Hacker Newsnew | past | comments | ask | show | jobs | submit | westurner's commentslogin

ScholarlyArticle:

"Additive-free and brine-discharge-free solar-thermal desalination with simultaneous complete mineral mining from ocean water" (2026) https://www.nature.com/articles/s41377-026-02315-4


Stephen Colbert: https://en.wikipedia.org/wiki/Stephen_Colbert :

> Colbert first performed professionally as an understudy for Steve Carell at Second City Chicago.



They used to have:

  - Organize the world's information
  
  - Don't be evil
Who was president at this time? Was this while they were denying students the option to code on the computers we bought for them because security?

From https://en.wikipedia.org/wiki/Don%27t_be_evil :

> Between 21 April and 4 May 2018, Google removed the motto from the preface, leaving a mention in the final line: "And remember... don't be evil, and if you see something that you think isn't right – speak up!"

That sounds like NY to me.


People think Google removing "Don't be evil" means Google somewhat implicitly told everyone they'll be evil from now on. That doesn't make much sense. It was likely a marketing decision since having "evil" in your motto is not good even if you explicitly state "DON'T BE" before it. Just like if I write in my profile text "I'm not a pedo", you'll think "this guy is likely a pedo"

I would have loved to be in the room when this decision was made, but my guess is that they decided that it was better to face some subtle nerd backlash now rather than having this quoted at the start of countless congressional hearings over the next decade.

Don't create the Torment Nexus

Wouldn't this be faster with an agent skill that has code?

/skill-creator [or /create-skill] Write an agent skill with code script(s) that use an existing user space IP library that works with your agent runtime, to [...]

ComposioHQ/awesome-claude-skills: https://github.com/ComposioHQ/awesome-claude-skills

anthopics/skills//skill-creator/SKILL.md: https://github.com/anthropics/skills/blob/main/skills/skill-...

/.agents/skills/skill-name/SKILL.md, scripts/{script_name.py,__init__.py}

https://agentskills.io/what-are-skills


Well, yeah, of course it would be.

Even faster would just to be use code in the first place!


But why would you use tokens instead of using the LLM to call code for this?

The minimum overhead to doing it with the LLM is the useful question


ScholarlyArticle: "Plasticity and language in the anaesthetized human hippocampus" (2026) https://www.nature.com/articles/s41586-026-10448-0

Sleep-learning https://en.wikipedia.org/wiki/Sleep-learning

Also, Sleep and learning https://en.wikipedia.org/wiki/Sleep_and_learning



A sphinxdoc ratex Makefile task for building from .rst and .md doctrees to PDF?

There are probably enough tests for a Rust rewrite of docutils and sphinx with Python extension compatibility; docutils.rs and sphinxdoc.rs?


What are the benchmarks for this, in terms of costs of computation and error; cost to converge?

Re: hyperparameter tuning and autoresearch: https://news.ycombinator.com/item?id=47444581

Parameter-free LLMs would be cool


Other examples of where non-LLM methods are and/or will always be more efficient than LLM methods?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: