illustrate.
But on this site specifically, we aspire to be mostly about truth-seeking. Stories don’t belong here without a pretty clear tie (preferably footnotes and links) to the theory and data you’re trying to illustrate.
This is the part that worries me. I think you should be able to start with a story (like meditations on Moloch) and only after get the theory (like inadequate equilibria).
At the top of Sailor Vulcan’s post he specifically talked about wanting to post the theory later. I do push against a culture that says the order in which you have to start with is the data then the story. I know this goes against concepts here like “Dark Arts” but I think a lot of that stuff is actually just wrong for truth seeking communities.
Medidations on Moloch and the stories in Inadequate Equilibria (and HPMOR, and Luminosity and Friendship is Optimal and pretty much all popular rationalist-centric stories) were written in a way that referenced and built on a whole lot of explicit theory.
And you’re right—I don’t want to mandate a sequence of publication. Everyone should do what works, and there are probably some story/theory pairs where it works to publish the story first. I can’t think of any, and I’d advise having the theory post ready if it turns out it’s needed sooner than you thought, but I won’t say “never”. This didn’t work for me. I think because I have pretty serious reservations about the theory (both whether it’s appropriate here and about the theory itself), but I can’t know whether those concerns are valid or not, as the story is fairly inexplicit.
I think story-first runs the very large risk that people will infer a theory different than you intend, and then downvote you for that (flawed interpretation of your) theory. I may be doing exactly this. It also runs the risk that if the theory has holes, people will retroactively feel tricked by the misleading story, and be angry at your presentation style in addition to respectfully disagreeing with your theory.
I think story-first runs the very large risk that people will infer a theory different than you intend, and then downvote you for that (flawed interpretation of your) theory. I may be doing exactly this. It also runs the risk that if the theory has holes, people will retroactively feel tricked by the misleading story, and be angry at your presentation style in addition to respectfully disagreeing with your theory
This seems like an important pitfall of the strategy and something people should be aware of. I agree it’s a problem especially in cases like this where the model is a bit opaque. I suspect that part of this is because the story was written for a different context than being a LW parable and therefore was less on the nose than lwers might like.
I think we mostly don’t disagree, I was mostly worried about people taking the critique and updating their general rules for what they should feel comfortable posting and in what order.
I don’t think Dagon saying not “don’t tell stories” but “if you want to make an argument by telling stories, please at least tell us what the argument is meant to be so that we can evaluate it with system 2 as well as system 1″.
Meditations on Moloch didn’t just quote Ginsberg and say “Lo!”, it explained what Scott was calling Moloch and why, and gave explicit concrete examples.
I think it’s entirely possible that Sailor Vulcan has something interesting and/or important to say here, but at least from my perspective (as I think from Dagon’s) it still needs actually saying rather than merely gesturing towards. Until we have at least some of those “nitty gritty details” we’re promised later, it’s hard to tell what bits of the story are intended more or less literally, what bits are intended as metaphors for other things, and what bits are mere window-dressing.
Indeed—I perhaps should have been more explicit when I wrote “STARTING WITH a story...” Stories are important in presenting, reifying, and exploring details of a theory. Stories are an important part of the way humans discover truth about the world and each other (and they’re often entertaining). I heartily encourage storytelling!
But on this site specifically, we aspire to be mostly about truth-seeking. Stories don’t belong here without a pretty clear tie (preferably footnotes and links) to the theory and data you’re trying to illustrate.
This is the part that worries me. I think you should be able to start with a story (like meditations on Moloch) and only after get the theory (like inadequate equilibria).
At the top of Sailor Vulcan’s post he specifically talked about wanting to post the theory later. I do push against a culture that says the order in which you have to start with is the data then the story. I know this goes against concepts here like “Dark Arts” but I think a lot of that stuff is actually just wrong for truth seeking communities.
Medidations on Moloch and the stories in Inadequate Equilibria (and HPMOR, and Luminosity and Friendship is Optimal and pretty much all popular rationalist-centric stories) were written in a way that referenced and built on a whole lot of explicit theory.
And you’re right—I don’t want to mandate a sequence of publication. Everyone should do what works, and there are probably some story/theory pairs where it works to publish the story first. I can’t think of any, and I’d advise having the theory post ready if it turns out it’s needed sooner than you thought, but I won’t say “never”. This didn’t work for me. I think because I have pretty serious reservations about the theory (both whether it’s appropriate here and about the theory itself), but I can’t know whether those concerns are valid or not, as the story is fairly inexplicit.
I think story-first runs the very large risk that people will infer a theory different than you intend, and then downvote you for that (flawed interpretation of your) theory. I may be doing exactly this. It also runs the risk that if the theory has holes, people will retroactively feel tricked by the misleading story, and be angry at your presentation style in addition to respectfully disagreeing with your theory.
This seems like an important pitfall of the strategy and something people should be aware of. I agree it’s a problem especially in cases like this where the model is a bit opaque. I suspect that part of this is because the story was written for a different context than being a LW parable and therefore was less on the nose than lwers might like.
I think we mostly don’t disagree, I was mostly worried about people taking the critique and updating their general rules for what they should feel comfortable posting and in what order.