Celarix

Karma: 451

Celarix 18 Jul 2026 20:37 UTC
1 point
0
in reply to: PatrickDFarley’s comment on: Connect to your past selves
I love this question! Pondering this helped me a bit.

I find it easier to take instruction from someone who, even if I don’t agree, seems to have solid, well-reasoned reasons to ask me to do something. At my job, the tickets are clear and usually pretty obvious to their value, and if not, I know I can ask for clarification. If someone asks for my help moving furniture, I may not want to, but I can see why they’d want it.

Applying those traits back to myself is not something I’ve thought of, and seems very promising! Usually:
- Past-me wanted to do $Thing because it was novel and cool.
- Current-me wants to do $OtherThing because it’s novel and cool and $Thing just isn’t.
Still, I think there’s definitely some stuff I can suss out about $Thing that could make me interested in doing it. Mentally give the part of me that wants to do $Thing a chance to state their case and see if that sways me.

Thank you!

Celarix 30 Jun 2026 23:15 UTC
2 points
0
on: Connect to your past selves
But they need to be in conversation with their old motivations to find the best path for their present self. Too much disconnection leads to the typical ADD pattern of always starting new plans but never following through.
oh_hey_its_me.jpg
I struggle with this quite a bit—any tips to help me look at my old motivations and say anything besides “I don’t care, not gonna do that, you can’t make me, I’m gonna do $ShinyNewThing, neener neener?” I remember my past motivations but since they aren’t as novel as the current one, they feel like a slog to get back into.

Celarix 30 May 2026 22:05 UTC
1 point
0
on: Use Decision Theory To Fix Your Bad Habits

Although I don’t exactly know why? Is this because of increasing marginal costs?

My guess: the worse a thing is, the fewer times you can do it before sustaining severe damage.

Celarix 10 May 2026 16:36 UTC
2 points
0
on: Sawtooth Problems
That tweet is the second worst thing Tim Urban ever did to me, lol.

I don’t think the blender scenario is quite equivalent unless it’s specified that the blender is off until everyone’s choice is revealed. And people do tend to be reticent about changing parameters too much.

I do love the charts and parameterization! That “regions” graph is quite good as well, articles like this say something useful about the question rather than just bicker about it ceaselessly.

Celarix 1 Apr 2026 22:16 UTC
1 point
0
on: Lesswrong Liberated
LessWrong Meets the NFL
Sonnet is very kind, giving us the lead in the 4th quarter against the Entire Internet.

Celarix 1 Apr 2026 22:13 UTC
1 point
0
on: “You Have Not Been a Good User” (LessWrong’s second album)
Not in-character, I Knew the Name Before It Had One is definitely not something I’d clock as AI on first glance, wow.

Celarix 23 Feb 2026 23:26 UTC
1 point
0
in reply to: romeostevensit’s comment on: Terminal Cynicism
I think I get it better, please forgive the repeated checking, I just want to be sure I have it clear since I love thinking about how emotions shape cognition and habit.

My read is now “I have done all I can think of/feel I have the capacity for, but the problem still isn’t solved, and advocacy is the only option I see at this very moment, so that’s what I’ll do for now?”

Celarix 22 Feb 2026 7:51 UTC
1 point
0
in reply to: romeostevensit’s comment on: Terminal Cynicism
This seems neat! Just to clarify, do you mean something like “I’ve explained seemingly all currently available facts, but I haven’t fully discharged my emotions on this, so there must be something more I’m missing?”

Celarix 17 Feb 2026 3:33 UTC
2 points
0
in reply to: FireStormOOO’s comment on: Stone Age Billionaire Can’t Words Good
Oh, okay! This is much clearer and I apologize for not getting your point the first time. I agree pretty much fully with this—insurance companies pushed the envelope until the pain got high enough that, sadly, a CEO lost his life without trial, and that poisoned the well and made it a lot harder on them going forward. It is debatable the exact cause of their retreat on this front, but I do agree more broadly with this.

Celarix 14 Feb 2026 0:09 UTC
2 points
1
in reply to: FireStormOOO’s comment on: Stone Age Billionaire Can’t Words Good
Then all the more reason to favor a reform to the system so that large companies can be restrained via due process. Everyone has a different idea of who the bad guys are, and many have very high priors that “they” are the ones responsible for all the problems, that “they” are unimpeded by the guardrails as well.

Tribalism makes information believable not because it is true, but because it is useful toward hurting the enemy. These feel the same from the inside. Hating UHC is a lot more justified than people on the Right hating LGBTQ+, by quite a large margin, but to give us the power to hunt down healthcare executives is a step toward giving them more power to hunt down LGBTQ+.

Celarix 11 Feb 2026 4:31 UTC
14 points
8
on: Stone Age Billionaire Can’t Words Good
I will throw my support toward a related position: Due process for all. Yes, even him.* To the anti-billionaires:

Is Brian Thompson culpable for mass murder? Great, try him in court and you even get the death penalty if you’d like. Is Luigi Mangione the one who shot him? Not yet proven beyond a reasonable doubt, if the state can’t prove it, then I support him walking free.

The system prevents trying rich people for their crimes? Reform the system. You do not want vigilantism. It will not attack only people you hate, the ones you think are the cause of all the world’s problems. Everyone has a different idea of who that is and open vigilantism means a lot of innocent people will die, including people like you!

An estimate of about 4% (1) of people executed by the state may be innocent. And this is after years of trials, appeals, and investigation! Do you think you can do better?

Love billionaires or hate them, fine, whatever, you can hate whoever you like. But flat-out advocacy for murder is not a door you want to open.

*what’s great about a non-specific “him” is that there’s so many people this applies to! Everyone will fill in their own “him”.

Celarix 28 Jan 2026 6:13 UTC
5 points
3
on: To be well-calibrated is to be punctual
The second half will definitely see some spirited debate, but I do want to point out how good that idea in the first half is—travel time is definitely one of the most easily trackable things to update on, could be good rationality practice to track it more even if you don’t have much of a problem with it. That would also let you be wrong in a guilt-free environment.

Celarix 28 Jan 2026 6:10 UTC
3 points
1
in reply to: NoUsernameSelected’s comment on: To be well-calibrated is to be punctual
The nerd in me wants to start keeping a spreadsheet and track that variance and get a 95% confidence interval. The rationalist in me knows to gently put that urge aside, haha.

Celarix 14 Jan 2026 7:01 UTC
11 points
2
on: Strong, bipartisan leadership for resistance to Trump.
Long comment is long. Most of this is either vague, wrong, or stuff you already knew 20 years ago, but I wanted to share my thoughts anyway.

Please note none of this is justification, you will likely think “no, that’s entirely wrong” a lot throughout this. I know, I’m sorry, I wish it could be different.

Most of this is inspired by Scott’s much better post “The Psychopolitics of Trauma”: https://www.astralcodexten.com/p/the-psychopolitics-of-trauma

You can feel badly hurt by something. You can feel badly hurt by something and have every single fact wrong. Diametrically wrong. That doesn’t matter. Emotional wounds run deeper than truth can reach.

We often hear “but Biden” or “but Hillary” or “but the WEF” from the right. Whenever you try to bring one of Trump’s many, many faults to the table, they come back with “but Biden broke it” or “but what about her e-mails?”. Why? What is it about these people that rile up the right make it seem like such a trump card against your argument?

I think everything is tied to everything else in their minds. I think it’s similar to a troubled relationship where any flaw of one person is met with the laundry list of past mistakes. Leave your socks on the floor and you get flak for not rotating the tires earlier or not raking the leaves or being too emotional or whatever.

The hate for Biden is especially humorous. He was President for only 4 years, elected for being the plainest option to beat Trump, and now the right consider him the evil old man who broke it all. I think that Biden isn’t special, nothing he did in particular draws their ire. He’s just the avatar of the left, after Hillary, after Obama, after Clinton. They hate the seat, not the person.

Republicans are hurt, badly hurt. Falsely hurt, but they’ve spent a long time hearing the false reasons. Social media algorithms are a lot of it, every flaw of the right is explained away with a falsity or a bogus study out of context. But I think talk radio had a huge impact first. We mostly kind of ignored it as a ranting person on AM, but I think hearing 3 hours a day of it for decades on end slowly breaks you.

When you listened to Rush Limbaugh, you got to feel like people were on your side. Like someone was pointing out the emperor’s new clothes. That every problem of the right was an inconvenience at worst, never heard about at best. And every flaw of the left, real or not, was amplified, made their whole character. Democrats stopped being bad for specific reasons, they just became bad. And the more bad they seemed, the more you’d believe about them.

And it was 3 hours a day, 5 days a week, for 30 years. I listened to talk radio growing up. I went from “wow, these people are so smart, fighting the left like this!” to “okay, ouch. We have good ideas but we have to present them better, why are you harping on this?” to “There is nothing here for me.”

Now, why Trump? Trump is an anathema to the purported Christian values of the right and yet it doesn’t matter. I don’t really get it myself. But I don’t think logic was ever in play—the farthest right believe in their own correctness first, the logic confabulated later, like the woman with the brain injury unable to update on her paralysis and her brain instead confabulating reasons why she can’t move it (https://www.lesswrong.com/posts/ZiQqsgGX6a42Sfpii/the-apologist-and-the-revolutionary).

Okay, yeah, maybe it’s not that bad. I just think similar mechanisms are in play. I’ve done it myself—my System 1 really wants to not do a thing, and I end up saying lots of words I don’t really believe just to get out of it. A lot of it is tribalism, but a lot of that tribalism here is built on the ceaseless message: “They hurt you. They hate you. They want you gone.”

It is not true. But when you’ve heard it for your whole life, when everyone you respect says it, when charismatic speakers put up logical but false arguments and posit immense conspiracies, and you really have been hurt by something (NAFTA or opioids or algorithms or Walmart or neap tides or whatever), it’s easy for someone to say “Trust me. They hurt you. If you hurt them, you will heal.” as a play for power.

Trump 2 is, god, I don’t know. I think of him more as an RNG than a shrewd planner. No Republican of 2012 (I know this, I was there) cared about Greenland. I don’t know why we want Greenland, even ignoring the horrible things we’d do for it. Why ICE doesn’t seem to be doing much deportation but is threatening a ton of random citizens and even killing them now. There’s a similarly to the paralyzed woman in Scott’s post (in kind but not in scale!) - what they do is random and punitive and senseless, and when asked WTF they are thinking, they confabulate answers based in their rage.

None of it is real. They thought the COVID vaccines were genocide, killing up to 90% of the population. I heard it often, but not one who believed it acted like they did. Imagine if you knew that a full 7,200,000,000 people were going to die! A death toll higher than all war combined. You’d probably be gathering supplies. Getting guns and ammo. Waking up in cold sweat, nightmares of bodies piled up everywhere, beyond too many to bury. The end of your old life. Almost everyone you love dying and you can’t do a damn thing about it. I mean, wouldn’t you at least try to declare war on Pfizer or firebomb a Walgreens or something, if you really believed it?

Some may have taken that idea seriously, I never met any. It wasn’t a belief evaluated, it was just a bunch of words modulated onto screaming, System 1 screaming “THEY HURT US! THEY HATE US!” and System 2 knowing it has to come up with something. Learning about the conspiracy was often a moment of smugness, a moment of “see! see! they do hate us!”, another little dopamine hit of “yep. turns out we’re right again!” rather than the intolerable horror it should have been.

None of their arguments, nor their beliefs, are the real reason they support Trump and his actions. To turn one would require a monumental effort—I think it is possible, but I don’t think most people are capable of it. You can only do it one-on-one, it will take a long time, and you will have to sit and smile to real attacks on your character, on people you love, groups you ally with. It is uncomfortable and I can’t ask anyone to do unless they really need to, and it will look less like “ah, I see now, you’re right” and more like a very slow shift over months or years, all of your ideas ones they think they came up with themselves. It’s not satisfying… but I think it is possible.

Celarix 11 Nov 2025 23:59 UTC
2 points
0
in reply to: dreeves’s comment on: Against Powerful Text Editors
Oh, yeah, AI can definitely introduce subtle errors, but maybe at a lower rate? I think the main value is seeing a big change at once, i.e.:
- 12345
- 12345
- 12b45
- 12345
- 12345
highlights the error immediately, whatever method is used to perform the quick or instant change.

Celarix 10 Nov 2025 23:38 UTC
2 points
0
on: Against Powerful Text Editors
One additional point in favor of using editing macros to do lots of work at once is that it reduces the odds you might make a mistake. Hand-editing dozens of the the lines of code, especially if it’s very tedious, could mean a typo or two slips in, which could be annoying to go back and fix and can hide in plain sight; an automated macro will either get them all right, or get them all wrong in a similar way, hopefully sticking out.

Celarix 27 Oct 2025 23:07 UTC
1 point
0
on: Credit goes to the presenter, not the inventor
It’s worth it to learn to sell your own ideas, not in a bullshitting or kissing-up sort of way, but just in learning to present what you believe in your ideas honestly. Then you can be the presenter and the inventor.

Celarix 12 Oct 2025 2:40 UTC
1 point
0
on: Hospitalization: A Review
Strong upvoted, these guides on what to expect for situation X are very useful, especially when detailed nuts-and-bolts like this. There’s a lot of good stuff here that anyone who has a pneumothorax can use to at least set a baseline of what to expect. I find that too many people stop at “eh, well, everyone kind of knows this” and thus few ever write about it.

Glad you’re okay and that you have such an awesome wife!

Celarix 4 Oct 2025 16:36 UTC
1 point
0
on: Omelas Is Perfectly Misread
“Imagine there’s a child drowning in a shallow pond. You’re wearing a swimsuit and could easily save them. Sounds impausible? No, really, you can just save the kid. Don’t trust it? Okay, let me make it more believable: imagine there’s also a cute puppy guarding the pond that you’d have to kill to reach the child. Would you do it?”
My gut reaction to reading this might illuminate why people don’t take stories about utopia or easily-accomplished things seriously: there doesn’t seem to be much benefit to be gained by telling a story where everything goes well, there’s no dilemma or moral problem, and nothing salient to latch on to.
Imagine there was a TV show about a starship. It is run well, the ship functions normally, the crew go about their days, there’s plenty of power and supplies, the journey is vast and not very notable, their mission of moving cargo between two planets back and forth goes fine with no complaints. Once you got past the 2- or 3-episode explanation and tour of the ship and her crew, there wouldn’t really be much else different on episode 105 as there was on episode 5.
What would it take to make this show better? Conflict of some sort, perhaps between the crew, or with the environment, or the government, or maybe the suppliers/buyers of the cargo, or maybe with the isolating environment of space and the ennui of staring out a mostly-black window ²⁴⁄₇.
I also posit that we see this pattern in our own lives; we need something to do with our time, even if that something is just a hobby project or reading a few books. Consider fresh retirees in good shape: home paid off, kids in their own careers and doing well, spouse happy. Many such retirees may choose to take it easy, resting and watching TV, and this often either makes them miserable enough to reconsider or saddles them with health problems from their sedentary lifestyle, sometimes leading to death. Contrast with retirees who still make and keep plans, still go traveling, work on projects.
So that’s my guess. People don’t take stories about happy worlds where nothing goes wrong very seriously because there just isn’t much to say about such worlds, so why write stories about them?

Celarix 25 Aug 2025 3:36 UTC
6 points
2
in reply to: FeepingCreature’s comment on: Banning Said Achmiz (and broader thoughts on moderation)
Okay, this is definitely true, too. I also do enjoy a more consistent ability to justify my actions and beliefs, which is far from nothing and not worth writing off. I guess, for me, the missing ingredient is that the other person gets it once I make a logical and reasonable justification; if that happens, I think it’s fine to be friends with a very critical person.