NoSignalNoNoise

• NoSignalNoNoise 7 Sep 2026 19:01 UTC

  9 points

  0

  on: Open Thread Au­tumn 2026

  Is there a repeatable process for getting stuff written in a reasonable amount of time?

  I take an excessively long time to write most things and I generally find the writing process unpleasant (I think these both cause each other), so I tend to avoid writing.

  I got an 800 on the writing SAT, which implies that I’m good at things like grammar and sentence structure, but if I have to write something longer than a paragraph, I get stuck and I often avoid texting due to similar difficulties).

  Is there a specific process (as opposed to something vague like “write a creepy rough draft and then edit it”) that you use that gets stuff written quickly? What step by step sequence of actions do you take when you want to write something? (I know the process will be different for different kinds of writing, but any example at all would be helpful)

• NoSignalNoNoise 1 Sep 2026 1:51 UTC

  2 points

  −1

  on: NoSig­nalNoNoise’s Shortform

  Are strong arguments persuasive by default?

  I’d like to posit the hypothesis (epistemic status: uncertain) that if a person who is paying attention is presented with a strong, comprehensible argument for a correct conclusion, they will usually be persuaded.

  Since the surgeon general’s report on the hazards of smoking, the rate of smoking among US adults has dropped fairly steadily from 42.4% to 11.6%, indicating that close to ³⁄₄ of people who would otherwise smoke were persuaded not to. Pretty much everyone knows that the Earth is round. 81% of Americans have a an unfavorable view of Communism (vs 10% favorable), and Fascism is at 80% unfavorable vs. 5% favorable.

  Basically, when the information is comprehensible and widely disseminated, a significant majority of people believe the sane thing and make the sane choice.

  Perhaps (bounded) rationality is the rule, and exceptions are what call for explanation (e.g. motivated reasoning, inferential distance, intentional disinformation, respectability bias, etc.).

NoSig­nalNoNoise’s Shortform

• NoSignalNoNoise 27 Aug 2026 1:08 UTC

  23 points

  9

  on: NoSig­nalNoNoise’s Shortform

  People need better heuristics, not different biases.

  Related: https://​​slatestarcodex.com/​​2014/​​03/​​24/​​should-you-reverse-any-advice-you-hear/​​

  A lot of the advice I’ve received, especially as a child, teen, and undergrad, was bad. A common theme in bad advice I’ve received is that it was premised on the idea that the recipient (maybe me personally, maybe a group I’m part of that it’s targeted at) is making a bad tradeoff and should make a different choice on that tradeoff.

  Typical such advice might be something like:

  • Think for yourself; don’t give in to peer pressure.

  • Stop arguing about every little thing (and by implication, the thing you’re currently trying to argue about is small and unimportant).

  • Body image issues are very damaging, so don’t worry about your weight.

  • Always try your hardest.

  On the margin, following one of these pieces of advice might cause someone to make better decisions, or it might push them further in a direction they’ve already gone too far in, e.g. telling younger me to be less susceptible to peer pressure. The target audience might be receptive to the advice, or they might interpret it as a personal attack from a perspective that’s biased against them, e.g. telling younger me not to get in so many arguments.

  The problem is that all these pieces of advice simply assert that the recipient is making a mistake but don’t provide the tools to help them identify when they are or aren’t making that particular mistake. Whether they take the advice becomes a matter of how much they trust the source and how predisposed they already are to follow it. I would expect trust for the source to be generally uncorrelated with whether the advice is good, and I would expect being predisposed to follow certain advice to be negatively correlated with being the sort of person who would benefit from following it.

  Teaching someone how to identify signs of groupthink and which situations depend on social harmony vs on getting the right answer (and which specific choices people tend to regret) is much more useful than some generic dictum about the dangers of peer pressure.

  As an adult, I’ve learned to ask the sorts of questions that help me turn advice I receive into better heuristics, but it’s still usually presented as “adjust your bias” rather than “here’s a better heuristic”.

• NoSignalNoNoise 4 Aug 2026 19:21 UTC

  2 points

  0

  in reply to: Aprillion’s comment on: Si­mu­lated Users & Sad AIs

  What I had in mind was integrating honeypot questions into larger evals. If most of the questions can be answered legitimately but a small number of them can’t, then you’re still mostly evaluating real (non-cheating) capabilities.

  I’m not sure how large hacks could get rewarded. If the only way to answer a question correctly is to cheat, then anything that gets it right is cheating. Unless you mean hacks large enough to straight-up directly change the evaluation score, in which case we have bigger problems.

• NoSignalNoNoise 27 Jul 2026 23:27 UTC

  4 points

  1

  on: Si­mu­lated Users & Sad AIs

  It is surprisingly easy for a conscientious, careful actor to construct an environment where the only way to pass is through reward-hacking.

  I know what you meant there is that it’s easy to do by mistake, but this got me wondering how difficult and how useful it would be to do this intentionally in order to create a honeypot to detect reward hacking.

• NoSignalNoNoise 22 Jul 2026 23:44 UTC

  3 points

  1

  in reply to: Richard_Kennaway’s comment on: We should push for no-fault li­a­bil­ity for ac­tions taken by AI

  Disclaimers can shield against liability for harm to willing users, but not harm to third parties, and harm to third parties is the main concern here.

• NoSignalNoNoise 17 Jul 2026 0:03 UTC

  12 points

  9

  on: Re­cap of bike trip/​street in­ter­views across America

  Strong upvoted for practicing the virtue of empiricism. We’re trying to solve complicated problems that nobody really understands, so it’s good to see people trying (safe) stuff and reporting what happens.

• NoSignalNoNoise 16 Jul 2026 23:53 UTC

  12 points

  3

  in reply to: Christopher King’s comment on: Guess on why ra­tio­nal­ity is not more pop­u­lar (there are no pam­phlets)

  There’s a vast oversupply of pamphlet-sizes documents that claim to teach you those skills. How does this one rise above the noise enough for someone who isn’t already familiar with its ideas to bother reading it?

• NoSignalNoNoise 29 Apr 2026 3:43 UTC

  2 points

  0

  on: Why no new no­ta­tions since 1960?

  Some common ones include:

  • Domain names and email addresses

  • @ to addresses a comment to a specific person in a context where that comment is visible to a wider audience

  • Hex codes for colors

• NoSignalNoNoise 8 Apr 2026 3:51 UTC

  8 points

  1

  on: Te­lescopes Need Good Lenses

  Israel is an interesting anomaly: a modern, technologically integrated society reproducing slightly above replacement. But the most persuasive explanation I’ve seen, NonZionism’s “trickle-down natalism,” attributes Israel’s fertility to the cultural influence of the Haredim, a community with precisely the direct-accountability structure this thesis predicts would be necessary.

  Aren’t the Haredi only able to live the way they do thanks to special treatment (welfare, subsidized religious schools, and until recently, draft exemptions) from the government? I wouldn’t hold them up as an example of a society that successfully sustains itself.

• NoSignalNoNoise 12 Mar 2026 23:49 UTC

  2 points

  0

  in reply to: jimmy’s comment on: Don’t Let LLMs Write For You

  What do you mean by “but English class”?

Ba­sics of Ra­tion­al­ist Discourse

• NoSignalNoNoise 18 Feb 2026 4:36 UTC

  2 points

  −2

  on: Most Ob­servers Are Alone: The Fermi Para­dox as Default

  the fraction of configurations capable of producing complex chemistry, stable stars, and long-lived planets is extraordinarily small. The fraction capable of producing sentient technological civilizations is smaller still. This gives us a distribution of expected civilizations per configuration that is overwhelmingly concentrated at zero, with a thin tail of configurations that produce any sentience at all

  This is a strong assertion that requires justification. How much more fine-tuning does it require to get a universe where 1 of every 1000 stars has a planet that can support intelligent life than a universe where 1 of every 10^30 stars does?

• NoSignalNoNoise 18 Feb 2026 4:31 UTC

  1 point

  0

  on: Most Ob­servers Are Alone: The Fermi Para­dox as Default

  (A3) The fine-tuning results from cosmology extend in the relevant way: the viable region of parameter space does not merely shrink as we add requirements for habitability, but shrinks fast enough that the tail of the distribution is thinner than linear.

  This assumption is rather underspecified; what does it mean to add requirements for habitability, and how does that translate the number of habitable (or intelligent-life-supporting) planets?

• NoSignalNoNoise 17 Feb 2026 5:45 UTC

  7 points

  2

  on: Con­tra Ca­plan on higher education

  college grades are far from the best way to signal intelligence; what he doesn’t discuss is that they’re even further from the best way to signal conscientiousness … an even better signal of conscientiousness would be acquiring all the same knowledge without attending college at all

  Acquiring the knowledge without the college structure would be a more accurate signal of conscientiousness, but it would be a less legible one. Evaluating potential employees involves sifting through a large number of applicants of mostly dubious quality, and unless you have reason to pay special attention to one of them, if a signal isn’t obvious, it probably gets rounded off to zero.

  A degree from Stanvard Institute of Technology gets an employer to treat your resume as worth reading (and not just skimming). Claiming (with evidence that require crossing a non-trivial inferential distance) to have gotten the same education for $1.50 in late fees at the library doesn’t. You get about 5 words.

[Question] Why did you buy Bit­coin?

• NoSignalNoNoise 17 Feb 2026 5:11 UTC

  3 points

  1

  on: Con­tra Ca­plan on higher education

  This post seems to implicitly assume a monocausal explanation. It gives good reasons why various proposed explanations are not sufficient to explain university attendance and dismisses those explanations because they aren’t sufficient. The reality is more likely a complex mix of different reasons, and the right question isn’t which one is the reason; it’s how much of the reason is each proposed explanation.

  The Sheepskin Effect makes it pretty clear that a significant portion of the value is signaling (Wikipedia cites Caplan as claiming that “over 60% of the economic benefit” is from the degree, though it’s hard to disentangle signaling effects from selection effects). Failure to complete an almost-completed degree seems more likely to be a failure of conscientiousness than of intelligence or conformity, so that part is probably mainly conscientiousness signaling.

• NoSignalNoNoise 15 Feb 2026 23:09 UTC

  2 points

  0

  in reply to: Linch’s comment on: In My Misan­thropy Era

  I’ve found my average conversation with Uber drivers to be more interesting and insightful than reading my phone.

  What have you talked with them about, and how did you start the conversations? What have you learned from those conversations?

• NoSignalNoNoise 11 Feb 2026 3:17 UTC

  4 points

  0

  on: End­ing Park­ing Space Saving

  To really put a stop to this practice, there needs to be a good chance that a space saver will be gone by the time you come back. In a college town, an explicit, well-publicized policy of “if a piece of furniture is being used to block a parking space, anyone who wants it can take it” would probably accomplish this.