From: Zack M. Davis Date: Fri, 4 Sep 2026 19:32:18 +0000 (-0700) Subject: check in X-Git-Url: http://unremediatedgender.space/source?a=commitdiff_plain;h=3605063790df31655a9963f181f97f7b4e6ca602;p=An_Algorithmic_Lucidity.git check in --- diff --git a/content/drafts/acausal-trade-as-limiting-case.md b/content/drafts/acausal-trade-as-limiting-case.md new file mode 100644 index 0000000..e74d967 --- /dev/null +++ b/content/drafts/acausal-trade-as-limiting-case.md @@ -0,0 +1,23 @@ +Title: Acausal Trade as Limiting Case +Date: 2027-01-01 +Status: draft +Category: philosophy +Tags: decision theory + +[TODO— +There's this idea of acausal trade, where I do something you want because I predict that you'll do something I want &c. A lot of people think this is spooky: what do you mean, you can trade without causality? You're open to trade offers from anyone in the universe (past or future), or even other universes? + +It's not that spooky. We're familiar with the concept of making deals, by using back-and-forth communication to pin down the terms of the deal. + +In acausal trade, you're _inferring_ the terms of the deal rather than communicating them using physical signals. + +But even in ordinary deals, there's some amount of inference involved. Real-world contracts don't pin down every detail; there's still an element of inferring what the other guy wants (because you think that he'll be trying to infer what you want to uphold his end of the deal). (And no matter how explicit a contract is, it's still your brain that has to interpret the words and infer the meaning; the symbols themselves are inert.) + +The amount that's spelled out in the contract vs. needing to be inferred is quantitative. Less extreme than the explicit contract, you can bring your items to the swap meet and say, "Trade?" + +Or prisoners deciding when to rush the guards and start a riot can give each other a knowing look. They don't need to negotiate a long contract of "I'll rush him if you do". A sufficient model of each other's goals makes the knowing-looking enough communication to make the deal go through. + +Acausal trade is just the limiting case where there's _no_ physical communication, and it's _all_ inference. But it's really not that much spookier than the prisoners (who only need a little bit of physical communication). + +] + diff --git a/content/drafts/contra-piper-on-when-conversation-is-possible.md b/content/drafts/contra-piper-on-when-conversation-is-possible.md index 58d86ed..832df46 100644 --- a/content/drafts/contra-piper-on-when-conversation-is-possible.md +++ b/content/drafts/contra-piper-on-when-conversation-is-possible.md @@ -24,9 +24,9 @@ However, I reject the premise that the accusation or its denial can be made just In the same way, people don't need to know what it feels like to be me in order to speak sensibly about rejecting my own understanding of my beliefs and motives. They could claim that my stated beliefs and motives are inconsistent with my behavior (which they can see) while remaining agnostic about how it feels to be me (which they don't have access to). Such a claim [probably wouldn't be based on nothing at all](/blog/2022/May/plea-bargaining/); indeed, the very existence of persistent disagreements (when you'd expect honest rational agents to simply update on each other's evidence and converge on the correct probability) [suggests that hidden motives are common](/blog/2023/Aug/assume-bad-faith/). -If I buy cookies for the [salon](https://en.wikipedia.org/wiki/Salon_(gathering)) from a bakery owned by my girlfriend in Australia (at a substantial markup for international shipping), it kind of makes sense for other salon members to wonder if my choice might have been biased by nepotism rather than reflecting true optimization towards the goal of the salon having the best desserts. +If I buy cookies for the [salon](https://en.wikipedia.org/wiki/Salon_(gathering)) from a bakery owned by my girlfriend in Australia (at a substantial markup for international shipping), it kind of makes sense for other salon members to wonder if my choice might have been biased by favoritism towards my beloved rather than reflecting true optimization towards the goal of the salon having the best desserts. -At the same time, my belief that I wasn't swayed by nepotism is something I can provide reasoning and evidence for: for example, I could try showing the other salon members the notes and spreadsheets I compiled while comparison shopping, which justify my belief that my girlfriend just happens to be the optimal dessert vendor for the salon according to the obvious objective criteria—price, quality, [Anzac biscuit](https://en.wikipedia.org/wiki/Anzac_biscuit) availability, and owner's prior familiarity with the salon's reading list. +At the same time, my belief that I wasn't swayed by favoritism is something I can provide reasoning and evidence for: for example, I could try showing the other salon members the notes and spreadsheets I compiled while comparison shopping, which justify my belief that my girlfriend just happens to be the optimal dessert vendor for the salon according to the obvious objective criteria—price, quality, [Anzac biscuit](https://en.wikipedia.org/wiki/Anzac_biscuit) availability, and owner's prior familiarity with the salon's reading list. I might not convince them, but at least we have _something to talk about_: it's not a deadlock between an unfalsifiable accusation of self-deception and an unfalsifiable denial, because my catering decision (and therefore the beliefs and motives that led to it) was tangled up with empirical realities of the world. If my comparison shopping was thorough and credibly documents the many ways in which my girlfriend's bakery is just the best, that makes my story about only wanting the best cookies for the salon more credible; if I didn't really shop around much, that makes my story less credible. The correctness of my understanding of my beliefs and motives is something I can rationally defend to skeptics in front of third parties; it's not something that needs to be assumed as a prerequisite for even talking at all. @@ -40,10 +40,10 @@ Notice how Piper writes about Ngo calling liberals self-deceived: warning him th But positing self-deception isn't just a "move" that one could decline to make while still transmitting the same information; the "frame" is the expression of a hypothesis that could be true or false, which Ngo believes to be true. (Piper characterizes Ngo as claiming that he "should make up beliefs [he thinks liberals] have", but [Ngo says](https://x.com/RichardMCNgo/status/2071372646517035120) his distrust of liberals' self-identifications is based on what he sees as their failure to "reason very explicitly about how much they've capitulated to wokeness"; reasoning from behavioral evidence aspires to be more disciplined than just "mak[ing] up" things.) -Furthermore, as I've discussed above, the cost of not being able to communicate with people you're accusing of self-deception isn't a logical or physical inevitability; it's a choice imposed by the people who don't want to talk to people who don't already trust them. +Furthermore, as I've discussed above, the cost of not being able to communicate with people you're accusing of self-deception isn't a logical or physical inevitability; it's a choice imposed by the people who don't want to talk to those who don't already trust them. -In light of these observations, I think Piper's apparently well-meaning advice is better understood as a negotiating threat: don't criticize us in this way, or we'll walk. Even if you privately think us self-decieved, you must not reveal that you think that, or we'll destroy the surplus of you being able to interact with us. +In light of these observations, I think Piper's apparently well-meaning advice is more naturally understood as a negotiating threat: don't criticize us in this way, or we'll walk. Even if you privately think us self-decieved, you must not reveal that you think that, or we'll destroy the surplus of you being able to interact with us. -If Piper doesn't think of herself as making a negotiating threat, I guess that implies that I'm rejecting her self-understanding? But, you know, I don't think I'm doing it maliciously. I'm just applying my knowledge of game theory, human psychology, _&c._ to the words I see before me and observing, hey, this looks to me like a veiled negotiating threat intended to suppress a class of criticism; that's an interesting phenomenon to analyze in a blog post. +If Piper doesn't think of herself as making a negotiating threat, I guess that implies that I'm rejecting her self-understanding? But, you know, I don't think I'm doing it maliciously. I'm just applying my knowledge of game theory, human psychology, _&c._ to the words I see before me and observing, hey, my max-likelihood interpretation of this behavior is that it looks like a veiled negotiating threat intended to suppress a class of criticism; that's an interesting phenomenon to analyze in a blog post. But if you think I'm kidding myself about that, then I'm happy to talk about it. diff --git a/content/drafts/listening-disorders-as-eating-disorders.md b/content/drafts/listening-disorders-as-eating-disorders.md index 443c8ac..f3e606b 100644 --- a/content/drafts/listening-disorders-as-eating-disorders.md +++ b/content/drafts/listening-disorders-as-eating-disorders.md @@ -36,13 +36,13 @@ The analogy is: just as I hold some people in absolute contempt for being so fin To be clear, it's not a great analogy. The domains of information and food intake aren't that similar, so of course there are disanalogies that run in my favor. For example, I could say that my health is no one's business but my own, but that falsehood in public discourse affects everyone. Or I could point out that I'm not _claiming_ to be some athlete with an optimally nourished body (where a suboptimal diet could call the claim into question), whereas the people I feel contempt for are claiming to be truthseeking intellectuals. -But the power of the analogy is that it forces me to attend to the practical weight of an involuntary Absolute No, rather than brushing it off with platitudes about idealized Bayesian reasoners. If someone tried to bully me into changing my diet with appeals to what an idealized healthy human would eat, I would be unmoved. I think whatever health deficits I'm suffering from my eating disorder are quantitatively small. Whatever nutrients I'm missing out on by not being able to tolerate most vegetables, my understanding is that I can fill out a relevantly similar profile with fruit, legumes, and whole grains. +But the power of the analogy is that it forces me to attend to the practical weight of an involuntary Absolute No, rather than brushing it off with platitudes about idealized Bayesian reasoners. If someone tried to bully me into changing my diet with appeals to what an idealized healthy human would eat, I would be unmoved. I think whatever health deficits I'm suffering from my eating disorder are quantitatively small. Whatever I'm missing out on by not being able to tolerate most vegetables, my understanding is that I can fill out a relevantly similar nutrient profile with fruit, legumes, and whole grains. But then the people I'm trying to bully into changing their information diet could say the same thing. They might concede in the abstract that they're theoretically missing something by not being an info-omnivore but insist that the practical effect is quantitatively small. Whatever information they're missing out on by not being able to tolerate Sneer Club, they think they can fill it out from other sources.[^sneer-deficit] [^sneer-deficit]: In the case of Sneer Club in particular, I mostly agree. -The analogy also stress-tests my haughty claim that the people I'm complaining about could be more like me if they wanted to. It's not that the claim is wrong, but the conditional is doing more work than I was giving it credit for. I think I _could_ beat my eating disorder—if I wanted to. There's lots of interventions to try: figure out what dimensions the aversion is responding to, do repeated exposure therapy in increasing doses, get skilled at cooking. Acquired tastes are a thing. (No one "innately" thinks coffee tastes good, and yet I drink decaf coffee, because at some point I acquired a coffee taste that remained even after I started being scared of caffeine.) I'm sure I could acquire a vegetable taste—if I made a project out of it. +The analogy also stress-tests my haughty claim that the people I'm complaining about could be more like me if they wanted to. It's not that the claim is wrong, but the conditional is doing more work than I was giving it credit for. I think I _could_ beat my eating disorder—if I wanted to. There's lots of interventions to try: figure out what dimensions the aversion is responding to, do repeated exposure therapy in increasing doses, get skilled at cooking. Acquired tastes are a thing. (No one "innately" thinks coffee tastes good, and yet I drink decaf coffee, because at some point I acquired a coffee taste that remained even after I started being scared of caffeine.) I bet I could acquire a vegetable taste—if I made a project out of it. But I don't want to make a project out of it! I just don't! One of the things that makes me most grateful to be an economically independent adult rather than a child is that I can control my own food supply. I buy and prepare the foods that I want to eat, and it's not a problem if I don't say it's a problem. @@ -65,6 +65,7 @@ So, should I have mercy on people with listening disorders? Should I not take pu ] + [TODO scrap or footnote— And somehow I got away with it, I won the childhood battle of wills? If that seems socially deviant, the secret is that there's only so much labor people are willing to do to force you. My father made me eat a carrot when I was ten, and I threw up on the spot. He didn't try again. Because, well—why would he? It's just less work to let me keep being childish. I suspect that this is more common than people admit? (link to mentions of vegetable aversion by Eigenrobot, Aella, Scott Alexander, Zvi Mowshowitz) diff --git a/content/drafts/the-archipelago-of-crybullying.md b/content/drafts/the-archipelago-of-crybullying.md new file mode 100644 index 0000000..60f1993 --- /dev/null +++ b/content/drafts/the-archipelago-of-crybullying.md @@ -0,0 +1,46 @@ +Title: The Archipelago of Crybullying +Date: 2027-01-01 +Status: draft +Category: philosophy +Tags: regret + +Oliver Habryka [writes](https://www.lesswrong.com/posts/EQJfdqSaMcJyR5k73/habryka-s-shortform-feed?commentId=C64s9RDszhBa5krjK): + +> [I]n part due to threads like this people have never felt sufficiently safe using a bunch of our moderation tools to the extend we want them to be used. It sucks. +> +> Even as the people running the site there are limitations to what degree we can make an option feel legitimate, and since indeed every time someone deletes anything tons of people will suddenly cry "CENSORSHIP!!! The site culture is dead, the author is evil", there is not that much we can do. +> +> It's not the most important fight to fight. It mostly results in annoying people getting site bans instead of author bans, which sucks, since I think an Archipelago would be better by the lights of almost everyone involved. + +Wei Dai [replied](https://www.lesswrong.com/posts/EQJfdqSaMcJyR5k73/habryka-s-shortform-feed?commentId=thsbwa8PEaRrnvEvs): + +> Have you considered that this seems an impossible goal? Whenever an author moderates someone, that person is very likely to think the author did so unfairly, and sometimes complain about it, and it's hard to defend against this charge since it's just so obvious that authors would likely be biased against their critics on priors, and there's not much evidence you can update on to overcome this presumption. + +Habryka [replied](https://www.lesswrong.com/posts/EQJfdqSaMcJyR5k73/habryka-s-shortform-feed?commentId=BJhx39ctrRg6qswot): + +> Yeah, I have! But then like 10 social media platforms pulled it off. They have other issues, but clearly it is not an impossible goal to create a place where people have tools to avoid having to engage with specific other people directly while still hosting enormous amounts of content. Facebook, Twitter, Substack. It's totally standard on the modern internet to give people their own moderation tools, indeed anything else seems very difficult. + + + + +A place where people _have tools_ + + + + + + + + +------ + +Archipelgo vs. Safe Archipelago + +https://www.lesswrong.com/posts/adk5xv5Q4hjvpEhhh/meta-new-moderation-tools-and-moderation-guidelines?commentId=KqtPsTP8TdWQnhjtD + +> Yep, moderation sucks, competing access needs are real, and not everyone can share the same space, even within a broader archipelago (especially if one is determined to tear down that very archipelago). + + +https://www.lesswrong.com/posts/EQJfdqSaMcJyR5k73/habryka-s-shortform-feed?commentId=PCrWkKmQS7Gy8H5YC + +> Yeah. And here is a cool idea, maybe the people who run those smaller subforums should have moderation privileges on their subforum, so they can set their own moderation guidelines, and cultivate an archipelago of cultures without needing to coordinate with site moderators all the time. /s