healthworker

l’altro ieridiscuss

This creates an incentive to post things that cause high volatility (even in the downward direction) as that creates a "subscribe if you want early warning of market dips" pressure factor.

healthworker

l’altro ieridiscuss

The human-targeted text in this design is using features with a low spatial frequency, compared to the robot-targeted text at a higher spatial frequency. Squinting blurs your vision so the high frequency details are lost.

healthworker

l’altro ieridiscuss

I think your comment is more persuasive than the original article. =)

healthworker

l’altro ieridiscuss

You didn't read the article. This has nothing to do with satellite-delivered Internet.

healthworker

l’altro ieridiscuss

Along these lines, it would be neat to use Raspberry Pi or something to make a "terminal to HDMI adapter", allowing me to have an unlimited number (tens or even hundreds) of external monitors for my laptop, each running a separate terminal-mode program such as a text editor, htop, Bloomberg, etc.

I often find fully working LCD monitors discarded by people who have no use for them, and could easily collect five or more of them that way if I had a convenient way to "wire them up" to my laptop.

healthworker

l’altro ieridiscuss

Agreed. As long as we understand that the map (the code) is not the territory (the spirit of what the code should do), e.g., I can't have type-checking for earthquake if I don't have any earthquake-related concepts in the code yet, but earthquake could and should still be part of the aspirational aspect of the inline spec.

healthworker

l’altro ieridiscuss

The dangerous weather example seems like a poor example. Most fields I would add would be orthogonal to existing functionality, and if they weren't, I would account for things I need to change (e.g. the is dangerous function) during the specification phase.

As an example of what I mean: The spec for "is dangerous" would have specified what "dangerous" means, and if that included high speeds, then that would have been known in advance when I made a plan to implement the speed feature.

The reason I mention this is that agentic coding seems to be requiring us to move to "spec-driven development" and it behooves us all to learn how best to adapt to this new landscape.

A simple example of an inline spec (i.e., a spec living as a comment in the code file, a bit like literate programming) for "is dangerous" would be something like:

    /*
     * dangerous = anything that would cause a risk of getting hurt for a healthy adult, e.g. heat stroke, lightning strike, hypothermia, gale force wind, flood, hurricane, tsunami, earthquake
     * Not all of this is implemented now, but that is what this function should try to do.
     */

healthworker

15 giorni fadiscuss

In this case he was investigating misuse of Pegasus spyware specifically, and was targeted with it while doing so. That's obstruction of justice, morally speaking, and would feel very scary, in that it would make you feel that this company might be so powerful that investigating it is personally dangerous.

healthworker

mese scorsodiscuss

As Majromax stated, if the game is framed such that losing vs MAD have the same penalty, then it is not at all representative of reality. It is set up to be a "you miss 100% of the shots you don't take" game.

As an important aside, I hope that everyone is aware that as of 2025, the annual US military funding law (the NDAA) established a prohibition by Congress prohibiting AI from automating nuclear launch. This was already US policy and is common sense; now it's also part of federal law.

10 USC Ch. 24: NUCLEAR POSTURE https://uscode.house.gov/view.xhtml?path=/prelim@title10/sub...

FY2025 NDAA, Section 1638 https://agora.eto.tech/instrument/1740

healthworker

2 mesi fadiscuss

This is strong evidence against LLMs experiencing qualia. (I know that that topic often gets people laughed out of the room but please don't jump on me for engaging in that debate. When we can collect evidence and be able to show it to people.)

healthworker

4 mesi fadiscuss

Do you prefer that we not even try to spread beyond our one planet, when an entire galaxy, or maybe even the neighboring ones, might be in reach if we try? What if someday at the very end of the lifetime of our sun and similar stars, we look back, and regret not trying?

healthworker

4 mesi fadiscuss

I think they were saying they would sign up just for the experience, even if it's unnecessary to the program.

healthworker

Submissions

Ask HN: Do you say please and thank you to your LLMs?

AI Will Not Start a Nuclear War, but Humans Might

[untitled]

comments