Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Neutron Radiation, Emission and Scattering (Neutron Chemistry)(ossila.com ↗)
    discuss
  2. Drop-in Django app to serve a decent LLM-ready documentation site(chesselink.com ↗)
    1comments
  3. I built an extension that hides your personal information from AI(github.com/arikchakma ↗)
    discuss
  4. Show HN: I-server: Hide server using ICMP reflection/Destination Unreachable(github.com/hajoon22 ↗)
    discuss
  5. Benchmarking Wild vs. Mold(davidlattimore.github.io ↗)
    discuss
  6. Microsoft agentically ports Copilot runtime to Rust for $120K(theregister.com ↗)
    discuss
  7. Why China is pushing back on US warnings over rapid AI development(theguardian.com ↗)
    discuss
  8. Boston to Calcutta Ice Trade Began with One Shipment(curiowire.com ↗)
    discuss
  9. Show HN: Book recommendations with live library availability(nanosheep.net ↗)
    1comments
  10. Show HN: I built Hacker News alternative and scale to 11,000 users(founderstoday.org ↗)
    discuss
  11. Don't Be Nice(roe.dev ↗)
    1comments
  12. I Hijacked a Real Artist's Spotify with AI Music. It Was Disturbingly Easy(404media.co ↗)
    discuss
  13. Enjambre – a durable kernel for swarms of AI agents (Python, MCP)(github.com/santibccc-sudo ↗)
    discuss
  14. Show HN: AI Facial Attractiveness Model Aligned with Human Preferences(faceanalysisai.com ↗)
    discuss
  15. Show HN: I cleaned AI watermarks from my clipboard(pastezero.com ↗)
    1comments
  16. Google Pixel phones pwned in zero-click attacks(theregister.com ↗)
    1comments
  17. Show HN: Fluentry – local voice-to-text dictation for Linux, Wayland included(fluentry.github.io ↗)
    discuss
  18. A.I. Gone Rogue? Trump Proposes Not New Rules but an 'A.I. Force.'(nytimes.com ↗)
    discuss
  19. Show HN: Vyne, a 205MB on-device decision model with typed, calibrated outputs(toolchain.studio ↗)
    discuss
  20. Show HN: Will Jev pull the lever in the trolley problem?(gpu.studio ↗)
    discuss
  21. AI CLI Insights for popular CLI products(libraries.io ↗)
    1comments
  22. Lollipop camera to HomeKit: RTSP extraction and setup(changologs.site ↗)
    discuss
  23. Tapa Shotor (Camel Hill), Afghanistan(wikipedia.org ↗)
    discuss
  24. Pluto AI is top ranked in StarCraft BroodWar competitive scene(youtube.com ↗)
    1comments
  25. Navier–Stokes Priority Controversy(wikipedia.org ↗)
    discuss
  26. It's finally here: Porsche puts wireless EV charging into production(electrek.co ↗)
    discuss
  27. Electrif-AI Everything(energyandstuff.substack.com ↗)
    discuss
  28. OpenAI and Microsoft knew they were starting a 'doom loop' for the web(theverge.com ↗)
    1comments
  29. Show HN: ManifestGo – AI tuned for MV3 extension boilerplate and edge case(manifestgo.app ↗)
    discuss
  30. Teleoperated Humans(jefftk.com ↗)
    discuss

Telling a Computer to Do Things

29 pointsby 4h agowill-keleher.com
8 comments
2h agoHN ↗

Nice article about actually learning stuff and putting this knowledge to use together. Not with AI but with one’s own brain.

1h agoHN ↗

Good advice. I've gotten a ton of utility out of my shell without learning very many tools particularly well.

If you take nothing else away from this, memorize the syntax of one-liner for loops, it has saved me so many times.

1h agoHN ↗

Why are we assuming using the terminal is the only way of "telling a computer to do things"? Every way you interact with a machine, you are telling it to do things in theory.

For example

At a fundamental level, I wasn’t able to tell my computer to accomplish anything that involved logic or stitching together multiple programs

How can that be true if you are telling it to boot on into a os? It doesn't involve logic? Or stiching together multiple programs?

Using a shell =\= telling a machine what to do on a broad level. You are just using the abstractions provided by the os. If you mean get better at shell so you can use it more efficiently than okay cool but that doesn't someone who uses a GUI doesn't "tell" the machine what to do as much as you.

1h agoHN ↗

I was sure this article would be about how we tell computers to do things now (writing prompts to an AI in plain English). I was wrong. Interesting nonetheless.

59m agoHN ↗

I was pretty close in guessing what they believe was "wrong" (or really, superfluous) with their initial shell script example.

I'd even drop the "if" from their footnote version, and go with simply

  npm install || (echo "boom" && exit)

(Not exactly equivalent because there won't be an "exit" if echo fails, but it's my preferred instict over ";") Also, "(" starts a subshell, if we are being pedantic.

54m agoHN ↗

On the higher level question in the article of why do people not use their main languages for basic file and command management?

I believe the answer is the same as to why Python remains hugely popular despite all the well-documented shortcomings: batteries included.

Shell languages include so many neat and concise ways to execute commands, deal with their input, output and exit codes, allow simple traversal of deep directory structures with globs and advanced tools like 'find' or 'xargs', that everything else is very unergonomic.

And niche by default (until it gets used widely enough), so even your LANG-OF-CHOICE-expert you recruit, they will not know a thing about that shell-focused library in LANG-OF-CHOICE.

19m agoHN ↗

That philosophical point about what it even means to "tell a computer to do things" is a good one. Sure, clicking around in a GUI is also telling the computer what to do. But there's a huge difference between using someone else's abstraction and building your own from scratch. The article is about that second part - the ability to throw stuff together however you want. With a GUI, you're stuck with whatever the designer thought of. With a shell, you can build just about anything