westurner

Stripping whitespace from JSON probably does save that many tokens?

/? fewer tokens: https://hn.algolia.com/?q=fewer+tokens

But `ls -print0` is not an option;

  set -x
  mkdir test; cd "$_"
  touch "test "$'\n'"12.txt"
  ls
  for name in `ls`; do echo "$name"; done
  find
  find . -exec echo {} \;
  find . -print0
  find . -print0 | xargs -n 1 -0 echo
  #
  find . -print0 | el -0 -v -x echo
  find . -print0 | el -0 -v -x echo "{0} #"
  find . -print0 | el -0 -v -x echo '"{0}" #'
  find . -print0 | el -0 -x sh -x -c 'echo "{0} #"'
  

  find . -print0 | el -0 -x -- sh -c 'set -x: echo "{0} #"'
  

`el -v/--verbose` is useful because it prints each input; though, I might not have written `el` if I had been aware of `xargs -n/--max-args=1`

- A Digital Twin of the facility in 3D?

- R3F and Drei for React-based 3D Three.js animations; but it might be best to pre-render at high FPS and encode with ffmpeg

- ProjectM, MilkDrop3, limit

- TIL about shuffle skating aka jam skating

- OpenNext NextJS

- Serwist for a PWA so that you can access your stats when not connected

- IndexedDb, sqlite-wasm, duckdb-wasm

- To run lights, projectors, etc. (with low enough hertz to avoid PSE Photosensitive Epilepsy risk) I have heard about: OSC (BespokeSynth, Resolume,), SMPTE LTC, Resolume projection mapping, mapmapteam/mapmap projection mapping,

Open Lighting Architecture has a list of Ethernet Protocols including OSC, and a list of USB/Serial/Network devices.

HyperHDR does ambient lightning to match displayed video.

Sometimes the glare from all of the lights is too much for visually impaired persons and animals FWIU; no varying lights hour might win too.

- If the machines are so old that all you need is a relay, there's possibly not as much need to model the whole system with TLA+ and check the proof.

Could ask the model about an "SIL" (Safety Integrity Level) controller or similar for a bowling alley. Rust RTIC and Rust embassy have RTOS features.

How to write unit and integration tests for the system without hardware simulators that can randomly fault.

- Soft-white backlit multicolor(?) ePaper displays would save energy and cut glare.

Most flying insects cannot see red, orange or yellow light;

Red preserves night vision. Blue keeps operators awake. Blue and white cause the worst glare. Jet instrument gauges are reddish/orange-ish.

- UV Blacklight paint, UV LED Black light.

Blue light glasses are great for reducing UV blacklight glare.

- Best LED projector 2026

- Gogs/Gitea/Forgejo host Issues, repos and can run Actions with nektos/act.

Cool project. Integrate an issue tracker with mobile notifications?

When I searched for that last sentence in full, I found this:

dj-bolt/django-bolt: Rust-powered API framework for Django achieving 60k+ RPS. Uses Actix Web for HTTP, PyO3 for Python bridging, msgspec for serialization. Decorator-based routing with built-in auth and middleware. https://github.com/dj-bolt/django-bolt :

Faster than FastAPI, but with Django ORM, Django Admin, and Django packages. [...] Django-Bolt is a high-performance API framework for Django, providing Rust-powered API endpoints capable of 188k+ RPS.

Here's a v0.0.2 jinja2rs built on minijinja with optional python compatibility; CompatMode::Django(…) Django template language (filters, app-directory loader, auto-escape) :

jinja2rs::filters : https://github.com/westurner/dsport/blob/main/src/jinja2rs/s...

jinja2rs::filters::django : https://github.com/westurner/dsport/blob/main/src/jinja2rs/s...

Which Rust components are like the Django ORM and django.contrib.admin and DRF/FastAPI; with convention over configuration and tests and great docs?

teal hydrogen: electrolytic catalysis, renewable feedstocks

turquoise hydrogen: high-temp pyrolysis of methane

So e.g. [1] is turquoise hydrogen.

[1] "Production of hydrogen and carbon nanotubes from methane using a multi-pass floating catalyst chemical vapour deposition reactor with process gas recycling" (2025) https://www.nature.com/articles/s41560-025-01925-3

A renewable feedstock process that yields Hydrogen and O-CNT would be yielding teal hydrogen.

A natural gas process that yields Hydrogen and O-CNT would be yielding turquoise hydrogen.

What would save tokens?

- "Groundtruth – checks your AI coding agent's claims against the Git diff" (2026-07) https://github.com/akahkhanna/groundtruth .. https://news.ycombinator.com/item?id=48809364

- "Schema Harness Achieves ~99% on Arc‑AGI‑3 Public" (2026-07) https://news.ycombinator.com/item?id=48935905

But do radio waves and photons have mass and why does gravity affect them?

Re: "Show HN: Frugon – Find which LLM calls a cheaper model could handle (local, MIT)" https://github.com/Rodiun/frugon and cost to converge: https://news.ycombinator.com/item?id=48816724#48877261

ScholarlyArticle: "Thermodynamics of Black Holes, Far from Equilibrium" (2026) https://journals.aps.org/prl/abstract/10.1103/3c1r-v8f1

NewsArticle: "Dynamic black holes explained by simple thermodynamics?" | Penn State University https://www.psu.edu/news/eberly-college-science/story/dynami...

NewsArticle: "Stephen Hawking's famous 'leaky' black hole theory gets much-needed update" (2026) Space https://www.space.com/astronomy/black-holes/stephen-hawkings...

Total cost estimates should included unassessed environmental hazard, carbon, energy mix

It's economical to capture and sell the Helium out of natural gas even without a flaring penalty to encourage to be efficient with heat and material outputs.

It's also already economical to make Hydrogen and Carbon-based products like Graphene and spec Oxidized Carbon Nanotubes from natural gas.

https://news.ycombinator.com/item?id=48946629

And, it's surely also profitable to make more He3, He4, and T with a fusion reactor: https://news.ycombinator.com/item?id=48946533

A teal Hydrogen company that splits Methane (CH4) into Hydrogen (H) and Oxidized Carbon Nanotubes (O-CNT) would be significantly more profitable than one that splits into Hydrogen and Carbon Black. (Because after production costs, O-CNT have a significantly higher market price than carbon black.)

But do our communities want fracking wastewater in our aquifer groundwater or on our farm fields?

When is the break-even point for cracking Helium from natural gas at current and predicted Helium market prices?

Given the neutron radiation from a fusion reactor, it's possible to synthesize Helium (He4) and more Tritium (T) and 3He using Lithium-6 with Boron-10 and Boron-11. An adjacent vertically-aligned (VA-CNT) battery would last for like 20 years and could power safety systems.

A Deuterated Vitrimer pipe with Li-6, B-10, and B-11 could be recycled to extract the T, 3He, and 4He.

Deuterated Vitrimers could also be disposed of in a waste burner.

So, a project can have 100% test coverage (the tests run each line or branch at least once without any errors or failures) but fail acceptance tests.

If there is prompt insufficiency, there is probably acceptance test insufficiency.

A more assuming agent could automatically develop a plan that includes presumptive acceptance tests and request feedback before spending tokens

Fearless refactoring is possible given test coverage? True in my experience; both with higher-risk human-developed software and with agent-coded software.

I haven't yet made myself learn the new swarm of concurrent agents with different specializations/agent_instructions methods yet.

Are multiple worktrees worth the cognitive burden and merge overhead?

A merge maintainer is always in code review mode

There is a market shortage of helium but shouldn't be:

There's also helium in methane, but unfortunately few places crack out the helium from natural gas.

TIL Helium kills Kudzu and powers fusion power plants.

If you have a "this is an incompetent team" attitude, that's what quality software you'll get back; whether or not there's LLM use.

If you have a "I trust my competent team to write good enough tests and test coverage isn't my responsibility" attitude, that's what quality software you'll get back.

There are people producing good and excellent quality software with LLMs. Presumably you must discard low-quality code in order to maintain quality.

There's certainly a limit to code quality with current models. On number of lines of code per unit of time, LLM tools certainly already win.

Can costly automated code review for PRs catch most of the problems before they're under consideration for merge?

For example, the vscode repo has extensive copilot integration. Every PR gets auto code reviewed. But with their tokens or the contributors'?

If I take poor quality code (AI-assisted or not) and spend a few hundred dollars on tokens for a next gen model and agent to get to 100% coverage and review for security bugs and CWE common weaknesses, what quality code will I have without refactoring with proven patterns and type annotations and polishing docstrings?

What new or existing metric would better assess test coverage than line coverage and branch coverage?

I wrote a tool called tert - I guess it's called an agent harness now - to run various test runners and log test output and coverage output to disk. FWIU stripping spaces from JSON does save tokens. It seems like feeding coverage lines-missing maps into the prompt results in better output, better LLM-authored tests.

"Refactor these tests for maintainability and coverage. Use fixtures, mocks, and parametrization"

Given that it's so easy to generate better than test stubs with a model and manually review what it thinks covers, I think it's fair to expect a manager to run and add to the tests.

If devops has done their job, it should be trivial for a manager to contribute to the tests and run the build on git push (or manually re-run the build with the web UI).

If a manager has deploy rights, they should be able to run the tests.

Causes of failure; Operator skill, prompt insufficiency, model insufficiency, agent insufficiency, insufficient code review, overloaded merge maintainer

Schema, the harness we introduce today, reaches 99% on the ARC‑AGI‑3 Public set using Claude Opus 4.8 and Fable 5, and 95.35% using GPT‑5.6 Sol.

Impressive results. Will this translate to coding agents (and training general purpose and for coding LLMs) too?

---

When Michelson and Morley could not detect the medium light was supposed to wave in, Lorentz took the first route: keep the aether, patch the rules with contraction hypotheses that absorbed the null result. Einstein took the second: in special relativity, he discarded the aether as part of the state and made simultaneity frame-relative, yielding a simple electrodynamics of moving bodies.

BECs, SVT, Superfluid Quantum Gravity

Massful photons are modeled with Proca fields. Like Einstein, Proca was also a student of Minkowski. The Mass-Equivalence principle ~~does not~~ still holds if photons have mass.

(edit) Energy-momentum relation: https://en.wikipedia.org/wiki/Energy%E2%80%93momentum_relati...

could not detect the medium light was supposed to wave in,

Superfluid Quantum Gravity (Fedi,) says that there is a medium that light waves through; there is not nothing in space, space is a quantum dilatant superfluid with near-zero viscosity.

Like becoming a manager?

When a person becomes a manager, they do or do not have enough time and expertise to review all of the code that they trust the team to produce.

Managers usually get into automated testing; unit tests, integration tests, acceptance tests, and maybe also BDD syntax

Managers and developers are responsible for setting a test coverage threshold for merge approval.

If there is 100% branch coverage test coverage for a codebase, what would coverage-guided fuzzing or property testing find? If there is 100% branch coverage test coverage for a codebase, what is the value of spending resources on formal verification?

How does the value of LLM-produced 100% branch coverage compare to no-LLM 100% branch coverage?

No worries. Here's this:

Public link: UBI Options on Interledger, W3C VC: https://share.gemini.google/3R53k83Dxfw2 ;

Review the Interledger spec, then develop time-limited options contracts for basic income on Interledger. Consider WebMonetization for how to implement Interledger. https://interledger.org/developers/rfcs/interledger-protocol...

[...] A UBI Fund would mint the "Streaming Zero-Strike Call Option" options contract by [...]

And then KYC and AML and grant accountability, so how to allow-list according to a merchant category etc but be Zero Trust without DNS?

Re: Clean Language from David Grove of NZ; and re: a Dr Sbaitso but with Clean Language: https://news.ycombinator.com/item?id=44188600

"Ask HN: Best online speech / public speaking course?" https://news.ycombinator.com/item?id=27754239

Setting aside concerns about observed outcomes from various basic income experiments; https://hn.algolia.com/?q=basic+income ,

So they're like expiring options contracts?

Which existing distributed ledger technology DLT are sufficient to host expiring options contracts for basic income?

If there was a CBDC (Central Bank Digital Currency), how would basic income and/or old age and disability social welfare (US: SNAP benefits, SSOADI) be distributed?

uptane is an updater system for automotive applications based on TUF

Audio diagnostics and prediction tied into the ECU/VCU would be of value

Would this work on a generator with a catalytic converter?

Looks like https://rusefi.com/ lists the features.

Re: STM32 but for real-time audio in Rust and RTOS / RTIC; from yesterday: https://news.ycombinator.com/item?id=48910657

The other day re: controlling the rear wheels of a truck like motorcycle trailer tied into the CAN bus: https://news.ycombinator.com/item?id=46283617 :

A modular truck bed could have Through-the-road TTR AWD (given a better VCU) and e.g. [radial axial flux] hub motors or an axle motor.

oxirs has CANbus support: oxirs-canbus. oxirs-tauri is a desktop UI with CAN bus support.

There are Open IP cores for RISC-V for many things.

Is this already modeled with TLA+?

Model this with TLA+, critically compare the spec with the implementation, revise, and run it?

MOF or COF for quantum computing?

Maybe "Radical COFs" or "Spintronic COFs", or Carbon-Linked Covalent Organic Frameworks for Spintronics, e.g. GQD Graphene Quantum Dots; GQD-COFs