I claim you are using semantic and formatting emphasis in a way that can be expected to increase rate of posts bypassing epistemic immune systems. More concisely, it reads like marketing slop. Maybe if you just want to manipulate public opinion this is fine for your purposes, though I doubt you’d want that if you thought about it enough—I expect that you’re getting backfires like me from it and that you’re better off with a mildly lower impact rate in exchange for a lower backfire rate. In this case, I think I already agreed with your post, but your writing didn’t move my opinion significantly and excessive candying of the words is part of why.
if you tell me it won’t be annoying to you for me to do so, I could leave many reacts on your post annotating the parts that lead me to this view; they’ll primarily be “citation needed”, “weak argument”, “soldier mindset”. Otherwise I’ll consider this to be the extent of my feedback and continue downvoting posts which feature-parse to me as attempting to be manipulative in this way.
To downvote based on this seems like clearly going downwards on Graham’s hierarchy of refutation.
You don’t refute the central point, identify a crucial mistake, or state an opposite case. It’s not even that you think I am bypassing your epistemic immune system.
You are instead expressing the concern that the tone I use may bypass someone else’s, which is ~irrefutable and pattern-matches to me like concern trolling.
To be clear, I do not think this behaviour is not specific to you. I think it is pervasive on LessWrong, and makes it a much worse intellectual community than it could be.
In my experience, when I have dug into them with people, such “backfires” are just normal disagreements or rejections of my points, but for reasons that do not sound “rational”, and then not owning up to it.
Nevertheless, I am thankful for your offer, and I am interested in it. This is much more detailed feedback than I normally get here. I would be prefer if you did it on a Google Doc though.
going downwards on Graham’s hierarchy of refutation.
This is likely to update my behavior somehow.
You don’t refute the central point, identify a crucial mistake, or state an opposite case.
I guess my claim boils down to “Okay, well, I mean, interesting hypothesis, but you haven’t provided much evidence for it and your emphasis is wildly overclaiming, which is very annoying for me to have to undo as I read.” This is how I feel after most of your posts. They read like the same genre as AI slop: some claims likely plausible, but I can’t trust that one claim will be true because the one next to it was, and claims are asserted rather than built up by argumentation, and visually emphasized. You give examples, but the ratio of examples to claims seems not favorable—many examples are also just claiming things without evidence. I don’t think it’s bad to claim things without evidence and cite them to “I just think this”, but I do think it’s very annoying to claim things without evidence without tagging that you’re doing so. Don’t make my vision system need to undo that. I compare to other people to describe part of why I think it’s worse than just how it affects me, but it definitely is first about its impact on my brain.
I would be prefer if you did it on a Google Doc though.
I’m likely to do that if you invite me to do so. dm me a link.
It feels like you are thoroughly misunderstanding the nature of the piece, and I am not sure what I have done for you to understand it that way.
From my point of view:
The examples are not evidence, they are examples. There are here to exemplify, illustrate and clarify the concepts that I introduce. Like, “Definitions are good, but it’s hard to get things from just a definition, so here are examples.”, or “Here are situations where I think using the concept is useful, as it may be hard to get an idea of the scope just from a definition.”
This is neither a Wikipedia Article nor it is a case for a specific thesis. [citation needed] and [weak argument] seem like not understanding the basic premise of the piece.
I am describing a pattern I have found useful, at the intersection of sociology and psychoanalysis. These are notoriously not fields with a standard evidence-based method to establish claims that broad.
Eliezer’s Sequences and Scott’s Codex are full of essays doing similar things, and I believe you would understand how much of an isolated demand for rigour it would be to go [citation needed] and [weak argument] on these essays. There’s just not a standard way to establish things in these fields, let alone based on citations and strong arguments.
So instead, like most people dealing with the topic, I am providing a frame. Good counters to a frame do not look like [citation needed] or [weak argument].
From my point of view, good counters are instead things like:
“When I see [this situation from the article], my ready-to-go explanation is [X] instead. I think it’s better because of [Y]”
“I have never seen observed this dynamic in my life. The examples feel forced, and when I try to proactively find new ones, I can’t. In my intuitive understanding of things, it feels like the opposite dynamic is more pervasive, wherein [X] happens instead”
“You diagnose [X] as the cause of [Y1 and Y2] in [a situation from article]. I think you are missing [Z], a much more important contributor given the considerations that you mention.”
“I / people I know have tried to use that frame, and they tended to make worse predictions and become less effective for it.”
I claim you are using semantic and formatting emphasis in a way that can be expected to increase rate of posts bypassing epistemic immune systems. More concisely, it reads like marketing slop. Maybe if you just want to manipulate public opinion this is fine for your purposes, though I doubt you’d want that if you thought about it enough—I expect that you’re getting backfires like me from it and that you’re better off with a mildly lower impact rate in exchange for a lower backfire rate. In this case, I think I already agreed with your post, but your writing didn’t move my opinion significantly and excessive candying of the words is part of why.
if you tell me it won’t be annoying to you for me to do so, I could leave many reacts on your post annotating the parts that lead me to this view; they’ll primarily be “citation needed”, “weak argument”, “soldier mindset”. Otherwise I’ll consider this to be the extent of my feedback and continue downvoting posts which feature-parse to me as attempting to be manipulative in this way.
To downvote based on this seems like clearly going downwards on Graham’s hierarchy of refutation.
You don’t refute the central point, identify a crucial mistake, or state an opposite case. It’s not even that you think I am bypassing your epistemic immune system.
You are instead expressing the concern that the tone I use may bypass someone else’s, which is ~irrefutable and pattern-matches to me like concern trolling.
To be clear, I do not think this behaviour is not specific to you. I think it is pervasive on LessWrong, and makes it a much worse intellectual community than it could be.
In my experience, when I have dug into them with people, such “backfires” are just normal disagreements or rejections of my points, but for reasons that do not sound “rational”, and then not owning up to it.
Nevertheless, I am thankful for your offer, and I am interested in it. This is much more detailed feedback than I normally get here. I would be prefer if you did it on a Google Doc though.
This is likely to update my behavior somehow.
I guess my claim boils down to “Okay, well, I mean, interesting hypothesis, but you haven’t provided much evidence for it and your emphasis is wildly overclaiming, which is very annoying for me to have to undo as I read.” This is how I feel after most of your posts. They read like the same genre as AI slop: some claims likely plausible, but I can’t trust that one claim will be true because the one next to it was, and claims are asserted rather than built up by argumentation, and visually emphasized. You give examples, but the ratio of examples to claims seems not favorable—many examples are also just claiming things without evidence. I don’t think it’s bad to claim things without evidence and cite them to “I just think this”, but I do think it’s very annoying to claim things without evidence without tagging that you’re doing so. Don’t make my vision system need to undo that. I compare to other people to describe part of why I think it’s worse than just how it affects me, but it definitely is first about its impact on my brain.
I’m likely to do that if you invite me to do so. dm me a link.
Thanks a lot for the response.
I am a bit baffled though.
It feels like you are thoroughly misunderstanding the nature of the piece, and I am not sure what I have done for you to understand it that way.
From my point of view:
The examples are not evidence, they are examples. There are here to exemplify, illustrate and clarify the concepts that I introduce. Like, “Definitions are good, but it’s hard to get things from just a definition, so here are examples.”, or “Here are situations where I think using the concept is useful, as it may be hard to get an idea of the scope just from a definition.”
This is neither a Wikipedia Article nor it is a case for a specific thesis. [citation needed] and [weak argument] seem like not understanding the basic premise of the piece.
I am describing a pattern I have found useful, at the intersection of sociology and psychoanalysis. These are notoriously not fields with a standard evidence-based method to establish claims that broad.
Eliezer’s Sequences and Scott’s Codex are full of essays doing similar things, and I believe you would understand how much of an isolated demand for rigour it would be to go [citation needed] and [weak argument] on these essays. There’s just not a standard way to establish things in these fields, let alone based on citations and strong arguments.
So instead, like most people dealing with the topic, I am providing a frame. Good counters to a frame do not look like [citation needed] or [weak argument].
From my point of view, good counters are instead things like:
“When I see [this situation from the article], my ready-to-go explanation is [X] instead. I think it’s better because of [Y]”
“I have never seen observed this dynamic in my life. The examples feel forced, and when I try to proactively find new ones, I can’t. In my intuitive understanding of things, it feels like the opposite dynamic is more pervasive, wherein [X] happens instead”
“You diagnose [X] as the cause of [Y1 and Y2] in [a situation from article]. I think you are missing [Z], a much more important contributor given the considerations that you mention.”
“I / people I know have tried to use that frame, and they tended to make worse predictions and become less effective for it.”
(DM-ing you with the GDoc link.)