Hacker News Reader

ukuinaon Ink and Switch Interactive Homepage

Seems like a subset of LiquidText

https://www.liquidtext.net is more PDF-focused, but the workspace is cool.

ukuinaon Thanks to Siri Recaps, your Apple Watch is always listening

Sadly not supported in the USA.

ukuinaon Actively exploited sandbox RCE in all Chromium versions

> Spacing around an em dash varies. Most newspapers insert a space before and after the dash, and many popular magazines do the same, but most books and journals omit spacing, closing whatever comes before and after the em dash right up next to it.

ukuinaon A week of using Codex more than Claude

Which models? It is not useful to compare harnesses without this information.

ukuinaon Grok 4.6 scores 61 on the Artificial Analysis Intelligence Index

Nissan would agree!

https://www.autoblog.com/news/nissan-reports-fifth-straight-...

ukuinaon Transcribe.cpp

What's the easiest way to add speaker separation to this?

ukuinaon I started a “dirt notebook”

It's just A6-sized. You can get a soft A6 binder including refills for a tenth that price.

ukuinaon GLM 5.2 beats Claude in our benchmarks

Why is Sonnet 4.6 ranked higher than Opus 4.6?

ukuinaon GLM-5.2 is a step change for open agents

Synthetic.new and Claude Code using GLM-5.2. Great model, but the harness will error out if using subagents. The base plan only allows one concurrent request at a time. Also, GLM will burn through your weekly quota in a day if you're not precise with your scope.

ukuinaon Bun has an open PR adding shared-memory threads to JavaScriptCore

Valid point, though not a good comparison: You can learn C++ and have a productive career without ever learning or writing a single line of C.

ukuinaon Claude Fable is relentlessly proactive

$12 and 200k tokens!

ukuinaon Harness engineering: Leveraging Codex in an agent-first world

Might as well be 2025.

ukuinaon The four-day workweek in Australia: insights from early adopters of 100:80:100

This is the definition of slack.

ukuinaon Codex is now in the ChatGPT mobile app

Is texting your Coding Agent really the final form? Something that watches your interactions or process execution to surface improvements, or whips up prototypes while you brainstorm seems like the next step.

ukuinaon 245TB Micron 6600 ION Data Center SSD Now Shipping

If you have to ask...

ukuinaon Understand Anything

Doesn't this video make a strong case that LLMs can think?

ukuinaon Claude Pro: Opus model will only be available if extra usage is enabled

Opus was consuming so much usage that it was basically unusable on anything but the Max plan.

ukuinaon Tell HN: An app is silently installing itself on my iPhone every day

I'm pretty sure it was last year, when the "no new features, bugfixes only" MacOS version was released.

ukuinaon MacBook Neo and how the iPad should be

Pour one out for Samsung Knox

ukuinaon Bring your own Agent to MS Teams

Why is there no Teams CLI?

ukuinaon Show HN: Smol machines – subsecond coldstart, portable virtual machines

Doesn't Docker's sbx do this?

https://docs.docker.com/reference/cli/sbx/

ukuinaon Qwen3.6-35B-A3B: Agentic coding power, now open to all

You'd do most of the planning/cognition yourself, down to the module/method signature level, and then have it loop through the plan to "fill in the code". Need a strong testing harness to loop effectively.

ukuinaon Show HN: boringBar – a taskbar-style dock replacement for macOS

GoodSync's pricing is notable: $20/year for five devices, but stackable. I've signed up for 10+ years. GoodSync needs central infrastructure to work, so the ongoing pricing makes sense.

ukuinaon Escaping the Ogallala Trap

Aren't gridlocks noisy due to honks?

ukuinaon Apple Business

Hence, "buggy".

ukuinaon Nanobrew: The fastest macOS package manager compatible with brew

Could you elaborate?

ukuinaon The gold standard of optimization: A look under the hood of RollerCoaster Tycoon

"30 seconds of fun"

https://youtu.be/0q69Msy8ttM?t=287

ukuinaon Professional video editing, right in the browser with WebGPU and WASM

What is the problem with targeting the most prevalent rendering engine?