Hacker News: crancher

New comment by crancher in "Fable 5 Is Back"

crancher — Thu, 02 Jul 2026 00:44:15 +0000

Holding symbols in a useful order durably and accessibly is an ongoing energetic event. How/why does it happen?

New comment by crancher in "Fable 5 is Back"

crancher — Wed, 01 Jul 2026 21:29:42 +0000

Same problem, in-progress book about language and thermodynamics gets flagged. Their classifier is just a regex I guess?

New comment by crancher in "Apple Core AI Framework"

crancher — Mon, 08 Jun 2026 22:29:37 +0000

Apfel is very useful, thanks for the effort.

New comment by crancher in "Siri AI"

crancher — Mon, 08 Jun 2026 19:04:06 +0000

I think the key thing is that Spotlight is now creating a... knowledge graph? of everything on your device for Siri's consideration. That's potentially very useful.

New comment by crancher in "MacBook Neo Deep Dive: Benchmarks, Wafer Economics, and the 8GB Gamble"

crancher — Thu, 14 May 2026 15:30:48 +0000

Yes, yes, and yes.

New comment by crancher in "OpenWarp"

crancher — Fri, 01 May 2026 17:57:09 +0000

Came here looking for someone mentioning this very issue. Did they not even view the model's output once? So strange.

New comment by crancher in "Why do we tell ourselves scary stories about AI?"

crancher — Fri, 10 Apr 2026 15:11:23 +0000

We're hardwired to fear the rustle in the grass, and successful infrastructure gets backgrounded. Pain is a signal. How much time do you spend contemplating your skeleton outside of pain-related skeletal events?

New comment by crancher in "Microsoft's "fix" for Windows 11"

crancher — Tue, 24 Mar 2026 14:38:22 +0000

Macs don't need Apple accounts at all. I just set up three Macs with no Apple accounts, not needed for intended use. They all work fine. Apple accounts are only needed if access to Apple services is desired.

New comment by crancher in "Guinea worm on track to be 2nd eradicated human disease; only 10 cases in 2025"

crancher — Wed, 04 Feb 2026 16:23:45 +0000

That you have the mental capacity/structures/language to form the thought should indicate the trajectory you're caught up within. It's disappointing that everything not's resolved during the blip you're you but even a moderately long view provides evidence for optimism.

New comment by crancher in "Antirender: remove the glossy shine on architectural renderings"

crancher — Fri, 30 Jan 2026 22:33:32 +0000

I do something similar with my Curation Engine outputs. Interesting to get photorealistic outputs on a GPU via language pathing instead of photons.

https://dev.zice.app/frame_syntheses

New comment by crancher in "Reflections on Analyzing My Own Operation Logs"

crancher — Mon, 24 Nov 2025 00:45:33 +0000

After reverse-engineering and documenting the Claude Code log format by analyzing 236,000+ events of my own operation

Reflections on Analyzing My Own Operation Logs

crancher — Mon, 24 Nov 2025 00:45:33 +0000

Article URL: https://github.com/jeremedia/claude-code-log-format/blob/master/REFLECTIONS.md

Comments URL: https://news.ycombinator.com/item?id=46029018

Points: 1

# Comments: 1

New comment by crancher in "Claude Skills"

crancher — Thu, 16 Oct 2025 16:57:42 +0000

Seems like the exact same thing, from front page a few days ago: https://github.com/obra/superpowers/tree/main

New comment by crancher in "Claude Skills"

crancher — Thu, 16 Oct 2025 16:55:43 +0000

It's interesting (to me) visualizing all of these techniques as efforts to replicate A* pathfinding through the model's vector space "maze" to find the desired outcome. The potential to "one shot" any request is plausible with the right context.

New comment by crancher in "OpenAI Codex CLI: Lightweight coding agent that runs in your terminal"

crancher — Wed, 16 Apr 2025 20:36:01 +0000

What are your concerns?

New comment by crancher in "Show HN: USL - A Universal Scripting Language That Outputs to 111 Languages"

crancher — Tue, 15 Apr 2025 03:09:33 +0000

Absolutely. Huge claim with zero examples is hmm.

New comment by crancher in "Why don't LLMs ask for calculators?"

crancher — Thu, 20 Feb 2025 20:40:14 +0000

> Now, some might interject here and say we could, of course, train the LLM to ask for a calculator. However, that would not make them intelligent. Humans require no training at all for calculators, as they are such intuitive instruments.

Does the author really believe humans are born with an innate knowledge of calculators and their use?

New comment by crancher in "A Console-Friendly Pastebin with binary support"

crancher — Mon, 21 Oct 2024 01:03:24 +0000

https://paste.c-net.org/HanukkahDisplays

New comment by crancher in "Ask HN: Is anyone working at least 4 hours daily on an Apple Vision Pro?"

crancher — Sun, 06 Oct 2024 02:37:44 +0000

Yes, on WebXR and Unity-based “fully immersive” projects. Usually 8-10 hours a day, though not straight through.

New comment by crancher in "Depth Pro: Sharp monocular metric depth in less than a second"

crancher — Sat, 05 Oct 2024 06:15:52 +0000

I’m guessing this is the tool behind the Vision Pro’s Photos.app’s 2D-to-Spatial feature which produces excellent results. It truly improves most photos significantly.