True[1] but unhelpful, like “your party lost the election because people didn’t vote for it” and “this species was evolutionarily successful because organisms of this species produced plenty of descendants-of-descendants-of-descendants”. I’m not very convinced by MikkW’s list of possible issues, but at least it makes some attempt to engage with why readers didn’t find the post valuable.
[1] Or at least true-ish. Some posts might get downvoted by, e.g., politically motivated people who downvote everything that looks like it disagrees with their politics, and a sufficiently determined group of highly-politicized downvoters could wipe out the effect of upvotes from people who found a post helpful. (So far as I know, LW doesn’t currently suffer from this specific problem. There might be others of similar shape. I am not claiming that they actually are, just adding a necessary qualification.)
but at least it makes some attempt to engage with why readers didn’t find the post valuable.
I don’t think so. To me, his list sounds like trying to understand reasons why the post is judged negatively.
People get trained in school and university to write without any considerations for whether or not their writing is valuable and get graded in a way that’s independent of the value provided by an essay. The essay gets read by the teacher because the teacher is paid to read it and not because it provides them with other value.
On the other hand, when writing posts on LessWrong it matters whether or not the reader gets value from reading it.
Larry McEnerney does a good job at laying out the difference:
I’m not very convinced by MikkW’s list of possible issues, but at least it makes some attempt to engage with why readers didn’t find the post valuable.
I would be interested to hear if there are any issues with the «Army of Jakoths» post that I didn’t identify here
A few that come to mind. I’m describing rather than endorsing here but these are all issues that I think it would be at least reasonable for a reader to have.
“It’s just not very well written”. A reader might have no problem with poetic or pseudopoetic style but might think you haven’t done a good job. (Cf. Viliam’s comment.)
“Specifically, it’s trying to look like poetry without in any useful sense being poetry”. (Cf. Viliam’s comment, again.)
“It’s making an analogy but the analogy isn’t actually a good match”. (E.g., because, as I pointed out in a comment, it’s not so hard to tell whether there’s an actual invading army that’s about to loot your home and massacre your family. Or because it conflates hostility with indifference and one might reasonably feel differently about someone who hates you and wants you dead, versus someone/something that has merely noticed that “you are used of atoms it can use for something else”.)
“The foolish neighbour is a straw man”. A reader might consider that AI doom naysayers typically have less silly things to say than just “it’s absurdly improbable”. (I have less sympathy for this one than for the others, because I do frequently hear people dismissing AI doom on grounds I can’t distinguish from “this obviously seems silly to me”.)
“It’s incorrectly written”. (Complaining about punctuation and grammar.)
A general pattern here: you listed a number of issues and it seems like you conspicuously avoided ones of the form “readers found the post to be of low quality”, as opposed to “readers had an irrational dislike to one of the reasonable stylistic choices made in writing the post”.
True[1] but unhelpful, like “your party lost the election because people didn’t vote for it” and “this species was evolutionarily successful because organisms of this species produced plenty of descendants-of-descendants-of-descendants”. I’m not very convinced by MikkW’s list of possible issues, but at least it makes some attempt to engage with why readers didn’t find the post valuable.
[1] Or at least true-ish. Some posts might get downvoted by, e.g., politically motivated people who downvote everything that looks like it disagrees with their politics, and a sufficiently determined group of highly-politicized downvoters could wipe out the effect of upvotes from people who found a post helpful. (So far as I know, LW doesn’t currently suffer from this specific problem. There might be others of similar shape. I am not claiming that they actually are, just adding a necessary qualification.)
I don’t think so. To me, his list sounds like trying to understand reasons why the post is judged negatively.
People get trained in school and university to write without any considerations for whether or not their writing is valuable and get graded in a way that’s independent of the value provided by an essay. The essay gets read by the teacher because the teacher is paid to read it and not because it provides them with other value.
On the other hand, when writing posts on LessWrong it matters whether or not the reader gets value from reading it.
Larry McEnerney does a good job at laying out the difference:
I would be interested to hear if there are any issues with the «Army of Jakoths» post that I didn’t identify here
A few that come to mind. I’m describing rather than endorsing here but these are all issues that I think it would be at least reasonable for a reader to have.
“It’s just not very well written”. A reader might have no problem with poetic or pseudopoetic style but might think you haven’t done a good job. (Cf. Viliam’s comment.)
“Specifically, it’s trying to look like poetry without in any useful sense being poetry”. (Cf. Viliam’s comment, again.)
“It’s making an analogy but the analogy isn’t actually a good match”. (E.g., because, as I pointed out in a comment, it’s not so hard to tell whether there’s an actual invading army that’s about to loot your home and massacre your family. Or because it conflates hostility with indifference and one might reasonably feel differently about someone who hates you and wants you dead, versus someone/something that has merely noticed that “you are used of atoms it can use for something else”.)
“The foolish neighbour is a straw man”. A reader might consider that AI doom naysayers typically have less silly things to say than just “it’s absurdly improbable”. (I have less sympathy for this one than for the others, because I do frequently hear people dismissing AI doom on grounds I can’t distinguish from “this obviously seems silly to me”.)
“It’s incorrectly written”. (Complaining about punctuation and grammar.)
A general pattern here: you listed a number of issues and it seems like you conspicuously avoided ones of the form “readers found the post to be of low quality”, as opposed to “readers had an irrational dislike to one of the reasonable stylistic choices made in writing the post”.